renewable energy research paper examples

IEEE Account

• Change Username/Password
• Update Address

Purchase Details

• Payment Options
• Order History
• View Purchased Documents

Profile Information

• Communications Preferences
• Profession and Education
• Technical Interests
• US & Canada: +1 800 678 4333
• Worldwide: +1 732 981 0060
• Contact & Support
• About IEEE Xplore
• Accessibility
• Terms of Use
• Nondiscrimination Policy
• Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity. © Copyright 2024 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.

• Open access
• Published: 07 January 2020

Renewable energy for sustainable development in India: current status, future prospects, challenges, employment, and investment opportunities

• Charles Rajesh Kumar. J   ORCID: orcid.org/0000-0003-2354-6463 1 &
• M. A. Majid 1  

Energy, Sustainability and Society volume  10 , Article number:  2 ( 2020 ) Cite this article

439k Accesses

316 Citations

83 Altmetric

Metrics details

The primary objective for deploying renewable energy in India is to advance economic development, improve energy security, improve access to energy, and mitigate climate change. Sustainable development is possible by use of sustainable energy and by ensuring access to affordable, reliable, sustainable, and modern energy for citizens. Strong government support and the increasingly opportune economic situation have pushed India to be one of the top leaders in the world’s most attractive renewable energy markets. The government has designed policies, programs, and a liberal environment to attract foreign investments to ramp up the country in the renewable energy market at a rapid rate. It is anticipated that the renewable energy sector can create a large number of domestic jobs over the following years. This paper aims to present significant achievements, prospects, projections, generation of electricity, as well as challenges and investment and employment opportunities due to the development of renewable energy in India. In this review, we have identified the various obstacles faced by the renewable sector. The recommendations based on the review outcomes will provide useful information for policymakers, innovators, project developers, investors, industries, associated stakeholders and departments, researchers, and scientists.

Introduction

The sources of electricity production such as coal, oil, and natural gas have contributed to one-third of global greenhouse gas emissions. It is essential to raise the standard of living by providing cleaner and more reliable electricity [ 1 ]. India has an increasing energy demand to fulfill the economic development plans that are being implemented. The provision of increasing quanta of energy is a vital pre-requisite for the economic growth of a country [ 2 ]. The National Electricity Plan [NEP] [ 3 ] framed by the Ministry of Power (MoP) has developed a 10-year detailed action plan with the objective to provide electricity across the country, and has prepared a further plan to ensure that power is supplied to the citizens efficiently and at a reasonable cost. According to the World Resource Institute Report 2017 [ 4 , 5 ], India is responsible for nearly 6.65% of total global carbon emissions, ranked fourth next to China (26.83%), the USA (14.36%), and the EU (9.66%). Climate change might also change the ecological balance in the world. Intended Nationally Determined Contributions (INDCs) have been submitted to the United Nations Framework Convention on Climate Change (UNFCCC) and the Paris Agreement. The latter has hoped to achieve the goal of limiting the rise in global temperature to well below 2 °C [ 6 , 7 ]. According to a World Energy Council [ 8 ] prediction, global electricity demand will peak in 2030. India is one of the largest coal consumers in the world and imports costly fossil fuel [ 8 ]. Close to 74% of the energy demand is supplied by coal and oil. According to a report from the Center for monitoring Indian economy, the country imported 171 million tons of coal in 2013–2014, 215 million tons in 2014–2015, 207 million tons in 2015–2016, 195 million tons in 2016–2017, and 213 million tons in 2017–2018 [ 9 ]. Therefore, there is an urgent need to find alternate sources for generating electricity.

In this way, the country will have a rapid and global transition to renewable energy technologies to achieve sustainable growth and avoid catastrophic climate change. Renewable energy sources play a vital role in securing sustainable energy with lower emissions [ 10 ]. It is already accepted that renewable energy technologies might significantly cover the electricity demand and reduce emissions. In recent years, the country has developed a sustainable path for its energy supply. Awareness of saving energy has been promoted among citizens to increase the use of solar, wind, biomass, waste, and hydropower energies. It is evident that clean energy is less harmful and often cheaper. India is aiming to attain 175 GW of renewable energy which would consist of 100 GW from solar energy, 10 GW from bio-power, 60 GW from wind power, and 5 GW from small hydropower plants by the year 2022 [ 11 ]. Investors have promised to achieve more than 270 GW, which is significantly above the ambitious targets. The promises are as follows: 58 GW by foreign companies, 191 GW by private companies, 18 GW by private sectors, and 5 GW by the Indian Railways [ 12 ]. Recent estimates show that in 2047, solar potential will be more than 750 GW and wind potential will be 410 GW [ 13 , 14 ]. To reach the ambitious targets of generating 175 GW of renewable energy by 2022, it is essential that the government creates 330,000 new jobs and livelihood opportunities [ 15 , 16 ].

A mixture of push policies and pull mechanisms, accompanied by particular strategies should promote the development of renewable energy technologies. Advancement in technology, proper regulatory policies [ 17 ], tax deduction, and attempts in efficiency enhancement due to research and development (R&D) [ 18 ] are some of the pathways to conservation of energy and environment that should guarantee that renewable resource bases are used in a cost-effective and quick manner. Hence, strategies to promote investment opportunities in the renewable energy sector along with jobs for the unskilled workers, technicians, and contractors are discussed. This article also manifests technological and financial initiatives [ 19 ], policy and regulatory framework, as well as training and educational initiatives [ 20 , 21 ] launched by the government for the growth and development of renewable energy sources. The development of renewable technology has encountered explicit obstacles, and thus, there is a need to discuss these barriers. Additionally, it is also vital to discover possible solutions to overcome these barriers, and hence, proper recommendations have been suggested for the steady growth of renewable power [ 22 , 23 , 24 ]. Given the enormous potential of renewables in the country, coherent policy measures and an investor-friendly administration might be the key drivers for India to become a global leader in clean and green energy.

Projection of global primary energy consumption

An energy source is a necessary element of socio-economic development. The increasing economic growth of developing nations in the last decades has caused an accelerated increase in energy consumption. This trend is anticipated to grow [ 25 ]. A prediction of future power consumption is essential for the investigation of adequate environmental and economic policies [ 26 ]. Likewise, an outlook to future power consumption helps to determine future investments in renewable energy. Energy supply and security have not only increased the essential issues for the development of human society but also for their global political and economic patterns [ 27 ]. Hence, international comparisons are helpful to identify past, present, and future power consumption.

Table 1 shows the primary energy consumption of the world, based on the BP Energy Outlook 2018 reports. In 2016, India’s overall energy consumption was 724 million tons of oil equivalent (Mtoe) and is expected to rise to 1921 Mtoe by 2040 with an average growth rate of 4.2% per annum. Energy consumption of various major countries comprises commercially traded fuels and modern renewables used to produce power. In 2016, India was the fourth largest energy consumer in the world after China, the USA, and the Organization for economic co-operation and development (OECD) in Europe [ 29 ].

The projected estimation of global energy consumption demonstrates that energy consumption in India is continuously increasing and retains its position even in 2035/2040 [ 28 ]. The increase in India’s energy consumption will push the country’s share of global energy demand to 11% by 2040 from 5% in 2016. Emerging economies such as China, India, or Brazil have experienced a process of rapid industrialization, have increased their share in the global economy, and are exporting enormous volumes of manufactured products to developed countries. This shift of economic activities among nations has also had consequences concerning the country’s energy use [ 30 ].

Projected primary energy consumption in India

The size and growth of a country’s population significantly affects the demand for energy. With 1.368 billion citizens, India is ranked second, of the most populous countries as of January 2019 [ 31 ]. The yearly growth rate is 1.18% and represents almost 17.74% of the world’s population. The country is expected to have more than 1.383 billion, 1.512 billion, 1.605 billion, 1.658 billion people by the end of 2020, 2030, 2040, and 2050, respectively. Each year, India adds a higher number of people to the world than any other nation and the specific population of some of the states in India is equal to the population of many countries.

The growth of India’s energy consumption will be the fastest among all significant economies by 2040, with coal meeting most of this demand followed by renewable energy. Renewables became the second most significant source of domestic power production, overtaking gas and then oil, by 2020. The demand for renewables in India will have a tremendous growth of 256 Mtoe in 2040 from 17 Mtoe in 2016, with an annual increase of 12%, as shown in Table 2 .

Table 3 shows the primary energy consumption of renewables for the BRIC countries (Brazil, Russia, India, and China) from 2016 to 2040. India consumed around 17 Mtoe of renewable energy in 2016, and this will be 256 Mtoe in 2040. It is probable that India’s energy consumption will grow fastest among all major economies by 2040, with coal contributing most in meeting this demand followed by renewables. The percentage share of renewable consumption in 2016 was 2% and is predicted to increase by 13% by 2040.

How renewable energy sources contribute to the energy demand in India

Even though India has achieved a fast and remarkable economic growth, energy is still scarce. Strong economic growth in India is escalating the demand for energy, and more energy sources are required to cover this demand. At the same time, due to the increasing population and environmental deterioration, the country faces the challenge of sustainable development. The gap between demand and supply of power is expected to rise in the future [ 32 ]. Table 4 presents the power supply status of the country from 2009–2010 to 2018–2019 (until October 2018). In 2018, the energy demand was 1,212,134 GWh, and the availability was 1,203,567 GWh, i.e., a deficit of − 0.7% [ 33 ].

According to the Load generation and Balance Report (2016–2017) of the Central Electricity Authority of India (CEA), the electrical energy demand for 2021–2022 is anticipated to be at least 1915 terawatt hours (TWh), with a peak electric demand of 298 GW [ 34 ]. Increasing urbanization and rising income levels are responsible for an increased demand for electrical appliances, i.e., an increased demand for electricity in the residential sector. The increased demand in materials for buildings, transportation, capital goods, and infrastructure is driving the industrial demand for electricity. An increased mechanization and the shift to groundwater irrigation across the country is pushing the pumping and tractor demand in the agriculture sector, and hence the large diesel and electricity demand. The penetration of electric vehicles and the fuel switch to electric and induction cook stoves will drive the electricity demand in the other sectors shown in Table 5 .

According to the International Renewable Energy Agency (IRENA), a quarter of India’s energy demand can be met with renewable energy. The country could potentially increase its share of renewable power generation to over one-third by 2030 [ 35 ].

Table 6 presents the estimated contribution of renewable energy sources to the total energy demand. MoP along with CEA in its draft national electricity plan for 2016 anticipated that with 175 GW of installed capacity of renewable power by 2022, the expected electricity generation would be 327 billion units (BUs), which would contribute to 1611 BU energy requirements. This indicates that 20.3% of the energy requirements would be fulfilled by renewable energy by 2022 and 24.2% by 2027 [ 36 ]. Figure 1 shows the ambitious new target for the share of renewable energy in India’s electricity consumption set by MoP. As per the order of revised RPO (Renewable Purchase Obligations, legal act of June 2018), the country has a target of a 21% share of renewable energy in its total electricity consumption by March 2022. In 2014, the same goal was at 15% and increased to 21% by 2018. It is India’s goal to reach 40% renewable sources by 2030.

Target share of renewable energy in India’s power consumption

Estimated renewable energy potential in India

The estimated potential of wind power in the country during 1995 [ 37 ] was found to be 20,000 MW (20 GW), solar energy was 5 × 10 15 kWh/pa, bioenergy was 17,000 MW, bagasse cogeneration was 8000 MW, and small hydropower was 10,000 MW. For 2006, the renewable potential was estimated as 85,000 MW with wind 4500 MW, solar 35 MW, biomass/bioenergy 25,000 MW, and small hydropower of 15,000 MW [ 38 ]. According to the annual report of the Ministry of New and Renewable Energy (MNRE) for 2017–2018, the estimated potential of wind power was 302.251 GW (at 100-m mast height), of small hydropower 19.749 GW, biomass power 17.536 GW, bagasse cogeneration 5 GW, waste to energy (WTE) 2.554 GW, and solar 748.990 GW. The estimated total renewable potential amounted to 1096.080 GW [ 39 ] assuming 3% wasteland, which is shown in Table 7 . India is a tropical country and receives significant radiation, and hence the solar potential is very high [ 40 , 41 , 42 ].

Gross installed capacity of renewable energy in India

As of June 2018 reports, the country intends to reach 225 GW of renewable power capacity by 2022 exceeding the target of 175 GW pledged during the Paris Agreement. The sector is the fourth most attractive renewable energy market in the world. As in October 2018, India ranked fifth in installed renewable energy capacity [ 43 ].

Gross installed capacity of renewable energy—according to region

Table 8 lists the cumulative installed capacity of both conventional and renewable energy sources. The cumulative installed capacity of renewable sources as on the 31 st of December 2018 was 74081.66 MW. Renewable energy (small hydropower, wind, biomass, WTE, solar) accounted for an approximate 21% share of the cumulative installed power capacity, and the remaining 78.791% originated from other conventional sources (coal, gas diesel, nuclear, and large hydropower) [ 44 ]. The best regions for renewable energy are the southern states that have the highest solar irradiance and wind in the country. When renewable energy alone is considered for analysis, the Southern region covers 49.121% of the cumulative installed renewable capacity, followed by the Western region (29.742%), the Northern region (18.890%), the Eastern region (1.836%), the North-Easter region 0.394%, and the Islands (0.017%). As far as conventional energy is concerned, the Western region with 33.452% ranks first and is followed by the Northern region with 28.484%, the Southern region (24.967%), the Eastern region (11.716%), the Northern-Eastern (1.366%), and the Islands (0.015%).

Gross installed capacity of renewable energy—according to ownership

State government, central government, and private players drive the Indian energy sector. The private sector leads the way in renewable energy investment. Table 9 shows the installed gross renewable energy and conventional energy capacity (percentage)—ownership wise. It is evident from Fig. 2 that 95% of the installed renewable capacity derives from private companies, 2% from the central government, and 3% from the state government. The top private companies in the field of non-conventional energy generation are Tata Power Solar, Suzlon, and ReNew Power. Tata Power Solar System Limited are the most significant integrated solar power players in the country, Suzlon realizes wind energy projects, and ReNew Power Ventures operate with solar and wind power.

Gross renewable energy installed capacity (percentage)—Ownership wise as per the 31.12.2018 [ 43 ]

Gross installed capacity of renewable energy—state wise

Table 10 shows the installed capacity of cumulative renewable energy (state wise), out of the total installed capacity of 74,081.66 MW, where Karnataka ranks first with 12,953.24 MW (17.485%), Tamilnadu second with 11,934.38 MW (16%), Maharashtra third with 9283.78 MW (12.532%), Gujarat fourth with 10.641 MW (10.641%), and Rajasthan fifth with 7573.86 MW (10.224%). These five states cover almost 66.991% of the installed capacity of total renewable. Other prominent states are Andhra Pradesh (9.829%), Madhya Pradesh (5.819%), Telangana (5.137%), and Uttar Pradesh (3.879%). These nine states cover almost 91.655%.

Gross installed capacity of renewable energy—according to source

Under union budget of India 2018–2019, INR 3762 crore (USD 581.09 million), was allotted for grid-interactive renewable power schemes and projects. As per the 31.12.2018, the installed capacity of total renewable power (excluding large hydropower) in the country amounted to 74.08166 GW. Around 9.363 GW of solar energy, 1.766 GW of wind, 0.105 GW of small hydropower (SHP), and biomass power of 8.7 GW capacity were added in 2017–2018. Table 11 shows the installed capacity of renewable energy over the last 10 years until the 31.12.2018. Wind energy continues to dominate the countries renewable energy industry, accounting for over 47% of cumulative installed renewable capacity (35,138.15 MW), followed by solar power of 34% (25,212.26 MW), biomass power/cogeneration of 12% (9075.5 MW), and small hydropower of 6% (4517.45 MW). In the renewable energy country attractiveness index (RECAI) of 2018, India ranked in fourth position. The installed renewable energy production capacity has grown at an accelerated pace over the preceding few years, posting a CAGR of 19.78% between 2014 and 2018 [ 45 ] .

Estimation of the installed capacity of renewable energy

Table 12 gives the share of installed cumulative renewable energy capacity, in comparison with the installed conventional energy capacity. In 2022 and 2032, the installed renewable energy capacity will account for 32% and 35%, respectively [ 46 , 47 ]. The most significant renewable capacity expansion program in the world is being taken up by India. The government is preparing to boost the percentage of clean energy through a tremendous push in renewables, as discussed in the subsequent sections.

Gross electricity generation from renewable energy in India

The overall generation (including the generation from grid-connected renewable sources) in the country has grown exponentially. Between 2014–2015 and 2015–2016, it achieved 1110.458 BU and 1173.603 BU, respectively. The same was recorded with 1241.689 BU and 1306.614 BU during 2015–2016 and 1306.614 BU from 2016–2017 and 2017–2018, respectively. Figure 3 indicates that the annual renewable power production increased faster than the conventional power production. The rise accounted for 6.47% in 2015–2016 and 24.88% in 2017–2018, respectively. Table 13 compares the energy generation from traditional sources with that from renewable sources. Remarkably, the energy generation from conventional sources reached 811.143 BU and from renewable sources 9.860 BU in 2010 compared to 1.206.306 BU and 88.945 BU in 2017, respectively [ 48 ]. It is observed that the price of electricity production using renewable technologies is higher than that for conventional generation technologies, but is likely to fall with increasing experience in the techniques involved [ 49 ].

The annual growth in power generation as per the 30th of November 2018

Gross electricity generation from renewable energy—according to regions

Table 14 shows the gross electricity generation from renewable energy-region wise. It is noted that the highest renewable energy generation derives from the southern region, followed by the western part. As of November 2018, 50.33% of energy generation was obtained from the southern area and 29.37%, 18.05%, 2%, and 0.24% from Western, Northern, North-Eastern Areas, and the Island, respectively.

Gross electricity generation from renewable energy—according to states

Table 15 shows the gross electricity generation from renewable energy—region-wise. It is observed that the highest renewable energy generation was achieved from Karnataka (16.57%), Tamilnadu (15.82%), Andhra Pradesh (11.92%), and Gujarat (10.87%) as per November 2018. While adding four years from 2015–2016 to 2018–2019 Tamilnadu [ 50 ] remains in the first position followed by Karnataka, Maharashtra, Gujarat and Andhra Pradesh.

Gross electricity generation from renewable energy—according to sources

Table 16 shows the gross electricity generation from renewable energy—source-wise. It can be concluded from the table that the wind-based energy generation as per 2017–2018 is most prominent with 51.71%, followed by solar energy (25.40%), Bagasse (11.63%), small hydropower (7.55%), biomass (3.34%), and WTE (0.35%). There has been a constant increase in the generation of all renewable sources from 2014–2015 to date. Wind energy, as always, was the highest contributor to the total renewable power production. The percentage of solar energy produced in the overall renewable power production comes next to wind and is typically reduced during the monsoon months. The definite improvement in wind energy production can be associated with a “good” monsoon. Cyclonic action during these months also facilitates high-speed winds. Monsoon winds play a significant part in the uptick in wind power production, especially in the southern states of the country.

Estimation of gross electricity generation from renewable energy

Table 17 shows an estimation of gross electricity generation from renewable energy based on the 2015 report of the National Institution for Transforming India (NITI Aayog) [ 51 ]. It is predicted that the share of renewable power will be 10.2% by 2022, but renewable power technologies contributed a record of 13.4% to the cumulative power production in India as of the 31st of August 2018. The power ministry report shows that India generated 122.10 TWh and out of the total electricity produced, renewables generated 16.30 TWh as on the 31st of August 2018. According to the India Brand Equity Foundation report, it is anticipated that by the year 2040, around 49% of total electricity will be produced using renewable energy.

Current achievements in renewable energy 2017–2018

India cares for the planet and has taken a groundbreaking journey in renewable energy through the last 4 years [ 52 , 53 ]. A dedicated ministry along with financial and technical institutions have helped India in the promotion of renewable energy and diversification of its energy mix. The country is engaged in expanding the use of clean energy sources and has already undertaken several large-scale sustainable energy projects to ensure a massive growth of green energy.

1. India doubled its renewable power capacity in the last 4 years. The cumulative renewable power capacity in 2013–2014 reached 35,500 MW and rose to 70,000 MW in 2017–2018.

2. India stands in the fourth and sixth position regarding the cumulative installed capacity in the wind and solar sector, respectively. Furthermore, its cumulative installed renewable capacity stands in fifth position globally as of the 31st of December 2018.

3. As said above, the cumulative renewable energy capacity target for 2022 is given as 175 GW. For 2017–2018, the cumulative installed capacity amounted to 70 GW, the capacity under implementation is 15 GW and the tendered capacity was 25 GW. The target, the installed capacity, the capacity under implementation, and the tendered capacity are shown in Fig. 4 .

4. There is tremendous growth in solar power. The cumulative installed solar capacity increased by more than eight times in the last 4 years from 2.630 GW (2013–2014) to 22 GW (2017–2018). As of the 31st of December 2018, the installed capacity amounted to 25.2122 GW.

5. The renewable electricity generated in 2017–2018 was 101839 BUs.

6. The country published competitive bidding guidelines for the production of renewable power. It also discovered the lowest tariff and transparent bidding method and resulted in a notable decrease in per unit cost of renewable energy.

7. In 21 states, there are 41 solar parks with a cumulative capacity of more than 26,144 MW that have already been approved by the MNRE. The Kurnool solar park was set up with 1000 MW; and with 2000 MW the largest solar park of Pavagada (Karnataka) is currently under installation.

8. The target for solar power (ground mounted) for 2018–2019 is given as 10 GW, and solar power (Rooftop) as 1 GW.

9. MNRE doubled the target for solar parks (projects of 500 MW or more) from 20 to 40 GW.

10. The cumulative installed capacity of wind power increased by 1.6 times in the last 4 years. In 2013–2014, it amounted to 21 GW, from 2017 to 2018 it amounted to 34 GW, and as of 31st of December 2018, it reached 35.138 GW. This shows that achievements were completed in wind power use.

11. An offshore wind policy was announced. Thirty-four companies (most significant global and domestic wind power players) competed in the “expression of interest” (EoI) floated on the plan to set up India’s first mega offshore wind farm with a capacity of 1 GW.

12. 682 MW small hydropower projects were installed during the last 4 years along with 600 watermills (mechanical applications) and 132 projects still under development.

13. MNRE is implementing green energy corridors to expand the transmission system. 9400 km of green energy corridors are completed or under implementation. The cost spent on it was INR 10141 crore (101,410 Million INR = 1425.01 USD). Furthermore, the total capacity of 19,000 MVA substations is now planned to be complete by March 2020.

14. MNRE is setting up solar pumps (off-grid application), where 90% of pumps have been set up as of today and between 2014–2015 and 2017–2018. Solar street lights were more than doubled. Solar home lighting systems have been improved by around 1.5 times. More than 2,575,000 solar lamps have been distributed to students. The details are illustrated in Fig. 5 .

15. From 2014–2015 to 2017–2018, more than 2.5 lakh (0.25 million) biogas plants were set up for cooking in rural homes to enable families by providing them access to clean fuel.

16. New policy initiatives revised the tariff policy mandating purchase and generation obligations (RPO and RGO). Four wind and solar inter-state transmission were waived; charges were planned, the RPO trajectory for 2022 and renewable energy policy was finalized.

17. Expressions of interest (EoI) were invited for installing solar photovoltaic manufacturing capacities associated with the guaranteed off-take of 20 GW. EoI indicated 10 GW floating solar energy plants.

18. Policy for the solar-wind hybrid was announced. Tender for setting up 2 GW solar-wind hybrid systems in existing projects was invited.

19. To facilitate R&D in renewable power technology, a National lab policy on testing, standardization, and certification was announced by the MNRE.

20. The Surya Mitra program was conducted to train college graduates in the installation, commissioning, operations, and management of solar panels. The International Solar Alliance (ISA) headquarters in India (Gurgaon) will be a new commencement for solar energy improvement in India.

21. The renewable sector has become considerably more attractive for foreign and domestic investors, and the country expects to attract up to USD 80 billion in the next 4 years from 2018–2019 to 2021–2022.

22. The solar power capacity expanded by more than eight times from 2.63 GW in 2013–2014 to 22 GW in 2017–2018.

23. A bidding for 115 GW renewable energy projects up to March 2020 was announced.

24. The Bureau of Indian Standards (BIS) acting for system/components of solar PV was established.

25. To recognize and encourage innovative ideas in renewable energy sectors, the Government provides prizes and awards. Creative ideas/concepts should lead to prototype development. The Name of the award is “Abhinav Soch-Nayi Sambhawanaye,” which means Innovative ideas—New possibilities.

Renewable energy target, installed capacity, under implementation and tendered [ 52 ]

Off-grid solar applications [ 52 ]

Solar energy

Under the National Solar Mission, the MNRE has updated the objective of grid-connected solar power projects from 20 GW by the year 2021–2022 to 100 GW by the year 2021–2022. In 2008–2009, it reached just 6 MW. The “Made in India” initiative to promote domestic manufacturing supported this great height in solar installation capacity. Currently, India has the fifth highest solar installed capacity worldwide. By the 31st of December 2018, solar energy had achieved 25,212.26 MW against the target of 2022, and a further 22.8 GW of capacity has been tendered out or is under current implementation. MNRE is preparing to bid out the remaining solar energy capacity every year for the periods 2018–2019 and 2019–2020 so that bidding may contribute with 100 GW capacity additions by March 2020. In this way, 2 years for the completion of projects would remain. Tariffs will be determined through the competitive bidding process (reverse e-auction) to bring down tariffs significantly. The lowest solar tariff was identified to be INR 2.44 per kWh in July 2018. In 2010, solar tariffs amounted to INR 18 per kWh. Over 100,000 lakh (10,000 million) acres of land had been classified for several planned solar parks, out of which over 75,000 acres had been obtained. As of November 2018, 47 solar parks of a total capacity of 26,694 MW were established. The aggregate capacity of 4195 MW of solar projects has been commissioned inside various solar parks (floating solar power). Table 18 shows the capacity addition compared to the target. It indicates that capacity addition increased exponentially.

Wind energy

As of the 31st of December 2018, the total installed capacity of India amounted to 35,138.15 MW compared to a target of 60 GW by 2022. India is currently in fourth position in the world for installed capacity of wind power. Moreover, around 9.4 GW capacity has been tendered out or is under current implementation. The MNRE is preparing to bid out for A 10 GW wind energy capacity every year for 2018–2019 and 2019–2020, so that bidding will allow for 60 GW capacity additions by March 2020, giving the remaining two years for the accomplishment of the projects. The gross wind energy potential of the country now reaches 302 GW at a 100 m above-ground level. The tariff administration has been changed from feed-in-tariff (FiT) to the bidding method for capacity addition. On the 8th of December 2017, the ministry published guidelines for a tariff-based competitive bidding rule for the acquisition of energy from grid-connected wind energy projects. The developed transparent process of bidding lowered the tariff for wind power to its lowest level ever. The development of the wind industry has risen in a robust ecosystem ensuring project execution abilities and a manufacturing base. State-of-the-art technologies are now available for the production of wind turbines. All the major global players in wind power have their presence in India. More than 12 different companies manufacture more than 24 various models of wind turbines in India. India exports wind turbines and components to the USA, Europe, Australia, Brazil, and other Asian countries. Around 70–80% of the domestic production has been accomplished with strong domestic manufacturing companies. Table 19 lists the capacity addition compared to the target for the capacity addition. Furthermore, electricity generation from the wind-based capacity has improved, even though there was a slowdown of new capacity in the first half of 2018–2019 and 2017–2018.

The national energy storage mission—2018

The country is working toward a National Energy Storage Mission. A draft of the National Energy Storage Mission was proposed in February 2018 and initiated to develop a comprehensive policy and regulatory framework. During the last 4 years, projects included in R&D worth INR 115.8 million (USD 1.66 million) in the domain of energy storage have been launched, and a corpus of INR 48.2 million (USD 0.7 million) has been issued. India’s energy storage mission will provide an opportunity for globally competitive battery manufacturing. By increasing the battery manufacturing expertise and scaling up its national production capacity, the country can make a substantial economic contribution in this crucial sector. The mission aims to identify the cumulative battery requirements, total market size, imports, and domestic manufacturing. Table 20 presents the economic opportunity from battery manufacturing given by the National Institution for Transforming India, also called NITI Aayog, which provides relevant technical advice to central and state governments while designing strategic and long-term policies and programs for the Indian government.

Small hydropower—3-year action agenda—2017

Hydro projects are classified as large hydro, small hydro (2 to 25 MW), micro-hydro (up to 100 kW), and mini-hydropower (100 kW to 2 MW) projects. Whereas the estimated potential of SHP is 20 GW, the 2022 target for India in SHP is 5 GW. As of the 31st of December 2018, the country has achieved 4.5 GW and this production is constantly increasing. The objective, which was planned to be accomplished through infrastructure project grants and tariff support, was included in the NITI Aayog’s 3-year action agenda (2017–2018 to 2019–2020), which was published on the 1st of August 2017. MNRE is providing central financial assistance (CFA) to set up small/micro hydro projects both in the public and private sector. For the identification of new potential locations, surveys and comprehensive project reports are elaborated, and financial support for the renovation and modernization of old projects is provided. The Ministry has established a dedicated completely automatic supervisory control and data acquisition (SCADA)—based on a hydraulic turbine R&D laboratory at the Alternate Hydro Energy Center (AHEC) at IIT Roorkee. The establishment cost for the lab was INR 40 crore (400 million INR, 95.62 Million USD), and the laboratory will serve as a design and validation facility. It investigates hydro turbines and other hydro-mechanical devices adhering to national and international standards [ 54 , 55 ]. Table 21 shows the target and achievements from 2007–2008 to 2018–2019.

National policy regarding biofuels—2018

Modernization has generated an opportunity for a stable change in the use of bioenergy in India. MNRE amended the current policy for biomass in May 2018. The policy presents CFA for projects using biomass such as agriculture-based industrial residues, wood produced through energy plantations, bagasse, crop residues, wood waste generated from industrial operations, and weeds. Under the policy, CFA will be provided to the projects at the rate of INR 2.5 million (USD 35,477.7) per MW for bagasse cogeneration and INR 5 million (USD 70,955.5) per MW for non-bagasse cogeneration. The MNRE also announced a memorandum in November 2018 considering the continuation of the concessional customs duty certificate (CCDC) to set up projects for the production of energy using non-conventional materials such as bio-waste, agricultural, forestry, poultry litter, agro-industrial, industrial, municipal, and urban wastes. The government recently established the National policy on biofuels in August 2018. The MNRE invited an expression of interest (EOI) to estimate the potential of biomass energy and bagasse cogeneration in the country. A program to encourage the promotion of biomass-based cogeneration in sugar mills and other industries was also launched in May 2018. Table 22 shows how the biomass power target and achievements are expected to reach 10 GW of the target of 2022 before the end of 2019.

The new national biogas and organic manure program (NNBOMP)—2018

The National biogas and manure management programme (NBMMP) was launched in 2012–2013. The primary objective was to provide clean gaseous fuel for cooking, where the remaining slurry was organic bio-manure which is rich in nitrogen, phosphorus, and potassium. Further, 47.5 lakh (4.75 million) cumulative biogas plants were completed in 2014, and increased to 49.8 lakh (4.98 million). During 2017–2018, the target was to establish 1.10 lakh biogas plants (1.10 million), but resulted in 0.15 lakh (0.015 million). In this way, the cost of refilling the gas cylinders with liquefied petroleum gas (LPG) was greatly reduced. Likewise, tons of wood/trees were protected from being axed, as wood is traditionally used as a fuel in rural and semi-urban households. Biogas is a viable alternative to traditional cooking fuels. The scheme generated employment for almost 300 skilled laborers for setting up the biogas plants. By 30th of May 2018, the Ministry had issued guidelines for the implementation of the NNBOMP during the period 2017–2018 to 2019–2020 [ 56 ].

The off-grid and decentralized solar photovoltaic application program—2018

The program deals with the energy demand through the deployment of solar lanterns, solar streetlights, solar home lights, and solar pumps. The plan intended to reach 118 MWp of off-grid PV capacity by 2020. The sanctioning target proposed outlay was 50 MWp by 2017–2018 and 68 MWp by 2019–2020. The total estimated cost amounted to INR 1895 crore (18950 Million INR, 265.547 million USD), and the ministry wanted to support 637 crores (6370 million INR, 89.263 million USD) by its central finance assistance. Solar power plants with a 25 KWp size were promoted in those areas where grid power does not reach households or is not reliable. Public service institutions, schools, panchayats, hostels, as well as police stations will benefit from this scheme. Solar study lamps were also included as a component in the program. Thirty percent of financial assistance was provided to solar power plants. Every student should bear 15% of the lamp cost, and the ministry wanted to support the remaining 85%. As of October 2018, lantern and lamps of more than 40 Lakhs (4 million), home lights of 16.72 lakhs (1.672 million) number, street lights of 6.40 lakhs (0.64 million), solar pumps of 1.96 lakhs (0.196 million), and 187.99 MWp stand-alone devices had been installed [ 57 , 58 ].

Major government initiatives for renewable energy

Technological initiatives.

The Technology Development and Innovation Policy (TDIP) released on the 6th of October 2017 was endeavored to promote research, development, and demonstration (RD&D) in the renewable energy sector [ 59 ]. RD&D intended to evaluate resources, progress in technology, commercialization, and the presentation of renewable energy technologies across the country. It aimed to produce renewable power devices and systems domestically. The evaluation of standards and resources, processes, materials, components, products, services, and sub-systems was carried out through RD&D. A development of the market, efficiency improvements, cost reductions, and a promotion of commercialization (scalability and bankability) were achieved through RD&D. Likewise, the percentage of renewable energy in the total electricity mix made it self-sustainable, industrially competitive, and profitable through RD&D. RD&D also supported technology development and demonstration in wind, solar, wind-solar hybrid, biofuel, biogas, hydrogen fuel cells, and geothermal energies. RD&D supported the R&D units of educational institutions, industries, and non-government organizations (NGOs). Sharing expertise, information, as well as institutional mechanisms for collaboration was realized by use of the technology development program (TDP). The various people involved in this program were policymakers, industrial innovators, associated stakeholders and departments, researchers, and scientists. Renowned R&D centers in India are the National Institute of Solar Energy (NISE), Gurgaon, the National Institute of Bio-Energy (NIBE), Kapurthala, and the National Institute of Wind Energy (NIWE), Chennai. The TDP strategy encouraged the exploration of innovative approaches and possibilities to obtain long-term targets. Likewise, it efficiently supported the transformation of knowledge into technology through a well-established monitoring system for the development of renewable technology that meets the electricity needs of India. The research center of excellence approved the TDI projects, which were funded to strengthen R&D. Funds were provided for conducting training and workshops. The MNRE is now preparing a database of R&D accomplishments in the renewable energy sector.

The Impacting Research Innovation and Technology (IMPRINT) program seeks to develop engineering and technology (prototype/process development) on a national scale. IMPRINT is steered by the Indian Institute of Technologies (IITs) and Indian Institute of science (IISCs). The expansion covers all areas of engineering and technology including renewable technology. The ministry of human resource development (MHRD) finances up to 50% of the total cost of the project. The remaining costs of the project are financed by the ministry (MNRE) via the RD&D program for renewable projects. Currently (2018–2019), five projects are under implementation in the area of solar thermal systems, storage for SPV, biofuel, and hydrogen and fuel cells which are funded by the MNRE (36.9 million INR, 0.518426 Million USD) and IMPRINT. Development of domestic technology and quality control are promoted through lab policies that were published on the 7th of December 2017. Lab policies were implemented to test, standardize, and certify renewable energy products and projects. They supported the improvement of the reliability and quality of the projects. Furthermore, Indian test labs are strengthened in line with international standards and practices through well-established lab policies. From 2015, the MNRE has provided “The New and Renewable Energy Young Scientist’s Award” to researchers/scientists who demonstrate exceptional accomplishments in renewable R&D.

Financial initiatives

One hundred percent financial assistance is granted by the MNRE to the government and NGOs and 50% financial support to the industry. The policy framework was developed to guide the identification of the project, the formulation, monitoring appraisal, approval, and financing. Between 2012 and 2017, a 4467.8 million INR, 62.52 Million USD) support was granted by the MNRE. The MNRE wanted to double the budget for technology development efforts in renewable energy for the current three-year plan period. Table 23 shows that the government is spending more and more for the development of the renewable energy sector. Financial support was provided to R&D projects. Exceptional consideration was given to projects that worked under extreme and hazardous conditions. Furthermore, financial support was applied to organizing awareness programs, demonstrations, training, workshops, surveys, assessment studies, etc. Innovative approaches will be rewarded with cash prizes. The winners will be presented with a support mechanism for transforming their ideas and prototypes into marketable commodities such as start-ups for entrepreneur development. Innovative projects will be financed via start-up support mechanisms, which will include an investment contract with investors. The MNRE provides funds to proposals for investigating policies and performance analyses related to renewable energy.

Technology validation and demonstration projects and other innovative projects with regard to renewables received a financial assistance of 50% of the project cost. The CFA applied to partnerships with industry and private institutions including engineering colleges. Private academic institutions, accredited by a government accreditation body, were also eligible to receive a 50% support. The concerned industries and institutions should meet the remaining 50% expenditure. The MNRE allocated an INR 3762.50 crore (INR 37625 million, 528.634 million USD) for the grid interactive renewable sources and an INR 1036.50 crore (INR 10365 million, 145.629 million USD) for off-grid/distributed and decentralized renewable power for the year 2018–2019 [ 60 ]. The MNRE asked the Reserve Bank of India (RBI), attempting to build renewable power projects under “priority sector lending” (priority lending should be done for renewable energy projects and without any limit) and to eliminate the obstacles in the financing of renewable energy projects. In July 2018, the Ministry of Finance announced that it would impose a 25% safeguard duty on solar panels and modules imported from China and Malaysia for 1 year. The quantum of tax might be reduced to 20% for the next 6 months, and 15% for the following 6 months.

Policy and regulatory framework initiatives

The regulatory interventions for the development of renewable energy sources are (a) tariff determination, (b) defining RPO, (c) promoting grid connectivity, and (d) promoting the expansion of the market.

Tariff policy amendments—2018

On the 30th of May 2018, the MoP released draft amendments to the tariff policy. The objective of these policies was to promote electricity generation from renewables. MoP in consultation with MNRE announced the long-term trajectory for RPO, which is represented in Table 24 . The State Electricity Regulatory Commission (SERC) achieved a favorable and neutral/off-putting effect in the growth of the renewable power sector through their RPO regulations in consultation with the MNRE. On the 25th of May 2018, the MNRE created an RPO compliance cell to reach India’s solar and wind power goals. Due to the absence of implementation of RPO regulations, several states in India did not meet their specified RPO objectives. The cell will operate along with the Central Electricity Regulatory Commission (CERC) and SERCs to obtain monthly statements on RPO compliance. It will also take up non-compliance associated concerns with the relevant officials.

Repowering policy—2016

On the 09th of August 2016, India announced a “repowering policy” for wind energy projects. An about 27 GW turnaround was possible according to the policy. This policy supports the replacing of aging wind turbines with more modern and powerful units (fewer, larger, taller) to raise the level of electricity generation. This policy seeks to create a simplified framework and to promote an optimized use of wind power resources. It is mandatory because the up to the year 2000 installed wind turbines were below 500 kW in sites where high wind potential might be achieved. It will be possible to obtain 3000 MW from the same location once replacements are in place. The policy was initially applied for the one MW installed capacity of wind turbines, and the MNRE will extend the repowering policy to other projects in the future based on experience. Repowering projects were implemented by the respective state nodal agencies/organizations that were involved in wind energy promotion in their states. The policy provided an exception from the Power Purchase Agreement (PPA) for wind farms/turbines undergoing repowering because they could not fulfill the requirements according to the PPA during repowering. The repowering projects may avail accelerated depreciation (AD) benefit or generation-based incentive (GBI) due to the conditions appropriate to new wind energy projects [ 61 ].

The wind-solar hybrid policy—2018

On the 14th of May 2018, the MNRE announced a national wind-solar hybrid policy. This policy supported new projects (large grid-connected wind-solar photovoltaic hybrid systems) and the hybridization of the already available projects. These projects tried to achieve an optimal and efficient use of transmission infrastructure and land. Better grid stability was achieved and the variability in renewable power generation was reduced. The best part of the policy intervention was that which supported the hybridization of existing plants. The tariff-based transparent bidding process was included in the policy. Regulatory authorities should formulate the necessary standards and regulations for hybrid systems. The policy also highlighted a battery storage in hybrid projects for output optimization and variability reduction [ 62 ].

The national offshore wind energy policy—2015

The National Offshore Wind Policy was released in October 2015. On the 19th of June 2018, the MNRE announced a medium-term target of 5 GW by 2022 and a long-term target of 30 GW by 2030. The MNRE called expressions of Interest (EoI) for the first 1 GW of offshore wind (the last date was 08.06.2018). The EoI site is located in Pipavav port at the Gulf of Khambhat at a distance of 23 km facilitating offshore wind (FOWIND) where the consortium deployed light detection and ranging (LiDAR) in November 2017). Pipavav port is situated off the coast of Gujarat. The MNRE had planned to install more such equipment in the states of Tamil Nadu and Gujarat. On the 14 th of December 2018, the MNRE, through the National Institute of Wind Energy (NIWE), called tender for offshore environmental impact assessment studies at intended LIDAR points at the Gulf of Mannar, off the coast of Tamil Nadu for offshore wind measurement. The timeline for initiatives was to firstly add 500 MW by 2022, 2 to 2.5 GW by 2027, and eventually reaching 5 GW between 2028 and 2032. Even though the installation of large wind power turbines in open seas is a challenging task, the government has endeavored to promote this offshore sector. Offshore wind energy would add its contribution to the already existing renewable energy mix for India [ 63 ] .

The feed-in tariff policy—2018

On the 28th of January 2016, the revised tariff policy was notified following the Electricity Act. On the 30th May 2018, the amendment in tariff policy was released. The intentions of this tariff policy are (a) an inexpensive and competitive electricity rate for the consumers; (b) to attract investment and financial viability; (c) to ensure that the perceptions of regulatory risks decrease through predictability, consistency, and transparency of policy measures; (d) development in quality of supply, increased operational efficiency, and improved competition; (e) increase the production of electricity from wind, solar, biomass, and small hydro; (f) peaking reserves that are acceptable in quantity or consistently good in quality or performance of grid operation where variable renewable energy source integration is provided through the promotion of hydroelectric power generation, including pumped storage projects (PSP); (g) to achieve better consumer services through efficient and reliable electricity infrastructure; (h) to supply sufficient and uninterrupted electricity to every level of consumers; and (i) to create adequate capacity, reserves in the production, transmission, and distribution that is sufficient for the reliability of supply of power to customers [ 64 ].

Training and educational initiatives

The MHRD has developed strong renewable energy education and training systems. The National Council for Vocational Training (NCVT) develops course modules, and a Modular Employable Skilling program (MES) in its regular 2-year syllabus to include SPV lighting systems, solar thermal systems, SHP, and provides the certificate for seven trades after the completion of a 2-year course. The seven trades are plumber, fitter, carpenter, welder, machinist, and electrician. The Ministry of Skill Development and Entrepreneurship (MSDE) worked out a national skill development policy in 2015. They provide regular training programs to create various job roles in renewable energy along with the MNRE support through a skill council for green jobs (SCGJ), the National Occupational Standards (NOS), and the Qualification Pack (QP). The SCGJ is promoted by the Confederation of Indian Industry (CII) and the MNRE. The industry partner for the SCGJ is ReNew Power [ 65 , 66 ].

The global status of India in renewable energy

Table 25 shows the RECAI (Renewable Energy Country Attractiveness Index) report of 40 countries. This report is based on the attractiveness of renewable energy investment and deployment opportunities. RECAI is based on macro vitals such as economic stability, investment climate, energy imperatives such as security and supply, clean energy gap, and affordability. It also includes policy enablement such as political stability and support for renewables. Its emphasis lies on project delivery parameters such as energy market access, infrastructure, and distributed generation, finance, cost and availability, and transaction liquidity. Technology potentials such as natural resources, power take-off attractiveness, potential support, technology maturity, and forecast growth are taken into consideration for ranking. India has moved to the fourth position of the RECAI-2018. Indian solar installations (new large-scale and rooftop solar capacities) in the calendar year 2017 increased exponentially with the addition of 9629 MW, whereas in 2016 it was 4313 MW. The warning of solar import tariffs and conflicts between developers and distribution firms are growing investor concerns [ 67 ]. Figure 6 shows the details of the installed capacity of global renewable energy in 2016 and 2017. Globally, 2017 GW renewable energy was installed in 2016, and in 2017, it increased to 2195 GW. Table 26 shows the total capacity addition of top countries until 2017. The country ranked fifth in renewable power capacity (including hydro energy), renewable power capacity (not including hydro energy) in fourth position, concentrating solar thermal power (CSP) and wind power were also in fourth position [ 68 ].

Globally installed capacity of renewable energy in 2017—Global 2018 status report with regard to renewables [ 68 ]

The investment opportunities in renewable energy in India

The investments into renewable energy in India increased by 22% in the first half of 2018 compared to 2017, while the investments in China dropped by 15% during the same period, according to a statement by the Bloomberg New Energy Finance (BNEF), which is shown in Table 27 [ 69 , 70 ]. At this rate, India is expected to overtake China and become the most significant growth market for renewable energy by the end of 2020. The country is eyeing pole position for transformation in renewable energy by reaching 175 GW by 2020. To achieve this target, it is quickly ramping up investments in this sector. The country added more renewable capacity than conventional capacity in 2018 when compared to 2017. India hosted the ISA first official summit on the 11.03.2018 for 121 countries. This will provide a standard platform to work toward the ambitious targets for renewable energy. The summit will emphasize India’s dedication to meet global engagements in a time-bound method. The country is also constructing many sizeable solar power parks comparable to, but larger than, those in China. Half of the earth’s ten biggest solar parks under development are in India.

In 2014, the world largest solar park was the Topaz solar farm in California with a 550 MW facility. In 2015, another operator in California, Solar Star, edged its capacity up to 579 MW. By 2016, India’s Kamuthi Solar Power Project in Tamil Nadu was on top with 648 MW of capacity (set up by the Adani Green Energy, part of the Adani Group, in Tamil Nadu). As of February 2017, the Longyangxia Dam Solar Park in China was the new leader, with 850 MW of capacity [ 71 ]. Currently, there are 600 MW operating units and 1400 MW units under construction. The Shakti Sthala solar park was inaugurated on 01.03.2018 in Pavagada (Karnataka, India) which is expected to become the globe’s most significant solar park when it accomplishes its full potential of 2 GW. Another large solar park with 1.5 GW is scheduled to be built in the Kadappa region [ 72 ]. The progress in solar power is remarkable and demonstrates real clean energy development on the ground.

The Kurnool ultra-mega solar park generated 800 million units (MU) of energy in October 2018 and saved over 700,000 tons of CO 2 . Rainwater was harvested using a reservoir that helps in cleaning solar panels and supplying water. The country is making remarkable progress in solar energy. The Kamuthi solar farm is cleaned each day by a robotic system. As the Indian economy expands, electricity consumption is forecasted to reach 15,280 TWh in 2040. With the government’s intent, green energy objectives, i.e., the renewable sector, grow considerably in an attractive manner with both foreign and domestic investors. It is anticipated to attract investments of up to USD 80 billion in the subsequent 4 years. The government of India has raised its 175 GW target to 225 GW of renewable energy capacity by 2022. The competitive benefit is that the country has sun exposure possible throughout the year and has an enormous hydropower potential. India was also listed fourth in the EY renewable energy country attractive index 2018. Sixty solar cities will be built in India as a section of MNRE’s “Solar cities” program.

In a regular auction, reduction in tariffs cost of the projects are the competitive benefits in the country. India accounts for about 4% of the total global electricity generation capacity and has the fourth highest installed capacity of wind energy and the third highest installed capacity of CSP. The solar installation in India erected during 2015–2016, 2016–2017, 2017–2018, and 2018–2019 was 3.01 GW, 5.52 GW, 9.36 GW, and 6.53 GW, respectively. The country aims to add 8.5 GW during 2019–2020. Due to its advantageous location in the solar belt (400 South to 400 North), the country is one of the largest beneficiaries of solar energy with relatively ample availability. An increase in the installed capacity of solar power is anticipated to exceed the installed capacity of wind energy, approaching 100 GW by 2022 from its current levels of 25.21226 GW as of December 2018. Fast falling prices have made Solar PV the biggest market for new investments. Under the Union Budget 2018–2019, a zero import tax on parts used in manufacturing solar panels was launched to provide an advantage to domestic solar panel companies [ 73 ].

Foreign direct investment (FDI) inflows in the renewable energy sector of India between April 2000 and June 2018 amounted to USD 6.84 billion according to the report of the department of industrial policy and promotion (DIPP). The DIPP was renamed (gazette notification 27.01.2019) the Department for the Promotion of Industry and Internal Trade (DPIIT). It is responsible for the development of domestic trade, retail trade, trader’s welfare including their employees as well as concerns associated with activities in facilitating and supporting business and startups. Since 2014, more than 42 billion USD have been invested in India’s renewable power sector. India reached US$ 7.4 billion in investments in the first half of 2018. Between April 2015 and June 2018, the country received USD 3.2 billion FDI in the renewable sector. The year-wise inflows expanded from USD 776 million in 2015–2016 to USD 783 million in 2016–2017 and USD 1204 million in 2017–2018. Between January to March of 2018, the INR 452 crore (4520 Million INR, 63.3389 million USD) of the FDI had already come in. The country is contributing with financial and promotional incentives that include a capital subsidy, accelerated depreciation (AD), waiver of inter-state transmission charges and losses, viability gap funding (VGF), and FDI up to 100% under the automated track.

The DIPP/DPIIT compiles and manages the data of the FDI equity inflow received in India [ 74 ]. The FDI equity inflow between April 2015 and June 2018 in the renewable sector is illustrated in Fig. 7 . It shows that the 2018–2019 3 months’ FDI equity inflow is half of that of the entire one of 2017–2018. It is evident from the figure that India has well-established FDI equity inflows. The significant FDI investments in the renewable energy sectors are shown in Table 28 . The collaboration between the Asian development bank and Renew Power Ventures private limited with 44.69 million USD ranked first followed by AIRRO Singapore with Diligent power with FDI equity inflow of 44.69 USD million.

The FDI equity inflow received between April 2015 and June 2018 in the renewable energy sector [ 73 ]

Strategies to promote investments

Strategies to promote investments (including FDI) by investors in the renewable sector:

Decrease constraints on FDI; provide open, transparent, and dependable conditions for foreign and domestic firms; and include ease of doing business, access to imports, comparatively flexible labor markets, and safeguard of intellectual property rights.

Establish an investment promotion agency (IPA) that targets suitable foreign investors and connects them as a catalyst with the domestic economy. Assist the IPA to present top-notch infrastructure and immediate access to skilled workers, technicians, engineers, and managers that might be needed to attract such investors. Furthermore, it should involve an after-investment care, recognizing the demonstration effects from satisfied investors, the potential for reinvestments, and the potential for cluster-development due to follow-up investments.

It is essential to consider the targeted sector (wind, solar, SPH or biomass, respectively) for which investments are required.

Establish the infrastructure needed for a quality investor, including adequate close-by transport facilities (airport, ports), a sufficient and steady supply of energy, a provision of a sufficiently skilled workforce, the facilities for the vocational training of specialized operators, ideally designed in collaboration with the investor.

Policy and other support mechanisms such as Power Purchase Agreements (PPA) play an influential role in underpinning returns and restricting uncertainties for project developers, indirectly supporting the availability of investment. Investors in renewable energy projects have historically relied on government policies to give them confidence about the costs necessary for electricity produced—and therefore for project revenues. Reassurance of future power costs for project developers is secured by signing a PPA with either a utility or an essential corporate buyer of electricity.

FiT have been the most conventional approach around the globe over the last decade to stimulate investments in renewable power projects. Set by the government concerned, they lay down an electricity tariff that developers of qualifying new projects might anticipate to receive for the resulting electricity over a long interval (15–20 years). These present investors in the tax equity of renewable power projects with a credit that they can manage to offset the tax burden outside in their businesses.

Table 29 presents the 2018 renewable energy investment report, source-wise, by the significant players in renewables according to the report of the Bloomberg New Energy Finance Report 2018. As per this report, global investment in renewable energy was USD of 279.8 billion in 2017. The top ten in the total global investments are China (126.1 $BN), the USA (40.5 $BN), Japan (13.4 $BN), India (10.9 $BN), Germany (10.4 $BN), Australia (8.5 $BN), UK (7.6 $BN), Brazil (6.0 $BN), Mexico (6.0 $BN), and Sweden (3.7 $BN) [ 75 ]. This achievement was possible since those countries have well-established strategies for promoting investments [ 76 , 77 ].

The appropriate objectives for renewable power expansion and investments are closely related to the Nationally Determined Contributions (NDCs) objectives, the implementation of the NDC, on the road to achieving Paris promises, policy competence, policy reliability, market absorption capacity, and nationwide investment circumstances that are the real purposes for renewable power expansion, which is a significant factor for the investment strategies, as is shown in Table 30 .

The demand for investments for building a Paris-compatible and climate-resilient energy support remains high, particularly in emerging nations. Future investments in energy grids and energy flexibility are of particular significance. The strategies and the comparison chart between China, India, and the USA are presented in Table 31 .

Table 32 shows France in the first place due to overall favorable conditions for renewables, heading the G20 in investment attractiveness of renewables. Germany drops back one spot due to a decline in the quality of the global policy environment for renewables and some insufficiencies in the policy design, as does the UK. Overall, with four European countries on top of the list, Europe, however, directs the way in providing attractive conditions for investing in renewables. Despite high scores for various nations, no single government is yet close to growing a role model. All countries still have significant room for increasing investment demands to deploy renewables at the scale required to reach the Paris objectives. The table shown is based on the Paris compatible long-term vision, the policy environment for renewable energy, the conditions for system integration, the market absorption capacity, and general investment conditions. India moved from the 11th position to the 9th position in overall investments between 2017 and 2018.

A Paris compatible long-term vision includes a de-carbonization plan for the power system, the renewable power ambition, the coal and oil decrease, and the reliability of renewables policies. Direct support policies include medium-term certainty of policy signals, streamlined administrative procedures, ensuring project realization, facilitating the use of produced electricity. Conditions for system integration include system integration-grid codes, system integration-storage promotion, and demand-side management policies. A market absorption capacity includes a prior experience with renewable technologies, a current activity with renewable installations, and a presence of major renewable energy companies. General investment conditions include non-financial determinants, depth of the financial sector as well, as an inflation forecast.

Employment opportunities for citizens in renewable energy in India

Global employment scenario.

According to the 2018 Annual review of the IRENA [ 78 ], global renewable energy employment touched 10.3 million jobs in 2017, an improvement of 5.3% compared with the quantity published in 2016. Many socio-economic advantages derive from renewable power, but employment continues to be exceptionally centralized in a handful of countries, with China, Brazil, the USA, India, Germany, and Japan in the lead. In solar PV employment (3.4 million jobs), China is the leader (65% of PV Jobs) which is followed by Japan, USA, India, Bangladesh, Malaysia, Germany, Philippines, and Turkey. In biofuels employment (1.9 million jobs), Brazil is the leader (41% of PV Jobs) followed by the USA, Colombia, Indonesia, Thailand, Malaysia, China, and India. In wind employment (1.1 million jobs), China is the leader (44% of PV Jobs) followed by Germany, USA, India, UK, Brazil, Denmark, Netherlands, France, and Spain.

Table 33 shows global renewable energy employment in the corresponding technology branches. As in past years, China maintained the most notable number of people employed (3880 million jobs) estimating for 43% of the globe’s total which is shown in Fig. 8 . In India, new solar installations touched a record of 9.6 GW in 2017, efficiently increasing the total installed capacity. The employment in solar PV improved by 36% and reached 164,400 jobs, of which 92,400 represented on-grid use. IRENA determines that the building and installation covered 46% of these jobs, with operations and maintenance (O&M) representing 35% and 19%, individually. India does not produce solar PV because it could be imported from China, which is inexpensive. The market share of domestic companies (Indian supplier to renewable projects) declined from 13% in 2014–2015 to 7% in 2017–2018. If India starts the manufacturing base, more citizens will get jobs in the manufacturing field. India had the world’s fifth most significant additions of 4.1 GW to wind capacity in 2017 and the fourth largest cumulative capacity in 2018. IRENA predicts that jobs in the wind sector stood at 60,500.

Renewable energy employment in selected countries [ 79 ]

The jobs in renewables are categorized into technological development, installation/de-installation, operation, and maintenance. Tables 34 , 35 , 36 , and 37 show the wind industry, solar energy, biomass, and small hydro-related jobs in project development, component manufacturing, construction, operations, and education, training, and research. As technology quickly evolves, workers in all areas need to update their skills through continuing training/education or job training, and in several cases could benefit from professional certification. The advantages of moving to renewable energy are evident, and for this reason, the governments are responding positively toward the transformation to clean energy. Renewable energy can be described as the country’s next employment boom. Renewable energy job opportunities can transform rural economy [ 79 , 80 ]. The renewable energy sector might help to reduce poverty by creating better employment. For example, wind power is looking for specialists in manufacturing, project development, and construction and turbine installation as well as financial services, transportation and logistics, and maintenance and operations.

The government is building more renewable energy power plants that will require a workforce. The increasing investments in the renewable energy sector have the potential to provide more jobs than any other fossil fuel industry. Local businesses and renewable sectors will benefit from this change, as income will increase significantly. Many jobs in this sector will contribute to fixed salaries, healthcare benefits, and skill-building opportunities for unskilled and semi-skilled workers. A range of skilled and unskilled jobs are included in all renewable energy technologies, even though most of the positions in the renewable energy industry demand a skilled workforce. The renewable sector employs semi-skilled and unskilled labor in the construction, operations, and maintenance after proper training. Unskilled labor is employed as truck drivers, guards, cleaning, and maintenance. Semi-skilled labor is used to take regular readings from displays. A lack of consistent data on the potential employment impact of renewables expansion makes it particularly hard to assess the quantity of skilled, semi-skilled, and unskilled personnel that might be needed.

Key findings in renewable energy employment

The findings comprise (a) that the majority of employment in the renewable sector is contract based, and that employees do not benefit from permanent jobs or security. (b) Continuous work in the industry has the potential to decrease poverty. (c) Most poor citizens encounter obstacles to entry-level training and the employment market due to lack of awareness about the jobs and the requirements. (d) Few renewable programs incorporate developing ownership opportunities for the citizens and the incorporation of women in the sector. (e) The inadequacy of data makes it challenging to build relationships between employment in renewable energy and poverty mitigation.

Recommendations for renewable energy employment

When building the capacity, focus on poor people and individuals to empower them with training in operation and maintenance.

Develop and offer training programs for citizens with minimal education and training, who do not fit current programs, which restrict them from working in renewable areas.

Include women in the renewable workforce by providing localized training.

Establish connections between training institutes and renewable power companies to guarantee that (a) trained workers are placed in appropriate positions during and after the completion of the training program and (b) training programs match the requirements of the renewable sector.

Poverty impact assessments might be embedded in program design to know how programs motivate poverty reduction, whether and how they influence the community.

Allow people to have a sense of ownership in renewable projects because this could contribute to the growth of the sector.

The details of the job being offered (part time, full time, contract-based), the levels of required skills for the job (skilled, semi-skilled and unskilled), the socio-economic status of the employee data need to be collected for further analysis.

Conduct investigations, assisted by field surveys, to learn about the influence of renewable energy jobs on poverty mitigation and differences in the standard of living.

Challenges faced by renewable energy in India

The MNRE has been taking dedicated measures for improving the renewable sector, and its efforts have been satisfactory in recognizing various obstacles.

Policy and regulatory obstacles

A comprehensive policy statement (regulatory framework) is not available in the renewable sector. When there is a requirement to promote the growth of particular renewable energy technologies, policies might be declared that do not match with the plans for the development of renewable energy.

The regulatory framework and procedures are different for every state because they define the respective RPOs (Renewable Purchase Obligations) and this creates a higher risk of investments in this sector. Additionally, the policies are applicable for just 5 years, and the generated risk for investments in this sector is apparent. The biomass sector does not have an established framework.

Incentive accelerated depreciation (AD) is provided to wind developers and is evident in developing India’s wind-producing capacity. Wind projects installed more than 10 years ago show that they are not optimally maintained. Many owners of the asset have built with little motivation for tax benefits only. The policy framework does not require the maintenance of the wind projects after the tax advantages have been claimed. There is no control over the equipment suppliers because they undertake all wind power plant development activities such as commissioning, operation, and maintenance. Suppliers make the buyers pay a premium and increase the equipment cost, which brings burden to the buyer.

Furthermore, ready-made projects are sold to buyers. The buyers are susceptible to this trap to save income tax. Foreign investors hesitate to invest because they are exempted from the income tax.

Every state has different regulatory policy and framework definitions of an RPO. The RPO percentage specified in the regulatory framework for various renewable sources is not precise.

RPO allows the SERCs and certain private firms to procure only a part of their power demands from renewable sources.

RPO is not imposed on open access (OA) and captive consumers in all states except three.

RPO targets and obligations are not clear, and the RPO compliance cell has just started on 22.05.2018 to collect the monthly reports on compliance and deal with non-compliance issues with appropriate authorities.

Penalty mechanisms are not specified and only two states in India (Maharashtra and Rajasthan) have some form of penalty mechanisms.

The parameter to determine the tariff is not transparent in the regulatory framework and many SRECs have established a tariff for limited periods. The FiT is valid for only 5 years, and this affects the bankability of the project.

Many SERCs have not decided on adopting the CERC tariff that is mentioned in CERCs regulations that deal with terms and conditions for tariff determinations. The SERCs have considered the plant load factor (PLF) because it varies across regions and locations as well as particular technology. The current framework does not fit to these issues.

Third party sale (TPS) is not allowed because renewable generators are not allowed to sell power to commercial consumers. They have to sell only to industrial consumers. The industrial consumers have a low tariff and commercial consumers have a high tariff, and SRCS do not allow OA. This stops the profit for the developers and investors.

Institutional obstacles

Institutes, agencies stakeholders who work under the conditions of the MNRE show poor inter-institutional coordination. The progress in renewable energy development is limited by this lack of cooperation, coordination, and delays. The delay in implementing policies due to poor coordination, decrease the interest of investors to invest in this sector.

The single window project approval and clearance system is not very useful and not stable because it delays the receiving of clearances for the projects ends in the levy of a penalty on the project developer.

Pre-feasibility reports prepared by concerned states have some deficiency, and this may affect the small developers, i.e., the local developers, who are willing to execute renewable projects.

The workforce in institutes, agencies, and ministries is not sufficient in numbers.

Proper or well-established research centers are not available for the development of renewable infrastructure.

Customer care centers to guide developers regarding renewable projects are not available.

Standards and quality control orders have been issued recently in 2018 and 2019 only, and there are insufficient institutions and laboratories to give standards/certification and validate the quality and suitability of using renewable technology.

Financial and fiscal obstacles

There are a few budgetary constraints such as fund allocation, and budgets that are not released on time to fulfill the requirement of developing the renewable sector.

The initial unit capital costs of renewable projects are very high compared to fossil fuels, and this leads to financing challenges and initial burden.

There are uncertainties related to the assessment of resources, lack of technology awareness, and high-risk perceptions which lead to financial barriers for the developers.

The subsidies and incentives are not transparent, and the ministry might reconsider subsidies for renewable energy because there was a sharp fall in tariffs in 2018.

Power purchase agreements (PPA) signed between the power purchaser and power generators on pre-determined fixed tariffs are higher than the current bids (Economic survey 2017–2018 and union budget on the 01.02.2019). For example, solar power tariff dropped to 2.44 INR (0. 04 USD) per unit in May 2017, wind power INR 3.46 per unit in February 2017, and 2.64 INR per unit in October 2017.

Investors feel that there is a risk in the renewable sector as this sector has lower gross returns even though these returns are relatively high within the market standards.

There are not many developers who are interested in renewable projects. While newly established developers (small and local developers) do not have much of an institutional track record or financial input, which are needed to develop the project (high capital cost). Even moneylenders consider it risky and are not ready to provide funding. Moneylenders look exclusively for contractors who have much experience in construction, well-established suppliers with proven equipment and operators who have more experience.

If the performance of renewable projects, which show low-performance, faces financial obstacles, they risks the lack of funding of renewable projects.

Financial institutions such as government banks or private banks do not have much understanding or expertise in renewable energy projects, and this imposes financial barriers to the projects.

Delay in payment by the SERCs to the developers imposes debt burden on the small and local developers because moneylenders always work with credit enhancement mechanisms or guarantee bonds signed between moneylenders and the developers.

Market obstacles

Subsidies are adequately provided to conventional fossil fuels, sending the wrong impression that power from conventional fuels is of a higher priority than that from renewables (unfair structure of subsidies)

There are four renewable markets in India, the government market (providing budgetary support to projects and purchase the output of the project), the government-driven market (provide budgetary support or fiscal incentives to promote renewable energy), the loan market (taking loan to finance renewable based applications), and the cash market (buying renewable-based applications to meet personal energy needs by individuals). There is an inadequacy in promoting the loan market and cash market in India.

The biomass market is facing a demand-supply gap which results in a continuous and dramatic increase in biomass prices because the biomass supply is unreliable (and, as there is no organized market for fuel), and the price fluctuations are very high. The type of biomass is not the same in all the states of India, and therefore demand and price elasticity is high for biomass.

Renewable power was calculated based on cost-plus methods (adding direct material cost, direct labor cost, and product overhead cost). This does not include environmental cost and shields the ecological benefits of clean and green energy.

There is an inadequate evacuation infrastructure and insufficient integration of the grid, which affects the renewable projects. SERCs are not able to use all generated power to meet the needs because of the non-availability of a proper evacuation infrastructure. This has an impact on the project, and the SERCs are forced to buy expensive power from neighbor states to fulfill needs.

Extending transmission lines is not possible/not economical for small size projects, and the seasonality of generation from such projects affect the market.

There are few limitations in overall transmission plans, distribution CapEx plans, and distribution licenses for renewable power. Power evacuation infrastructure for renewable energy is not included in the plans.

Even though there is an increase in capacity for the commercially deployed renewable energy technology, there is no decline in capital cost. This cost of power also remains high. The capital cost quoted by the developers and providers of equipment is too high due to exports of machinery, inadequate built up capacity, and cartelization of equipment suppliers (suppliers join together to control prices and limit competition).

There is no adequate supply of land, for wind, solar, and solar thermal power plants, which lead to poor capacity addition in many states.

Technological obstacles

Every installation of a renewable project contributes to complex risk challenges from environmental uncertainties, natural disasters, planning, equipment failure, and profit loss.

MNRE issued the standardization of renewable energy projects policy on the 11th of December 2017 (testing, standardization, and certification). They are still at an elementary level as compared to international practices. Quality assurance processes are still under starting conditions. Each success in renewable energy is based on concrete action plans for standards, testing and certification of performance.

The quality and reliability of manufactured components, imported equipment, and subsystems is essential, and hence quality infrastructure should be established. There is no clear document related to testing laboratories, referral institutes, review mechanism, inspection, and monitoring.

There are not many R&D centers for renewables. Methods to reduce the subsidies and invest in R&D lagging; manufacturing facilities are just replicating the already available technologies. The country is dependent on international suppliers for equipment and technology. Spare parts are not manufactured locally and hence they are scarce.

Awareness, education, and training obstacles

There is an unavailability of appropriately skilled human resources in the renewable energy sector. Furthermore, it faces an acute workforce shortage.

After installation of renewable project/applications by the suppliers, there is no proper follow-up or assistance for the workers in the project to perform maintenance. Likewise, there are not enough trained and skilled persons for demonstrating, training, operation, and maintenance of the plant.

There is inadequate knowledge in renewables, and no awareness programs are available to the general public. The lack of awareness about the technologies is a significant obstacle in acquiring vast land for constructing the renewable plant. Moreover, people using agriculture lands are not prepared to give their land to construct power plants because most Indians cultivate plants.

The renewable sector depends on the climate, and this varying climate also imposes less popularity of renewables among the people.

The per capita income is low, and the people consider that the cost of renewables might be high and they might not be able to use renewables.

The storage system increases the cost of renewables, and people believe it too costly and are not ready to use them.

The environmental benefits of renewable technologies are not clearly understood by the people and negative perceptions are making renewable technologies less prevalent among them.

Environmental obstacles

A single wind turbine does not occupy much space, but many turbines are placed five to ten rotor diameters from each other, and this occupies more area, which include roads and transmission lines.

In the field of offshore wind, the turbines and blades are bigger than onshore wind turbines, and they require a substantial amount of space. Offshore installations affect ocean activities (fishing, sand extraction, gravel extraction, oil extraction, gas extraction, aquaculture, and navigation). Furthermore, they affect fish and other marine wildlife.

Wind turbines influence wildlife (birds and bats) because of the collisions with them and due to air pressure changes caused by wind turbines and habitat disruption. Making wind turbines motionless during times of low wind can protect birds and bats but is not practiced.

Sound (aerodynamic, mechanical) and visual impacts are associated with wind turbines. There is poor practice by the wind turbine developers regarding public concerns. Furthermore, there are imperfections in surfaces and sound—absorbent material which decrease the noise from turbines. The shadow flicker effect is not taken as severe environmental impact by the developers.

Sometimes wind turbine material production, transportation of materials, on-site construction, assembling, operation, maintenance, dismantlement, and decommissioning may be associated with global warming, and there is a lag in this consideration.

Large utility-scale solar plants require vast lands that increase the risk of land degradation and loss of habitat.

The PV cell manufacturing process includes hazardous chemicals such as 1-1-1 Trichloroethene, HCL, H 2 SO 4 , N 2 , NF, and acetone. Workers face risks resulting from inhaling silicon dust. The manufacturing wastes are not disposed of properly. Proper precautions during usage of thin-film PV cells, which contain cadmium—telluride, gallium arsenide, and copper-indium-gallium-diselenide are missing. These materials create severe public health threats and environmental threats.

Hydroelectric power turbine blades kill aquatic ecosystems (fish and other organisms). Moreover, algae and other aquatic weeds are not controlled through manual harvesting or by introducing fish that can eat these plants.

Discussion and recommendations based on the research

Policy and regulation advancements.

The MNRE should provide a comprehensive action plan or policy for the promotion of the renewable sector in its regulatory framework for renewables energy. The action plan can be prepared in consultation with SERCs of the country within a fixed timeframe and execution of the policy/action plan.

The central and state government should include a “Must run status” in their policy and follow it strictly to make use of renewable power.

A national merit order list for renewable electricity generation will reduce power cost for the consumers. Such a merit order list will help in ranking sources of renewable energy in an ascending order of price and will provide power at a lower cost to each distribution company (DISCOM). The MNRE should include that principle in its framework and ensure that SERCs includes it in their regulatory framework as well.

SERCs might be allowed to remove policies and regulatory uncertainty surrounding renewable energy. SERCs might be allowed to identify the thrust areas of their renewable energy development.

There should be strong initiatives from municipality (local level) approvals for renewable energy-based projects.

Higher market penetration is conceivable only if their suitable codes and standards are adopted and implemented. MNRE should guide minimum performance standards, which incorporate reliability, durability, and performance.

A well-established renewable energy certificates (REC) policy might contribute to an efficient funding mechanism for renewable energy projects. It is necessary for the government to look at developing the REC ecosystem.

The regulatory administration around the RPO needs to be upgraded with a more efficient “carrot and stick” mechanism for obligated entities. A regulatory mechanism that both remunerations compliance and penalizes for non-compliance may likely produce better results.

RECs in India should only be traded on exchange. Over-the-counter (OTC) or off-exchange trading will potentially allow greater participation in the market. A REC forward curve will provide further price determination to the market participants.

The policymakers should look at developing and building the REC market.

Most states have defined RPO targets. Still, due to the absence of implemented RPO regulations and the inadequacy of penalties when obligations are not satisfied, several of the state DISCOMs are not complying completely with their RPO targets. It is necessary that all states adhere to the RPO targets set by respective SERCs.

The government should address the issues such as DISCOM financials, must-run status, problems of transmission and evacuation, on-time payments and payment guarantees, and deemed generation benefits.

Proper incentives should be devised to support utilities to obtain power over and above the RPO mandated by the SERC.

The tariff orders/FiTs must be consistent and not restricted for a few years.

Transmission requirements

The developers are worried that transmission facilities are not keeping pace with the power generation. Bays at the nearest substations are occupied, and transmission lines are already carrying their full capacity. This is due to the lack of coordination between MNRE and the Power Grid Corporation of India (PGCIL) and CEA. Solar Corporation of India (SECI) is holding auctions for both wind and solar projects without making sure that enough evacuation facilities are available. There is an urgent need to make evacuation plans.

The solution is to develop numerous substations and transmission lines, but the process will take considerably longer time than the currently under-construction projects take to get finished.

In 2017–2018, transmission lines were installed under the green energy corridor project by the PGCIL, with 1900 circuit km targeted in 2018–2019. The implementation of the green energy corridor project explicitly meant to connect renewable energy plants to the national grid. The budget allocation of INR 6 billion for 2018–2019 should be increased to higher values.

The mismatch between MNRE and PGCIL, which are responsible for inter-state transmission, should be rectified.

State transmission units (STUs) are responsible for the transmission inside the states, and their fund requirements to cover the evacuation and transmission infrastructure for renewable energy should be fulfilled. Moreover, STUs should be penalized if they fail to fulfill their responsibilities.

The coordination and consultation between the developers (the nodal agency responsible for the development of renewable energy) and STUs should be healthy.

Financing the renewable sector

The government should provide enough budget for the clean energy sector. China’s annual budget for renewables is 128 times higher than India’s. In 2017, China spent USD 126.6 billion (INR 9 lakh crore) compared to India’s USD 10.9 billion (INR 75500 crore). In 2018, budget allocations for grid interactive wind and solar have increased but it is not sufficient to meet the renewable target.

The government should concentrate on R&D and provide a surplus fund for R&D. In 2017, the budget allotted was an INR 445 crore, which was reduced to an INR 272.85 crore in 2016. In 2017–2018, the initial allocation was an INR 144 crore that was reduced to an INR 81 crore during the revised estimates. Even the reduced amounts could not be fully used, there is an urgent demand for regular monitoring of R&D and the budget allocation.

The Goods and Service Tax (GST) that was introduced in 2017 worsened the industry performance and has led to an increase in costs and poses a threat to the viability of the ongoing projects, ultimately hampering the target achievement. These GST issues need to be addressed.

Including the renewable sector as a priority sector would increase the availability of credit and lead to a more substantial participation by commercial banks.

Mandating the provident funds and insurance companies to invest the fixed percentage of their portfolio into the renewable energy sector.

Banks should allow an interest rebate on housing loans if the owner is installing renewable applications such as solar lights, solar water heaters, and PV panels in his house. This will encourage people to use renewable energy. Furthermore, income tax rebates also can be given to individuals if they are implementing renewable energy applications.

Improvement in manufacturing/technology

The country should move to domestic manufacturing. It imports 90% of its solar cell and module requirements from Malaysia, China, and Taiwan, so it is essential to build a robust domestic manufacturing basis.

India will provide “safeguard duty” for merely 2 years, and this is not adequate to build a strong manufacturing basis that can compete with the global market. Moreover, safeguard duty would work only if India had a larger existing domestic manufacturing base.

The government should reconsider the safeguard duty. Many foreign companies desiring to set up joint ventures in India provide only a lukewarm response because the given order in its current form presents inadequate safeguards.

There are incremental developments in technology at regular periods, which need capital, and the country should discover a way to handle these factors.

To make use of the vast estimated renewable potential in India, the R&D capability should be upgraded to solve critical problems in the clean energy sector.

A comprehensive policy for manufacturing should be established. This would support capital cost reduction and be marketed on a global scale.

The country should initiate an industry-academia partnership, which might promote innovative R&D and support leading-edge clean power solutions to protect the globe for future generations.

Encourage the transfer of ideas between industry, academia, and policymakers from around the world to develop accelerated adoption of renewable power.

Awareness about renewables

Social recognition of renewable energy is still not very promising in urban India. Awareness is the crucial factor for the uniform and broad use of renewable energy. Information about renewable technology and their environmental benefits should reach society.

The government should regularly organize awareness programs throughout the country, especially in villages and remote locations such as the islands.

The government should open more educational/research organizations, which will help in spreading knowledge of renewable technology in society.

People should regularly be trained with regard to new techniques that would be beneficial for the community.

Sufficient agencies should be available to sell renewable products and serve for technical support during installation and maintenance.

Development of the capabilities of unskilled and semiskilled workers and policy interventions are required related to employment opportunities.

An increase in the number of qualified/trained personnel might immediately support the process of installations of renewables.

Renewable energy employers prefer to train employees they recruit because they understand that education institutes fail to give the needed and appropriate skills. The training institutes should rectify this issue. Severe trained human resources shortages should be eliminated.

Upgrading the ability of the existing workforce and training of new professionals is essential to achieve the renewable goal.

Hybrid utilization of renewables

The country should focus on hybrid power projects for an effective use of transmission infrastructure and land.

India should consider battery storage in hybrid projects, which support optimizing the production and the power at competitive prices as well as a decrease of variability.

Formulate mandatory standards and regulations for hybrid systems, which are lagging in the newly announced policies (wind-solar hybrid policy on 14.05.2018).

The hybridization of two or more renewable systems along with the conventional power source battery storage can increase the performance of renewable technologies.

Issues related to sizing and storage capacity should be considered because they are key to the economic viability of the system.

Fiscal and financial incentives available for hybrid projects should be increased.

The renewable sector suffers notable obstacles. Some of them are inherent in every renewable technology; others are the outcome of a skewed regulative structure and marketplace. The absence of comprehensive policies and regulation frameworks prevent the adoption of renewable technologies. The renewable energy market requires explicit policies and legal procedures to enhance the attention of investors. There is a delay in the authorization of private sector projects because of a lack of clear policies. The country should take measures to attract private investors. Inadequate technology and the absence of infrastructure required to establish renewable technologies should be overcome by R&D. The government should allow more funds to support research and innovation activities in this sector. There are insufficiently competent personnel to train, demonstrate, maintain, and operate renewable energy structures and therefore, the institutions should be proactive in preparing the workforce. Imported equipment is costly compared to that of locally manufactured; therefore, generation of renewable energy becomes expensive and even unaffordable. Hence, to decrease the cost of renewable products, the country should become involve in the manufacturing of renewable products. Another significant infrastructural obstacle to the development of renewable energy technologies is unreliable connectivity to the grid. As a consequence, many investors lose their faith in renewable energy technologies and are not ready to invest in them for fear of failing. India should work on transmission and evacuation plans.

Inadequate servicing and maintenance of facilities and low reliability in technology decreases customer trust in some renewable energy technologies and hence prevent their selection. Adequate skills to repair/service the spare parts/equipment are required to avoid equipment failures that halt the supply of energy. Awareness of renewable energy among communities should be fostered, and a significant focus on their socio-cultural practices should be considered. Governments should support investments in the expansion of renewable energy to speed up the commercialization of such technologies. The Indian government should declare a well-established fiscal assistance plan, such as the provision of credit, deduction on loans, and tariffs. The government should improve regulations making obligations under power purchase agreements (PPAs) statutorily binding to guarantee that all power DISCOMs have PPAs to cover a hundred percent of their RPO obligation. To accomplish a reliable system, it is strongly suggested that renewables must be used in a hybrid configuration of two or more resources along with conventional source and storage devices. Regulatory authorities should formulate the necessary standards and regulations for hybrid systems. Making investments economically possible with effective policies and tax incentives will result in social benefits above and beyond the economic advantages.

Availability of data and materials

Not applicable.

Abbreviations

Accelerated depreciation

Billion units

Central Electricity Authority of India

Central electricity regulatory commission

Central financial assistance

Expression of interest

Foreign direct investment

Feed-in-tariff

Ministry of new and renewable energy

Research and development

Renewable purchase obligations

State electricity regulatory

Small hydropower

Terawatt hours

Waste to energy

Chr.Von Zabeltitz (1994) Effective use of renewable energies for greenhouse heating. Renewable Energy 5:479-485.

Article   Google Scholar  

Charles Rajesh Kumar. J, Vinod Kumar.D, M.A. Majid (2019) Wind energy programme in India: emerging energy alternatives for sustainable growth. Energy & Environment 30(7):1135-1189.

National electricity plan (2016), Volume 1, Generation, Central Electricity Authority (CEA),Ministry of Power, GOI . Available at http://www.cea.nic.in/reports/committee/nep/nep_dec.pdf .Accessed 31 Jan 2018.

Canadian environmental sustainability indicators (2017), Global greenhouse gas emissions. Available at http://www.ec.gc.ca/indicateurs-indicators/54C061B5-44F7-4A93-A3EC-5F8B253A7235/GlobalGHGEmissions_EN.pdf . Accessed 27 June.2017.

Pappas D (2017) Energy and Industrial Growth in India: The Next Emissions Superpower? Energy procedia 105:3656–3662

Agreement P (2015) Available at ttps://unfccc.int/sites/default/files/english_paris_agreement.pdf.Accessed 20. Aug 2017

Aggarwal P (2017) 2 °C target, India’s climate action plan and urban transport sector. Travel Behavior and Society 6:110–116

World Energy Scenarios Composing energy futures to 2050 (2013), World energy Council. https://www.worldenergy.org/wp-content/uploads/2013/09/World-Energy-Scenarios_Composing-energy-futures-to-2050_Full-report.pdf .Accessed 01 Jan 2017.

Blondeel M, Van de Graaf T (2018) Toward a global coal mining moratorium? A comparative analysis of coal mining policies in the USA, China, India and Australia. Climatic Change 150(1-2):89–101

Kumar S (2016) CO2 emission reduction potential assessment using renewable energy in India. Energy 97:273–282

Charles Rajesh Kumar. J, Mary Arunsi. B, Jenova. R, M.A.Majid (2019) Sustainable waste management through waste to energy technologies in India—opportunities and environmental impacts .International journal of renewable energy research 9(1): 309-342.

National Institution for Transforming India(2015), Government of India, Report of the Expert group on 175 GW RE by 2022,Available at http://niti.gov.in/writereaddata/files/writereaddata/files/document_publication/report-175-GW-RE.pdf.Accessed 31 Dec 2016.

Sholapurkar RB, Mahajan YS (2015) Review of wind energy development and policy in India. Energy Technology & Policy 2:122–132

India Energy scenarios 2047 (2015), ISGF for planning commission. Available at http://www.indiaenvironmentportal.org.in/files/file/ISGF_IES%202047%20Documentation.pdf.Accessed 01 Jan 2017].

Harrison T, Kostka G (2014) Balancing priorities, aligning interests: developing mitigation capacity in China and India. Comparative Political Studies 47:450-480

Akash KumarShukl (2017) Renewable energy resources in South Asian countries: challenges, policy and recommendations.Resource-Efficient Technologies 3: 342-346.

Schmid G (2012) The development of renewable energy power in India: which policies have been effective? Energy Policy 45:317–326

Vikas Khare, SavitaNema,PrashantBareda (2013) Status of solar wind renewable energy in India, Renewable and Sustainable Energy Reviews. 27: 1-10

Singh R (2015) India's renewable energy targets: How to overcome a $200 billion funding gap. Renewable Energy Focus. 16(4):60–61

T.Blenkinsopp, S.RColes ,K.Kirwan (2013) Renewable energy for rural communities in Maharashtra, India, Energy Policy .60:192-199

Kandp TC, Garg HP (1998) Renewable energy education for technicians/mechanics. Renewable Energy . (14(1–4):393–400

Subhes C,Bhattacharyya , Shaping a sustainable energy future for India: management challenges, Energy Policy .38(8):4173-4185

Swaran Singh, BoparaiK.C.Secretary, India and renewable energy: a future challenge, Renewable Energy . 15(1–4 ): 16-21.

Rehman S, Hussain Z (2017) Renewable energy governance in India: challenges and prospects for achieving the 2022 energy goals Journal of Resources, Energy and Development . 14(1):13–22

K Kaygusuz, S. BilgenEnergy related environmental policies in Turkey Energy Sources Part B, 3 (2008), pp. 396-410.

Y Chang, J Lee, H. Yoon Alternative projection of the world energy consumption-in comparison with the 2010 international energy outlook Energy Policy, 50 (2010), pp. 154-160.

Wang W (2014) M Zhang. P. Li Exploring temporal and spatial evolution of global energy production and consumption Renew Sustain Energy Rev 30:943–949

Google Scholar  

BP Energy Outlook country and regional insights-India (2018) https://www.bp.com/content/dam/bp/en/corporate/pdf/energy-economics/energy-outlook/bp-energy-outlook-2018-country-insight-india.pdf.Accessed 30 Jun 2018.

EIA Energy outlook 2019 with projections to 2050 (2019), Available at https://www.eia.gov/outlooks/aeo/pdf/aeo2019.pdf . .

International energy outlook 2018 (IEO2018), EIA Energy outlook 2018(2018), Available at https://www.eia.gov/pressroom/presentations/capuano_07242018.pdf .Accessed 30.07.2018.

World meters (2019).Available at http://www.worldometers.info/world-population/india-population.Accessed 24 Jan 2019.

Inaki Arto (2016) The energy requirements of a developed world", Energy for Sustainable Development.33: 1-13.

Power sector at a glance all India (2019), Ministry of Power, Government of India. Available at https://powermin.nic.in/en/content/power-sector-glance-all-india.Accessed 31 Oct 2018

VikasKhare (2013) Status of solar wind renewable energy in India Renewable and Sustainable Energy Reviews.27:1-10.

REMAP, renewable energy prospects for India (2017), The International renewable energy agency (IRENA). Available at https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2017/May/IRENA_REmap_India_paper_2017.pdf.Accessed 23 Aug 2017.

Draft national electricity plan, Volume 1, Generation, Central Electricity Authority (CEA), Ministry of Power, GOI Report 2016. http://www.cea.nic.in/reports/committee/nep/nep_dec.pdf .Accessed 26.06.2017.

BSK Naidu,Indian scenario of renewable energy for sustainable development, Energy policy ,Vol 24 ,N0 6,pp 575-581,1996.

Ashwani Kumar,Kapil Kumar, Naresh Kaushik, Satyawati Sharma,Saroj Mishra, Renewable energy in India:Current status and future potentials,Journal of renewable and sustainable energy reviews,14(2010),2434-2442

Bandyopadhyay S (2017) Renewable targets for India. Clean Technologies and Environmental Policy 19(2):293–294

Ministry of New and Renewable Energy (2017),Annual report 2016-17,Available at http://mnre.gov.in/file-manager/annual-report/2016-2017/EN/pdf/1.pdf.Accessed 31 April 2017.

Nimish Kumar, Nitai Pal, The existence of barriers and proposed recommendations for the development of renewable energy in Indian perspective, Environment, Development and Sustainability ,pp1-19.2018.

Yearend review (2017), Government of India, Ministry of New and Renewable Energy. Available at https://mnre.gov.in/file-manager/akshay-urja/april-2018/Images/44-45.pdf.Accessed 27 Jan 2018.

Central Electricity authority of India (2018), Government of India. Available at http://www.cea.nic.in/reports/monthly/executivesummary/2018/exe_summary-12.pdf.Accessed 31 Jan 2018.

The growth of Electricity sector in India from 1947-2017(2017), Central Electricity Authority of India, Government of India. Available at http://www.cea.nic.in/reports/others/planning/pdm/growth_2017.pdf . Accessed 31 June 2017.

Renewable energy, Indian brand equity foundation (2018). Available at https://www.ibef.org/download/renewable-energy-dec-2018.pdf.Accessed 05 Jan 2019.

Prayas energy group (2018) .Available at http://www.prayaspune.org/peg/re-capacity.html.Accessed 31 June 2018.

Subhojit Dawn, Prashant Kumar Tiwari, Arup Kumar Goswami, Ankit Kumar, Singh Rajesh Panda (2019) Wind power: Existing status, achievements and government's initiative towards renewable power dominating India Energy Strategy Reviews.23:178-199.

Generation monthly reports, Central electricity Authority of India (2018) Available at http://www.cea.nic.in/reports/monthly/renewable/2018/overview-11.pdf . .

Ian Partridge (2013) Renewable electricity generation in India—a learning rate analysis Energy Policy.60:906:915.

J. Jeslin Drusila Nesamalar, P. Venkatesh, S. Charles Raja (2017) The drive of renewable energy in Tamilnadu: Status, barriers and future prospect Renewable and Sustainable Energy Reviews.73:115-124.

A Report on Energy Efficiency and Energy Mix in the Indian Energy System (2030) Using India Energy Security Scenarios 2047 (2017), Available at http://niti.gov.in/writereaddata/files/document_publication/Energy_Efficiency.pdf.Accessed 06 April 2017.

Initiatives and achievements, MNRE (2018).Available at https://mnre.gov.in/sites/default/files/uploads/MNRE-4-Year-Achievement-Booklet.pdf . .

Acheievements of ministry of new and renewable energy during 2018. (2019).Available at https://pibindia.wordpress.com/2018/12/11/achievements-of-ministry-of-new-renewable-energy-during-2018/.Accessed 09 Jan 2019.

Standing committee on energy (2017-18), sixteenth lok sabha, MNRE, demands for grands (2018-19), 39th (2018). Available at http://164.100.47.193/lsscommittee/Energy/16_Energy_39.pdf . .

Mukesh KumarMishra, NilayKhare,Alka BaniAgrawa (2015) Small hydro power in India: Current status and future perspectives Renewable and Sustainable Energy Reviews .51:101-115.

Tara Chandra Kandpal, Bharati Joshi , Chandra ShekharSinha (1991) Economics of family sized biogas plants in India Energy Conversion and Management .32:101-113.

Sravanthi Choragudi (2013) Off-grid solar lighting systems: a way align India's sustainable and inclusive development goals Renewable and Sustainable Energy Reviews .28:890-899.

Abhigyan Singh,Alex T.Stratin,N.A.Romero Herrera,Debotosh Mahato,David V.Keyson,Hylke W.van Dijk (2018) Exploring peer-to-peer returns in off-grid renewable energy systems in rural India: an anthropological perspective on local energy sharing and trading Energy Research & Social Science .46:194-213.

Draft Technology Development and Innovation Policy (TDIP) for New &Renewable Energy, MNRE (2017). Available at https://mnre.gov.in/file-manager/UserFiles/Draft-TDIP_RE.pdf .Accessed 31 Jan 2018.

Demands for grants (2018-19), MNRE, Standing committee on energy, 16 th lok sabha, 39th Report (2018). Available at http://164.100.47.193/lsscommittee/Energy/16_Energy_39.pdf.Accessed 31 Oct 2018.

Mohit Goyal (2010) Repowering—next big thing in India Renewable and Sustainable Energy Reviews.1 4(5):1400-1409.

B.N.Prashanth,R.Pramod,G.B. VeereshKumar (2018) Design and development of hybrid wind and solar energy system for power generation. 5(5):11415-11422.

Swaminathan mani,Tarun Dhingra (2013) Policies to accelerate the growth of offshore wind energy sector in India Renewable and Sustainable Energy Reviews . 24 : 473-482

Draft amendments to tariff policy, MNRE (2018). Available at https://powermin.nic.in/sites/default/files/webform/notices/Proposed_amendments_in_Tariff_Policy_0.pdf . .

T.C.Kandpal, H.P.Garg (1998) Renewable energy education for technicians/mechanics Renewable Energy . Volume 14(1–4 ): 393-400.

T.Blenkinsopp, S.RColes K.Kirwan (2013) Renewable energy for rural communities in Maharashtra, India Energy policy.60:192-199.

Renewable Energy Country Attractiveness Index (RECAI) (2018), Available at https://www.ey.com/Publication/vwLUAssets/ey-recai-issue-52-index-scores/$File/ey-recai-issue-52-index-scores.pdf . .

Renewables 2018, global status report, Renewable energy policy network for the 21 st century (REN21) (2018). Available at http://www.ren21.net/wp-content/uploads/2018/06/17-8652_GSR2018_FullReport_web_final_.pdf . .

Clean Energy Investment trends, 2Q 2018, BloombergNEF (2018) .Available at https://data.bloomberglp.com/bnef/sites/14/2018/07/BNEF-Clean-Energy-Investment-Trends-1H-2018.pdf .Accessed 12 Dec 2018.

Rolf Wüstenhagen, Emanuela Menichetti (2012) Strategic choices for renewable energy investment: conceptual framework and opportunities for further research Energy Policy.40:1-10.

Longayanxia dam solar park, earth observatory (NASA) (2017). Available at https://earthobservatory.nasa.gov/images/89668/longyangxia-dam-solar-park . .

Dawn, Subhojit Dawn, Prashant Kumar Tiwari,Arup Kumar Goswami, Manash Kumar Mishra. (2016).Recent developments of solar energy in India: perspectives, strategies and future goals. Renewable and Sustainable Energy Reviews.62:215-235.

Press information Bureau, MNRE FDI in renewable energy sector (2018). Available at http://pib.nic.in/newsite/PrintRelease.aspx?relid=186849.Accessed 01 Feb 2019.

Renewable energy, India brand equity foundation, Report August 2018. https://www.ibef.org/download/Renewable-Energy-Report-August-2018.pdf . Accessed 31.12.2018.

Global trends in renewable energy investment 2018, Bloomberg new energy (2018).Available at http://www.iberglobal.com/files/2018/renewable_trends.pdf . .

Deep dive –Allianz climate and energy monitor (2017).Available at https://newclimate.org/wp-content/uploads/2017/04/allianz-climate-and-energy-monitor-deep-dive-2017.pdf . .

Allianz climate and energy monitor (2018), Available at https://www.allianz.com/content/dam/onemarketing/azcom/Allianz_com/sustainability/documents/Allianz_Climate_and_Energy_Monitor_2018.pdf . .

Renewable energy and Jobs –Annual review 2018 (2018). Available at https://irena.org/-/media/Files/IRENA/Agency/Publication/2018/May/IRENA_RE_Jobs_Annual_Review_2018.pdf .Accessed Jan 2019.

A. Bergmann, N. Hanley, R. Wright Valuing the attributes of renewable energy investments .Energy Policy, 34 (9) (2006), pp. 1004-1014

József Benedek, Tihamér-Tibor Sebestyén, BlankaBartók (2018) Evaluation of renewable energy sources in peripheral areas and renewable energy-based rural development Renewable and Sustainable Energy Reviews . 90 :516-535.

Download references

Acknowledgments

The authors gratefully acknowledge the support provided by the Research Consultancy Institute (RCI) and the department of Electrical and Computer Engineering of Effat University, Saudi Arabia.

Author information

Authors and affiliations.

Department of Electrical and Computer Engineering, College of Engineering, Effat University, Box 34689, Jeddah, PO, 21478, Saudi Arabia

Charles Rajesh Kumar. J & M. A. Majid

You can also search for this author in PubMed   Google Scholar

Contributions

CRK conceptualized the research, undertook fieldwork, analyzed the data, and wrote the manuscript. MAM conceptualized the research, wrote the manuscript, and supervised the research. Both authors have read and approved the final manuscript.

Corresponding author

Correspondence to Charles Rajesh Kumar. J .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

The authors declared that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Cite this article.

Kumar. J, C.R., Majid, M.A. Renewable energy for sustainable development in India: current status, future prospects, challenges, employment, and investment opportunities. Energ Sustain Soc 10 , 2 (2020). https://doi.org/10.1186/s13705-019-0232-1

Download citation

Received : 15 September 2018

Accepted : 27 November 2019

Published : 07 January 2020

DOI : https://doi.org/10.1186/s13705-019-0232-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Sustainable
• Renewable energy
• Achievements
• Initiatives
• Recommendations
• Policymakers

Energy, Sustainability and Society

ISSN: 2192-0567

• General enquiries: [email protected]

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Perspective
• Published: 18 October 2022

Machine learning for a sustainable energy future

• Zhenpeng Yao   ORCID: orcid.org/0000-0001-8286-8257 1 , 2 , 3 , 4   na1 ,
• Yanwei Lum   ORCID: orcid.org/0000-0001-7261-2098 5 , 6   na1 ,
• Andrew Johnston 6   na1 ,
• Luis Martin Mejia-Mendoza 2 ,
• Xin Zhou 7 ,
• Yonggang Wen 7 ,
• Alán Aspuru-Guzik   ORCID: orcid.org/0000-0002-8277-4434 2 , 8 ,
• Edward H. Sargent   ORCID: orcid.org/0000-0003-0396-6495 6 &
• Zhi Wei Seh   ORCID: orcid.org/0000-0003-0953-567X 5  

Nature Reviews Materials volume  8 ,  pages 202–215 ( 2023 ) Cite this article

34k Accesses

117 Citations

24 Altmetric

Metrics details

• Computer science

Electrocatalysis

• Energy grids and networks
• Solar cells

Transitioning from fossil fuels to renewable energy sources is a critical global challenge; it demands advances — at the materials, devices and systems levels — for the efficient harvesting, storage, conversion and management of renewable energy. Energy researchers have begun to incorporate machine learning (ML) techniques to accelerate these advances. In this Perspective, we highlight recent advances in ML-driven energy research, outline current and future challenges, and describe what is required to make the best use of ML techniques. We introduce a set of key performance indicators with which to compare the benefits of different ML-accelerated workflows for energy research. We discuss and evaluate the latest advances in applying ML to the development of energy harvesting (photovoltaics), storage (batteries), conversion (electrocatalysis) and management (smart grids). Finally, we offer an overview of potential research areas in the energy field that stand to benefit further from the application of ML.

Similar content being viewed by others

A review of the recent progress in battery informatics

Predicting the state of charge and health of batteries using data-driven machine learning

Fundamental theory on multiple energy resources and related case studies

Introduction.

The combustion of fossil fuels, used to fulfill approximately 80% of the world’s energy needs, is the largest single source of rising greenhouse gas emissions and global temperature 1 . The increased use of renewable sources of energy, notably solar and wind power, is an economically viable path towards meeting the climate goals of the Paris Agreement 2 . However, the rate at which renewable energy has grown has been outpaced by ever-growing energy demand, and as a result the fraction of total energy produced by renewable sources has remained constant since 2000 (ref. 3 ). It is thus essential to accelerate the transition towards sustainable sources of energy 4 . Achieving this transition requires energy technologies, infrastructure and policies that enable and promote the harvest, storage, conversion and management of renewable energy.

In sustainable energy research, suitable material candidates (such as photovoltaic materials) must first be chosen from the combinatorial space of possible materials, then synthesized at a high enough yield and quality for use in devices (such as solar panels). The time frame of a representative materials discovery process is 15–20 years 5 , 6 , leaving considerable room for improvement. Furthermore, the devices have to be optimized for robustness and reproducibility to be incorporated into energy systems (such as in solar farms) 7 , where management of energy usage and generation patterns is needed to further guarantee commercial success.

Here we explore the extent to which machine learning (ML) techniques can help to address many of these challenges 8 , 9 , 10 . ML models can be used to predict specific properties of new materials without the need for costly characterization; they can generate new material structures with desired properties; they can understand patterns in renewable energy usage and generation; and they can help to inform energy policy by optimizing energy management at both device and grid levels.

In this Perspective, we introduce Acc(X)eleration Performance Indicators (XPIs), which can be used to measure the effectiveness of platforms developed for accelerated energy materials discovery. Next, we discuss closed-loop ML frameworks and evaluate the latest advances in applying ML to the development of energy harvesting, storage and conversion technologies, as well as the integration of ML into a smart power grid. Finally, we offer an overview of energy research areas that stand to benefit further from ML.

Performance indicators

Because many reports discuss ML-accelerated approaches for materials discovery and energy systems management, we posit that there should be a consistent baseline from which these reports can be compared. For energy systems management, performance indicators at the device, plant and grid levels have been reported 11 , 12 , yet there are no equivalent counterparts for accelerated materials discovery.

The primary goal in materials discovery is to develop efficient materials that are ready for commercialization. The commercialization of a new material requires intensive research efforts that can span up to two decades: the goal of every accelerated approach should be to accomplish commercialization an order-of-magnitude faster. The materials science field can benefit from studying the case of vaccine development. Historically, new vaccines take 10 years from conception to market 13 . However, after the start of the COVID-19 pandemic, several companies were able to develop and begin releasing vaccines in less than a year. This achievement was in part due to an unprecedented global research intensity, but also to a shift in the technology: after a technological breakthrough in 2008, the cost of sequencing DNA began decreasing exponentially 14 , 15 , enabling researchers to screen orders-of-magnitude more vaccines than was previously possible.

ML for energy technologies has much in common with ML for other fields like biomedicine, sharing the same methodology and principles. However, in practice, ML models for different technologies are exposed to additional unique requirements. For example, ML models for medical applications usually have complex structures that take into account regulatory oversight and ensure the safe development, use and monitoring of systems, which usually does not happen in the energy field 16 . Moreover, data availability varies substantially from field to field; biomedical researchers can work with a relatively large amount of data that energy researchers usually lack. This limited data accessibility can constrain the usage of sophisticated ML models (such as deep learning models) in the energy field. However, adaptation has been quick in all energy subfields, with a rapidly increased number of groups recognizing the importance of statistical methods and starting to use them for various problems. We posit that the use of high-throughput experimentation and ML in materials discovery workflows can result in breakthroughs in accelerating development, but the field first needs a set of metrics with which ML models can be evaluated and compared.

Accelerated materials discovery methods should be judged based on the time it takes for a new material to be commercialized. We recognize that this is not a useful metric for new platforms, nor is it one that can be used to decide quickly which platform is best suited for a particular scenario. We therefore propose here XPIs that new materials discovery platforms should report.

Acceleration factor of new materials, XPI-1

This XPI is evaluated by dividing the number of new materials that are synthesized and characterized per unit time with the accelerated platform by the number of materials that are synthesized and characterized with traditional methods. For example, an acceleration factor of ten means that for a given time period, the accelerated platform can evaluate ten times more materials than a traditional platform. For materials with multiple target properties, researchers should report the rate-limiting acceleration factor.

Number of new materials with threshold performance, XPI-2

This XPI tracks the number of new materials discovered with an accelerated platform that have a performance greater than the baseline value. The selection of this baseline value is critical: it should be something that fairly captures the standard to which new materials need to be compared. As an example, an accelerated platform that seeks to discover new perovskite solar cell materials should track the number of devices made with new materials that have a better performance than the best existing solar cell 17 .

Performance of best material over time, XPI-3

This XPI tracks the absolute performance — whether it is Faradaic efficiency, power conversion efficiency or other — of the best material as a function of time. For the accelerated framework, the evolution of the performance should increase faster than the performance obtained by traditional methods 18 .

Repeatability and reproducibility of new materials, XPI-4

This XPI seeks to ensure that the new materials discovered are consistent and repeatable: this is a key consideration to screen out materials that would fail at the commercialization stage. The performance of a new material should not vary by more than x % of its mean value (where x is the standard error): if it does, this material should not be included in either XPI-2 (number of new materials with threshold performance) or XPI-3 (performance of best material over time).

Human cost of the accelerated platform, XPI-5

This XPI reports the total costs of the accelerated platform. This should include the total number of researcher hours needed to design and order the components for the accelerated system, develop the programming and robotic infrastructure, develop and maintain databases used in the system and maintain and run the accelerated platform. This metric would provide researchers with a realistic estimate of the resources required to adapt an accelerated platform for their own research.

Use of the XPIs

Each of these XPIs can be measured for computational, experimental or integrated accelerated systems. Consistently reporting each of these XPIs as new accelerated platforms are developed will allow researchers to evaluate the growth of these platforms and will provide a consistent metric by which different platforms can be compared. As a demonstration, we applied the XPIs to evaluate the acceleration performance of several typical platforms: Edisonian-like trial-test, robotic photocatalysis development 19 and design of a DNA-encoded-library-based kinase inhibitor 20 (Table  1 ). To obtain a comprehensive performance estimate, we define one overall acceleration score S adhering to the following rules. The dependent acceleration factors (XPI-1 and XPI-2), which function in a synergetic way, are added together to reflect their contribution as a whole. The independent acceleration factors (XPI-3, XPI-4 and XPI-5), which may function in a reduplicated way, are multiplied together to value their contributions respectively. As a result, the overall acceleration score can be calculated as S  = (XPI-1 + XPI-2) × XPI-3 × XPI-4 ÷ XPI-5. As the reference, the Edisonian-like approach has a calculated overall XPIs score of around 1, whereas the most advanced method, the DNA-encoded-library-based drug design, exhibits an overall XPIs score of 10 7 . For the sustainability field, the robotic photocatalysis platform has an overall XPIs score of 10 5 .

For energy systems, the most frequently reported XPI is the acceleration factor, in part because it is deterministic, but also because it is easy to calculate at the end of the development of a workflow. In most cases, we expect that authors report the acceleration factor only after completing the development of the platform. Reporting the other suggested XPIs will provide researchers with a better sense of both the time and human resources required to develop the platform until it is ready for publication. Moving forward, we hope that other researchers adopt the XPIs — or other similar metrics — to allow for fair and consistent comparison between the different methods and algorithms that are used to accelerate materials discovery.

Closed-loop ML for materials discovery

The traditional approach to materials discovery is often Edisonian-like, relying on trial and error to develop materials with specific properties. First, a target application is identified, and a starting pool of possible candidates is selected (Fig.  1a ). The materials are then synthesized and incorporated into a device or system to measure their properties. These results are then used to establish empirical structure–property relationships, which guide the next round of synthesis and testing. This slow process goes through as many iterations as required and each cycle can take several years to complete.

a | Traditional Edisonian-like approach, which involves experimental trial and error. b | High-throughput screening approach involving a combination of theory and experiment. c | Machine learning (ML)-driven approach whereby theoretical and experimental results are used to train a ML model for predicting structure–property relationships. d | ML-driven approach for property-directed and automatic exploration of the chemical space using optimization ML (such as genetic algorithms or generative models) that solve the ‘inverse’ design problem.

A computation-driven, high-throughput screening strategy (Fig.  1b ) offers a faster turnaround. To explore the overall vast chemical space (~10 60 possibilities), human intuition and expertise can be used to create a library with a substantial number of materials of interest (~10 4 ). Theoretical calculations are carried out on these candidates and the top performers (~10 2 candidates) are then experimentally verified. With luck, the material with the desired functionality is ‘discovered’. Otherwise, this process is repeated in another region of the chemical space. This approach can still be very time-consuming and computationally expensive and can only sample a small region of the chemical space.

ML can substantially increase the chemical space sampled, without costing extra time and effort. ML is data-driven, screening datasets to detect patterns, which are the physical laws that govern the system. In this case, these laws correspond to materials structure–property relationships. This workflow involves high-throughput virtual screening (Fig.  1c ) and begins by selecting a larger region (~10 6 ) of the chemical space of possibilities using human intuition and expertise. Theoretical calculations are carried out on a representative subset (~10 4 candidates) and the results are used for training a discriminative ML model. The model can then be used to make predictions on the other candidates in the overall selected chemical space 9 . The top ~10 2 candidates are experimentally verified, and the results are used to improve the predictive capabilities of the model in an iterative loop. If the desired material is not ‘discovered’, the process is repeated on another region of the chemical space.

An improvement on the previous approaches is a framework that requires limited human intuition or expertise to direct the chemical space search: the automated virtual screening approach (Fig.  1d ). To begin with, a region of the chemical space is picked at random to initiate the process. Thereafter, this process is similar to the previous approach, except that the computational and experimental data is also used to train a generative learning model. This generative model solves the ‘inverse’ problem: given a required property, the goal is to predict an ideal structure and composition in the chemical space. This enables a directed, automated search of the chemical space, towards the goal of ‘discovering’ the ideal material 8 .

ML for energy

ML has so far been used to accelerate the development of materials and devices for energy harvesting (photovoltaics), storage (batteries) and conversion (electrocatalysis), as well as to optimize power grids. Besides all the examples discussed here, we summarize the essential concepts in ML (Box  1 ), the grand challenges in sustainable materials research (Box  2 ) and the details of key studies (Table  2 ).

Box 1 Essential concepts in ML

With the availability of large datasets 122 , 125 and increased computing power, various machine learning (ML) algorithms have been developed to solve diverse problems in energy. Below, we provide a brief overview of the types of problem that ML can solve in energy technology, and we then summarize the status of ML-driven energy research. More detailed information about the nuts and bolts of ML techniques can be found in previous reviews 173 , 174 , 175 .

Property prediction

Supervised learning models are predictive (or discriminative) models that are given a datapoint x , and seek to predict a property y (for example, the bandgap 27 ) after being trained on a labelled dataset. The property y can be either continuous or discrete. These models have been used to aid or even replace physical simulations or measurements under certain circumstances 176 , 177 .

Generative materials design

Unsupervised learning models are generative models that can generate or output new examples x ′ (such as new molecules 104 ) after being trained on an unlabelled dataset. This generation of new examples can be further enhanced with additional information (physical properties) to condition or bias the generative process, allowing the models to generate examples with improved properties and leading to the property-to-structure approach called inverse design 52 , 178 .

Self-driving laboratories

Self-driving or autonomous laboratories 19 use ML models to plan and perform experiments, including the automation of retrosynthesis analysis (such as in reinforcement-learning-aided synthesis planning 124 , 179 ), prediction of reaction products (such as in convolutional neural networks (CNNs) for reaction prediction 137 , 138 ) and reaction condition optimization (such as in robotic workflows optimized by active learning 19 , 160 , 180 , 181 , 182 , 183 ). Self-driving laboratories, which use active learning for iterating through rounds of synthesis and measurements, are a key component in the closed-loop inverse design 52 .

Aiding characterization

ML models have been used to aid the quantitative or qualitative analysis of experimental observations and measurements, including assisting in the determination of crystal structure from transmission electron microscopy images 184 , identifying coordination environment 81 and structural transition 83 from X-ray absorption spectroscopy and inferring crystal symmetry from electron diffraction 176 .

Accelerating theoretical computations

ML models can enable otherwise intractable simulations by reducing the computational cost (processor core amount and time) for systems with increased length and timescales 69 , 70 and providing potentials and functionals for complex interactions 68 .

Optimizing system management

ML models can aid the management of energy systems at the device or grid power level by predicting lifetimes (such as battery life 43 , 44 ), adapting to new loads (such as in long short-term memory for building load prediction 95 ) and optimizing performance (such as in reinforcement learning for smart grid control 94 ).

Box 2 Grand challenges in energy materials research

Photovoltaics.

Discover non-toxic (Pd- and Cd-free) materials with good optoelectronic properties

Identify and minimize materials defects in light-absorbing materials

Design effective recombination-layer materials for tandem solar cells

Develop materials design strategies for long-term operational stability 125

Develop (hole/electron) transport materials with high carrier mobility 125

Optimize cell structure for maximum light absorption and minimum use of active materials

Tune materials bandgaps for optimal solar-harvesting performance under complex operation conditions 21 , 22

Develop Earth-abundant cathode materials (Co-free) with high reversibility and charge capacity 4

Design electrolytes with wider electrochemical windows and high conductivity 4

Identify electrolyte systems to boost battery performance and lifetime 4

Discover new molecules for redox flow batteries with suitable voltage 4

Understand correlation between defect growth in battery materials and overall degradation process of battery components

Tune operando (dis)charging protocol for minimized capacity loss, (dis)charging rate and optimal battery life under diversified conditions 7 , 53

Design materials with optimal adsorption energy for maximized catalytic activity 60 , 61

Identify and study active sites on catalytic materials 58

Engineer catalytic materials for extended durability 58 , 60 , 61

Identify a fuller set of materials descriptors that relate to catalytic activity 60 , 61

Engineer multiple catalytic functionalities into the same material 60 , 61

Design multiscale electrode structures for optimized catalytic activity

Correlate atomistic contamination and growth of catalyst particles with electrode degradation process

Tune operando (dis)charging protocol for minimized capacity loss and optimal cell life

ML is accelerating the discovery of new optoelectronic materials and devices for photovoltaics, but major challenges are still associated with each step.

Photovoltaics materials discovery

One materials class for which ML has proved particularly effective is perovskites, because these materials have a vast chemical space from which the constituents may be chosen. Early representations of perovskite materials for ML were atomic-feature representations, in which each structure is encoded as a fixed-length vector comprised of an average of certain atomic properties of the atoms in the crystal structure 21 , 22 . A similar technique was used to predict new lead-free perovskite materials with the proper bandgap for solar cells 23 (Fig.  2a ). These representations allowed for high accuracy but did not account for any spatial relation between atoms 24 , 25 . Materials systems can also be represented as images 26 or as graphs 27 , enabling the treatment of systems with diverse number of atoms. The latter representation is particularly compelling, as perovskites, particularly organic–inorganic perovskites, have crystal structures that incorporate a varying number of atoms, and the organic molecules can vary in size.

a | Energy harvesting 23 . b | Energy storage 38 . c | Energy conversion 76 . d | Energy management 93 . ICSD, Inorganic Crystal Structure Database; ML, machine learning.

Although bandgap prediction is an important first step, this parameter alone is not sufficient to indicate a useful optoelectronic material; other parameters, including electronic defect density and stability, are equally important. Defect energies are addressable with computational methods, but the calculation of defects in structures is extremely computationally expensive, which inhibits the generation of a dataset of defect energies from which an ML model can be trained. To expedite the high-throughput calculation of defect energies, a Python toolkit has been developed 28 that will be pivotal in building a database of defect energies in semiconductors. Researchers can then use ML to predict both the formation energy of defects and the energy levels of these defects. This knowledge will ensure that the materials selected from high-throughput screening will not only have the correct bandgap but will also either be defect-tolerant or defect-resistant, finding use in commercial optoelectronic devices.

Even without access to a large dataset of experimental results, ML can accelerate the discovery of optoelectronic materials. Using a self-driving laboratory approach, the number of experiments required to optimize an organic solar cell can be reduced from 500 to just 60 (ref. 29 ). This robotic synthesis method accelerates the learning rate of the ML models and drastically reduces the cost of the chemicals needed to run the optimization.

Solar device structure and fabrication

Photovoltaic devices require optimization of layers other than the active layer to maximize performance. One component is the top transparent conductive layer, which needs to have both high optical transparency and high electronic conductivity 30 , 31 . A genetic algorithm that optimized the topology of a light-trapping structure enabled a broadband absorption efficiency of 48.1%, which represents a more than threefold increase over the Yablonovitch limit, the 4 n 2 factor (where n is the refractive index of the material) theoretical limit for light trapping in photovoltaics 32 .

A universal standard irradiance spectrum is usually used by researchers to determine optimal bandgaps for solar cell operation 33 . However, actual solar irradiance fluctuates based on factors such as the position of the Sun, atmospheric phenomena and the season. ML can reduce yearly spectral sets into a few characteristic spectra 33 , allowing for the calculation of optimal bandgaps for real-world conditions.

To optimize device fabrication, a CNN was used to predict the current–voltage characteristics of as-cut Si wafers based on their photoluminescence images 34 . Additionally, an artificial neural network was used to predict the contact resistance of metallic front contacts for Si solar cells, which is critical for the manufacturing process 35 .

Although successful, these studies appear to be limited to optimizing structures and processes that are already well established. We suggest that, in future work, ML could be used to augment simulations, such as the multiphysics models for solar cells. Design of device architecture could begin from such simulation models, coupled with ML in an iterative process to quickly optimize design and reduce computational time and cost. In addition, optimal conditions for the scaling-up of device area and fabrication processes are likely to be very different from those for laboratory-scale demonstrations. However, determining these optimal conditions could be expensive in terms of materials cost and time, owing to the need to construct much larger devices. In this regard, ML, together with the strategic design of experiments, could greatly accelerate the optimization of process conditions (such as the annealing temperatures and solvent choice).

Electrochemical energy storage

Electrochemical energy storage is an essential component in applications such as electric vehicles, consumer electronics and stationary power stations. State-of-the-art electrochemical energy storage solutions have varying efficacy in different applications: for example, lithium-ion batteries exhibit excellent energy density and are widely used in electronics and electric vehicles, whereas redox flow batteries have drawn substantial attention for use in stationary power storage. ML approaches have been widely employed in the field of batteries, including for the discovery of new materials such as solid-state ion conductors 36 , 37 , 38 (Fig.  2b ) and redox active electrolytes for redox flow batteries 39 . ML has also aided battery management, for example, through state-of-charge determination 40 , state-of-health evaluation 41 , 42 and remaining-life prediction 43 , 44 .

Electrode and electrolyte materials design

Layered oxide materials, such as LiCoO 2 or LiNi x Mn y Co 1- x - y O 2 , have been used extensively as cathode materials for alkali metal-ion (Li/Na/K) batteries. However, developing new Li-ion battery materials with higher operating voltages, enhanced energy densities and longer lifetimes is of paramount interest. So far, universal design principles for new battery materials remain undefined, and hence different approaches have been explored. Data from the Materials Project have been used to model the electrode voltage profile diagrams for different materials in alkali metal-ion batteries (Na and K) 45 , leading to the proposition of 5,000 different electrode materials with appropriate moderate voltages. ML was also employed to screen 12,000 candidates for solid Li-ion batteries, resulting in the discovery of ten new Li-ion conducting materials 46 , 47 .

Flow batteries consist of active materials dissolved in electrolytes that flow into a cell with electrodes that facilitate redox reactions. Organic flow batteries are of particular interest. In flow batteries, the solubility of the active material in the electrolyte and the charge/discharge stability dictate performance. ML methods have explored the chemical space to find suitable electrolytes for organic redox flow batteries 48 , 49 . Furthermore, a multi-kernel-ridge regression method accelerated the discovery of active organic molecules using multiple feature training 48 . This method also helped in predicting the solubility dependence of anthraquinone molecules with different numbers and combinations of sulfonic and hydroxyl groups on pH. Future opportunities lie in the exploration of large combinatorial spaces for the inverse design of high-entropy electrodes 50 and high-voltage electrolytes 51 . To this end, deep generative models can assist the discovery of new materials based on the simplified molecular input line entry system (SMILES) representation of molecules 52 .

Battery device and stack management

A combination of mechanistic and semi-empirical models is currently used to estimate capacity and power loss in lithium-ion batteries. However, the models are applicable only to specific failure mechanisms or situations and cannot predict the lifetimes of batteries at the early stages of usage. By contrast, mechanism-agnostic models based on ML can accurately predict battery cycle life, even at an early stage of a battery’s life 43 . A combined early-prediction and Bayesian optimization model has been used to rapidly identify the optimal charging protocol with the longest cycle life 44 . ML can be used to accelerate the optimization of lithium-ion batteries for longer lifetimes 53 , but it remains to be seen whether these models can be generalized to different battery chemistries 54 .

ML methods can also predict important properties of battery storage facilities. A neural network was used to predict the charge/discharge profiles in two types of stationary battery systems, lithium iron phosphate and vanadium redox flow batteries 55 . Battery power management techniques must also consider the uncertainty and variability that arise from both the environment and the application. An iterative Q -learning ( reinforcement learning ) method was also designed for battery management and control in smart residential environments 56 . Given the residential load and the real-time electricity rate, the method is effective at optimizing battery charging/discharging/idle cycles. Discriminative neural network-based models can also optimize battery usage in electric vehicles 57 .

Although ML is able to predict the lifetime of batteries, the underlying degradation mechanisms are difficult to identify and correlate to the state of health and lifetime. To this end, incorporation of domain knowledge into a hybrid physics-based ML model can provide insight and reduce overfitting 53 . However, incorporating the physics of battery degradation processes into a hybrid model remains challenging; representation of electrode materials that encode both compositional and structural information is far from trivial. Validation of these models also requires the development of operando characterization techniques, such as liquid-phase transmission electron microscopy and ambient-pressure X-ray absorption spectroscopy (XAS), that reflect true operating conditions as closely as possible 54 . Ideally, these characterization techniques should be carried out in a high-throughput manner, using automated sample changers, for example, in order to generate large datasets for ML.

Electrocatalysts

Electrocatalysis enables the conversion of simple feedstocks (such as water, carbon dioxide and nitrogen) into valuable chemicals and/or fuels (such as hydrogen, hydrocarbons and ammonia), using renewable energy as an input 58 . The reverse reactions are also possible in a fuel cell, and hydrogen can be consumed to produce electricity 59 . Active and selective electrocatalysts must be developed to improve the efficiency of these reactions 60 , 61 . ML has been used to accelerate electrocatalyst development and device optimization.

Electrocatalyst materials discovery

The most common descriptor of catalytic activity is the adsorption energy of intermediates on a catalyst 61 , 62 . Although these adsorption energies can be calculated using density functional theory (DFT), catalysts possess multiple surface binding sites, each with different adsorption energies 63 . The number of possible sites increases dramatically if alloys are considered, and thus becomes intractable with conventional means 64 .

DFT calculations are critical for the search of electrocatalytic materials 65 and efforts have been made to accelerate the calculations and to reduce their computational cost by using surrogate ML models 66 , 67 , 68 , 69 . Complex reaction mechanisms involving hundreds of possible species and intermediates can also be simplified using ML, with a surrogate model predicting the most important reaction steps and deducing the most likely reaction pathways 70 . ML can also be used to screen for active sites across a random, disordered nanoparticle surface 71 , 72 . DFT calculations are performed on only a few representative sites, which are then used to train a neural network to predict the adsorption energies of all active sites.

Catalyst development can benefit from high-throughput systems for catalyst synthesis and performance evaluation 73 , 74 . An automatic ML-driven framework was developed to screen a large intermetallic chemical space for CO 2 reduction and H 2 evolution 75 . The model predicted the adsorption energy of new intermetallic systems and DFT was automatically performed on the most promising candidates to verify the predictions. This process went on iteratively in a closed feedback loop. 131 intermetallic surfaces across 54 alloys were ultimately identified as promising candidates for CO 2 reduction. Experimental validation 76 with Cu–Al catalysts yielded an unprecedented Faradaic efficiency of 80% towards ethylene at a high current density of 400 mA cm – 2 (Fig.  2c ).

Because of the large number of properties that electrocatalysts may possess (such as shape, size and composition), it is difficult to do data mining on the literature 77 . Electrocatalyst structures are complex and difficult to characterize completely; as a result, many properties may not be fully characterized by research groups in their publications. To avoid situations in which potentially promising compositions perform poorly as a result of non-ideal synthesis or testing conditions, other factors (such as current density, particle size and pH value) that affect the electrocatalyst performance must be kept consistent. New approaches such as carbothermal shock synthesis 78 , 79 may be a promising avenue, owing to its propensity to generate uniformly sized and shaped alloy nanoparticles, regardless of composition.

XAS is a powerful technique, especially for in situ measurements, and has been widely employed to gain crucial insight into the nature of active sites and changes in the electrocatalyst over time 80 . Because the data analysis relies heavily on human experience and expertise, there has been interest in developing ML tools for interpreting XAS data 81 . Improved random forest models can predict the Bader charge (a good approximation of the total electronic charge of an atom) and nearest-neighbour distances, crucial factors that influence the catalytic properties of the material 82 . The extended X-ray absorption fine structure (EXAFS) region of XAS spectra is known to contain information on bonding environments and coordination numbers. Neural networks can be used to automatically interpret EXAFS data 83 , permitting the identification of the structure of bimetallic nanoparticles using experimental XAS data, for example 84 . Raman and infrared spectroscopy are also important tools for the mechanistic understanding of electrocatalysis. Together with explainable artificial intelligence (AI), which can relate the results to underlying physics, these analyses could be used to discover descriptors hidden in spectra that could lead to new breakthroughs in electrocatalyst discovery and optimization.

Fuel cell and electrolyser device management

A fuel cell is an electrochemical device that can be used to convert the chemical energy of a fuel (such as hydrogen) into electrical energy. An electrolyser transforms electrical energy into chemical energy (such as in water splitting to generate hydrogen). ML has been used to optimize and manage their performance, predict degradation and device lifetime as well as detect and diagnose faults. Using a hybrid method consisting of an extreme learning machine, genetic algorithms and wavelet analysis, the degradation in proton-exchange membrane fuel cells has been predicted 85 , 86 . Electrochemical impedance measurements used as input for an artificial neural network have enabled fault detection and isolation in a high-temperature stack of proton-exchange membrane fuel cells 87 , 88 .

ML approaches can also be employed to diagnose faults, such as fuel and air leakage issues, in solid oxide fuel cell stacks. Artificial neural networks can predict the performance of solid oxide fuel cells under different operating conditions 89 . In addition, ML has been applied to optimize the performance of solid oxide electrolysers, for CO 2 /H 2 O reduction 90 , and chloralkali electrolysers 91 .

In the future, the use of ML for fuel cells could be combined with multiscale modelling to improve their design, for example to minimize Ohmic losses and optimize catalyst loading. For practical applications, fuel cells may be subject to fluctuations in energy output requirements (for example, when used in vehicles). ML models could be used to determine the effects of such fluctuations on the long-term durability and performance of fuel cells, similar to what has been done for predicting the state of health and lifetime for batteries. Furthermore, it remains to be seen whether the ML techniques for fuel cells can be easily generalized to electrolysers and vice versa, using transfer learning for example, given that they are essentially reactions in reverse.

Smart power grids

A power grid is responsible for delivering electrical energy from producers (such as power plants and solar farms) to consumers (such as homes and offices). However, energy fluctuations from intermittent renewable energy generators can render the grid vulnerable 92 . ML algorithms can be used to optimize the automatic generation control of power grids, which controls the power output of multiple generators in an energy system. For example, when a relaxed deep learning model was used as a unified timescale controller for the automatic generation control unit, the total operational cost was reduced by up to 80% compared with traditional heuristic control strategies 93 (Fig.  2d ). A smart generation control strategy based on multi-agent reinforcement learning was found to improve the control performance by around 10% compared with other ML algorithms 94 .

Accurate demand and load prediction can support decision-making operations in energy systems for proper load scheduling and power allocation. Multiple ML methods have been proposed to precisely predict the demand load: for example, long short-term memory was used to successfully and accurately predict hourly building load 95 . Short-term load forecasting of diverse customers (such as retail businesses) using a deep neural network and cross-building energy demand forecasting using a deep belief network have also been demonstrated effectively 96 , 97 .

Demand-side management consists of a set of mechanisms that shape consumer electricity consumption by dynamically adjusting the price of electricity. These include reducing (peak shaving), increasing (load growth) and rescheduling (load shifting) the energy demand, which allows for flexible balancing of renewable electricity generation and load 98 . A reinforcement-learning-based algorithm resulted in substantial cost reduction for both the service provider and customer 99 . A decentralized learning-based residential demand scheduling technique successfully shifted up to 35% of the energy demand to periods of high wind availability, substantially saving power costs compared with the unscheduled energy demand scenario 100 . Load forecasting using a multi-agent approach integrates load prediction with reinforcement learning algorithms to shift energy usage (for example, to different electrical devices in a household) for its optimization 101 . This approach reduced peak usage by more than 30% and increased off-peak usage by 50%, reducing the cost and energy losses associated with energy storage.

Opportunities for ML in renewable energy

ML provides the opportunity to enable substantial further advances in different areas of the energy materials field, which share similar materials-related challenges (Fig.  3 ). There are also grand challenges for ML application in smart grid and policy optimization.

a | Energy materials present additional modelling challenges. Machine learning (ML) could help in the representation of structurally complex structures, which can include disordering, dislocations and amorphous phases. b | Flexible models that scale efficiently with varied dataset sizes are in demand, and ML could help to develop robust predictive models. The yellow dots stand for the addition of unreliable datasets that could harm the prediction accuracy of the ML model. c | Synthesis route prediction remains to be solved for the design of a novel material. In the ternary phase diagram, the dots stand for the stable compounds in that corresponding phase space and the red dot for the targeted compound. Two possible synthesis pathways are compared for a single compound. The score obtained would reflect the complexity, cost and so on of one synthesis pathway. d | ML-aided phase degradation prediction could boost the development of materials with enhanced cyclability. The shaded region represents the rocksalt phase, which grows inside the layered phase. The arrow marks the growth direction. e | The use of ML models could help in optimizing energy generation and energy consumption. Automating the decision-making processes associated with dynamic power supplies using ML will make the power distribution more efficient. f | Energy policy is the manner in which an entity (for example, a government) addresses its energy issues, including conversion, distribution and utilization, where ML could be used to optimize the corresponding economy.

Materials with novel geometries

A ML representation is effective when it captures the inherent properties of the system (such as its physical symmetries) and can be utilized in downstream ancillary tasks, such as transfer learning to new predictive tasks, building new knowledge using visualization or attribution and generating similar data distributions with generative models 102 .

For materials, the inputs are molecules or crystal structures whose physical properties are modelled by the Schrödinger equation. Designing a general representation of materials that reflects these properties is an ongoing research problem. For molecular systems, several representations have been used successfully, including fingerprints 103 , SMILES 104 , self-referencing embedded strings (SELFIES) 105 and graphs 106 , 107 , 108 . Representing crystalline materials has the added complexity of needing to incorporate periodicity in the representation. Methods like the smooth overlap of atomic positions 109 , Voronoi tessellation 110 , 111 , diffraction images 112 , multi-perspective fingerprints 113 and graph-based algorithms 27 , 114 have been suggested, but typically lack the capability for structure reconstruction.

Complex structural systems found in energy materials present additional modelling challenges (Fig.  3a ): a large number of atoms (such as in reticular frameworks or polymers), specific symmetries (such as in molecules with a particular space group and for reticular frameworks belonging to a certain topology), atomic disordering, partial occupancy, or amorphous phases (leading to an enormous combinatorial space), defects and dislocations (such as interfaces and grain boundaries) and low-dimensionality materials (as in nanoparticles). Reduction approximations alleviate the first issue (using, for example, RFcode for reticular framework representation) 8 , but the remaining several problems warrant intensive future research efforts.

Self- supervised learning , which seeks to lever large amounts of synthetic labels and tasks to continue learning without experimental labels 115 , multi-task learning 116 , in which multiple material properties can be modelled jointly to exploit correlation structure between properties, and meta-learning 117 , which looks at strategies that allow models to perform better in new datasets or in out-of-distribution data, all offer avenues to build better representations. On the modelling front, new advances in attention mechanisms 118 , 119 , graph neural networks 120 and equivariant neural networks 121 expand our range of tools with which to model interactions and expected symmetries.

Robust predictive models

Predictive models are the first step when building a pipeline that seeks materials with desired properties. A key component for building these models is training data; more data will often translate into better-performing models, which in turn will translate into better accuracy in the prediction of new materials. Deep learning models tend to scale more favourably with dataset size than traditional ML approaches (such as random forests). Dataset quality is also essential. However, experiments are usually conducted under diverse conditions with large variation in untracked variables (Fig.  3b ). Additionally, public datasets are more likely to suffer from publication bias, because negative results are less likely to be published even though they are just as important as positive results when training statistical models 122 .

Addressing these issues require transparency and standardization of the experimental data reported in the literature. Text and natural language processing strategies could then be employed to extract data from the literature 77 . Data should be reported with the belief that it will eventually be consolidated in a database, such as the MatD3 database 123 . Autonomous laboratory techniques will help to address this issue 19 , 124 . Structured property databases such as the Materials Project 122 and the Harvard Clean Energy Project 125 can also provide a large amount of data. Additionally, different energy fields — energy storage, harvesting and conversion — should converge upon a standard and uniform way to report data. This standard should be continuously updated; as researchers continue to learn about the systems they are studying, conditions that were previously thought to be unimportant will become relevant.

New modelling approaches that work in low-data regimes, such as data-efficient models, dataset-building strategies (active sampling) 126 and data-augmentation techniques, are also important 127 . Uncertainty quantification , data efficiency, interpretability and regularization are important considerations that improve the robustness of ML models. These considerations relate to the notion of generalizability: predictions should generalize to a new class of materials that is out of the distribution of the original dataset. Researchers can attempt to model how far away new data points are from the training set 128 or the variability in predicted labels with uncertainty quantification 129 . Neural networks are a flexible model class, and often models can be underspecified 130 . Incorporating regularization, inductive biases or priors can boost the credibility of a model. Another way to create trustable models could be to enhance the interpretability of ML algorithms by deriving feature relevance and scoring their importance 131 . This strategy could help to identify potential chemically meaningful features and form a starting point for understanding latent factors that dominate material properties. These techniques can also identify the presence of model bias and overfitting, as well as improving generalization and performance 132 , 133 , 134 .

Stable and synthesizable new materials

The formation energy of a compound is used to estimate its stability and synthesizability 135 , 136 . Although negative values usually correspond to stable or synthesizable compounds, slightly positive formation energies below a limit lead to metastable phases with unclear synthesizability 137 , 138 . This is more apparent when investigating unexplored chemical spaces with undetermined equilibrium ground states; yet often the metastable phases exhibit superior properties, as seen in photovoltaics 136 , 139 and ion conductors 140 , for example. It is thus of interest to develop a method to evaluate the synthesizability of metastable phases (Fig.  3c ). Instead of estimating the probability that a particular phase can be synthesized, one can instead evaluate its synthetic complexity using ML. In organic chemistry, synthesis complexity is evaluated according to the accessibility of the phases’ synthesis route 141 or precedent reaction knowledge 142 . Similar methodologies can be applied to the inorganic field with the ongoing design of automated synthesis-planning algorithms for inorganic materials 143 , 144 .

Synthesis and evaluation of a new material alone does not ensure that material will make it to market; material stability is a crucial property that takes a long time to evaluate. Degradation is a generally complex process that occurs through the loss of active matter or growth of inactive phases (such as the rocksalt phases formed in layered Li-ion battery electrodes 145 (Fig.  3d ) or the Pt particle agglomeration in fuel cells 146 ) and/or propagation of defects (such as cracks in cycled battery electrode 147 ). Microscopies such as electron microscopy 148 and simulations such as continuum mechanics modelling 149 are often used to investigate growth and propagation dynamics (that is, phase boundary and defect surface movements versus time). However, these techniques are usually expensive and do not allow rapid degradation prediction. Deep learning techniques such as convolutional neural networks and recurrent neural networks may be able to predict the phase boundary and/or defect pattern evolution under certain conditions after proper training 150 . Similar models can then be built to understand multiple degradation phenomena and aid the design of materials with improved cycle life.

Optimized smart power grids

A promising prospect of ML in smart grids is automating the decision-making processes that are associated with dynamic power supplies to distribute power most efficiently (Fig.  3e ). Practical deployment of ML technologies into physical systems remains difficult because of data scarcity and the risk-averse mindset of policymakers. The collection of and access to large amounts of diverse data is challenging owing to high cost, long delays and concerns over compliance and security 151 . For instance, to capture the variation of renewable resources owing to peak or off-peak and seasonal attributes, long-term data collections are implemented for periods of 24 hours to several years 152 . Furthermore, although ML algorithms are ideally supposed to account for all uncertainties and unpredictable situations in energy systems, the risk-adverse mindset in the energy management industry means that implementation still relies on human decision-making 153 .

An ML-based framework that involves a digital twin of the physical system can address these problems 154 , 155 . The digital twin represents the digitalized cyber models of the physical system and can be constructed from physical laws and/or ML models trained using data sampled from the physical system. This approach aims to accurately simulate the dynamics of the physical system, enabling relatively fast generation of large amounts of high-quality synthetic data at low cost. Notably, because ML model training and validation is performed on the digital twin, there is no risk to the actual physical system. Based on the prediction results, suitable actions can be suggested and then implemented in the physical system to ensure stability and/or improve system operation.

Policy optimization

Finally, research is generally focused on one narrow aspect of a larger problem; we argue that energy research needs a more integrated approach 156 (Fig.  3f ). Energy policy is the manner in which an entity, such as the government, addresses its energy issues, including conversion, distribution and utilization. ML has been used in the fields of energy economics finance for performance diagnostics (such as for oil wells), energy generation (such as wind power) and consumption (such as power load) forecasts and system lifespan (such as battery cell life) and failure (such as grid outage) prediction 157 . They have also been used for energy policy analysis and evaluation (for example, for estimating energy savings). A natural extension of ML models is to use them for policy optimization 158 , 159 , a concept that has not yet seen widespread use. We posit that the best energy policies — including the deployment of the newly discovered materials — can be improved and augmented with ML and should be discussed in research reporting accelerated energy technology platforms.

Conclusions

To summarize, ML has the potential to enable breakthroughs in the development and deployment of sustainable energy techniques. There have been remarkable achievements in many areas of energy technology, from materials design and device management to system deployment. ML is particularly well suited to discovering new materials, and researchers in the field are expecting ML to bring up new materials that may revolutionize the energy industry. The field is still nascent, but there is conclusive evidence that ML is at least able to expose the same trends that human researchers have noticed over decades of research. The ML field itself is still seeing rapid development, with new methodologies being reported daily. It will take time to develop and adopt these methodologies to solve specific problems in materials science. We believe that for ML to truly accelerate the deployment of sustainable energy, it should be deployed as a tool, similar to a synthesis procedure, characterization equipment or control apparatus. Researchers using ML to accelerate energy technology discovery should judge the success of the method primarily on the advances it enables. To this end, we have proposed the XPIs and some areas in which we hope to see ML deployed.

Davidson, D. J. Exnovating for a renewable energy transition. Nat. Energy 4 , 254–256 (2019).

Article   Google Scholar  

Horowitz, C. A. Paris agreement. Int. Leg. Mater. 55 , 740–755 (2016).

International Energy Agency 2018 World Energy Outlook: Executive Summary https://www.iea.org/reports/world-energy-outlook-2018 (OECD/IEA, 2018).

Chu, S., Cui, Y. & Liu, N. The path towards sustainable energy. Nat. Mater. 16 , 16–22 (2017).

Maine, E. & Garnsey, E. Commercializing generic technology: the case of advanced materials ventures. Res. Policy 35 , 375–393 (2006).

De Luna, P., Wei, J., Bengio, Y., Aspuru-Guzik, A. & Sargent, E. Use machine learning to find energy materials. Nature 552 , 23–27 (2017).

Wang, H., Lei, Z., Zhang, X., Zhou, B. & Peng, J. A review of deep learning for renewable energy forecasting. Energy Convers. Manag. 198 , 111799–111814 (2019).

Yao, Z. et al. Inverse design of nanoporous crystalline reticular materials with deep generative models. Nat. Mach. Intell. 3 , 76–86 (2021).

Rosen, A. S. et al. Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery. Matter 4 , 1578–1597 (2021).

Article   CAS   Google Scholar  

Jordan, M. I. & Mitchell, T. M. Machine learning: trends, perspectives, and prospects. Science 349 , 255–260 (2015).

Personal, E., Guerrero, J. I., Garcia, A., Peña, M. & Leon, C. Key performance indicators: a useful tool to assess Smart Grid goals. Energy 76 , 976–988 (2014).

Helmus, J. & den Hoed, R. Key performance indicators of charging infrastructure. World Electr. Veh. J. 8 , 733–741 (2016).

Struck, M.-M. Vaccine R&D success rates and development times. Nat. Biotechnol. 14 , 591–593 (1996).

Moore, G. E. Cramming more components onto integrated circuits. Electronics 38 , 114–116 (1965).

Google Scholar  

Wetterstrand, K. A. DNA sequencing costs: data. NHGRI Genome Sequencing Program (GSP) www.genome.gov/sequencingcostsdata (2020).

Rajkomar, A., Dean, J. & Kohane, I. Machine learning in medicine. N. Engl. J. Med. 380 , 1347–1358 (2019).

Jeong, J. et al. Pseudo-halide anion engineering for α-FAPbI3 perovskite solar cells. Nature 592 , 381–385 (2021).

NREL. Best research-cell efficiency chart. NREL https://www.nrel.gov/pv/cell-efficiency.html (2021).

Burger, B. et al. A mobile robotic chemist. Nature 583 , 237–241 (2020).

Clark, M. A. et al. Design, synthesis and selection of DNA-encoded small-molecule libraries. Nat. Chem. Biol. 5 , 647–654 (2009).

Pilania, G., Gubernatis, J. E. & Lookman, T. Multi-fidelity machine learning models for accurate bandgap predictions of solids. Comput. Mater. Sci. 129 , 156–163 (2017).

Pilania, G. et al. Machine learning bandgaps of double perovskites. Sci. Rep. 6 , 19375 (2016).

Lu, S. et al. Accelerated discovery of stable lead-free hybrid organic-inorganic perovskites via machine learning. Nat. Commun. 9 , 3405 (2018).

Askerka, M. et al. Learning-in-templates enables accelerated discovery and synthesis of new stable double perovskites. J. Am. Chem. Soc. 141 , 3682–3690 (2019).

Jain, A. & Bligaard, T. Atomic-position independent descriptor for machine learning of material properties. Phys. Rev. B 98 , 214112 (2018).

Choubisa, H. et al. Crystal site feature embedding enables exploration of large chemical spaces. Matter 3 , 433–448 (2020).

Xie, T. & Grossman, J. C. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties. Phys. Rev. Lett. 120 , 145301–145306 (2018).

Broberg, D. et al. PyCDT: a Python toolkit for modeling point defects in semiconductors and insulators. Comput. Phys. Commun. 226 , 165–179 (2018).

Roch, L. M. et al. ChemOS: an orchestration software to democratize autonomous discovery. PLoS ONE 15 , 1–18 (2020).

Wei, L., Xu, X., Gurudayal, Bullock, J. & Ager, J. W. Machine learning optimization of p-type transparent conducting films. Chem. Mater. 31 , 7340–7350 (2019).

Schubert, M. F. et al. Design of multilayer antireflection coatings made from co-sputtered and low-refractive-index materials by genetic algorithm. Opt. Express 16 , 5290–5298 (2008).

Wang, C., Yu, S., Chen, W. & Sun, C. Highly efficient light-trapping structure design inspired by natural evolution. Sci. Rep. 3 , 1025 (2013).

Ripalda, J. M., Buencuerpo, J. & García, I. Solar cell designs by maximizing energy production based on machine learning clustering of spectral variations. Nat. Commun. 9 , 5126 (2018).

Demant, M., Virtue, P., Kovvali, A., Yu, S. X. & Rein, S. Learning quality rating of As-Cut mc-Si wafers via convolutional regression networks. IEEE J. Photovolt. 9 , 1064–1072 (2019).

Musztyfaga-Staszuk, M. & Honysz, R. Application of artificial neural networks in modeling of manufactured front metallization contact resistance for silicon solar cells. Arch. Metall. Mater. 60 , 1673–1678 (2015).

Sendek, A. D. et al. Holistic computational structure screening of more than 12000 candidates for solid lithium-ion conductor materials. Energy Environ. Sci. 10 , 306–320 (2017).

Ahmad, Z., Xie, T., Maheshwari, C., Grossman, J. C. & Viswanathan, V. Machine learning enabled computational screening of inorganic solid electrolytes for suppression of dendrite formation in lithium metal anodes. ACS Cent. Sci. 4 , 996–1006 (2018).

Zhang, Y. et al. Unsupervised discovery of solid-state lithium ion conductors. Nat. Commun. 10 , 5260 (2019).

Doan, H. A. et al. Quantum chemistry-informed active learning to accelerate the design and discovery of sustainable energy storage materials. Chem. Mater. 32 , 6338–6346 (2020).

Chemali, E., Kollmeyer, P. J., Preindl, M. & Emadi, A. State-of-charge estimation of Li-ion batteries using deep neural networks: a machine learning approach. J. Power Sources 400 , 242–255 (2018).

Richardson, R. R., Osborne, M. A. & Howey, D. A. Gaussian process regression for forecasting battery state of health. J. Power Sources 357 , 209–219 (2017).

Berecibar, M. et al. Online state of health estimation on NMC cells based on predictive analytics. J. Power Sources 320 , 239–250 (2016).

Severson, K. A. et al. Data-driven prediction of battery cycle life before capacity degradation. Nat. Energy 4 , 383–391 (2019).

Attia, P. M. et al. Closed-loop optimization of fast-charging protocols for batteries with machine learning. Nature 578 , 397–402 (2020).

Joshi, R. P. et al. Machine learning the voltage of electrode materials in metal-ion batteries. ACS Appl. Mater. Interf. 11 , 18494–18503 (2019).

Cubuk, E. D., Sendek, A. D. & Reed, E. J. Screening billions of candidates for solid lithium-ion conductors: a transfer learning approach for small data. J. Chem. Phys. 150 , 214701 (2019).

Sendek, A. D. et al. Machine learning-assisted discovery of solid Li-ion conducting materials. Chem. Mater. 31 , 342–352 (2019).

Kim, S., Jinich, A. & Aspuru-Guzik, A. MultiDK: a multiple descriptor multiple kernel approach for molecular discovery and its application to organic flow battery electrolytes. J. Chem. Inf. Model. 57 , 657–668 (2017).

Jinich, A., Sanchez-Lengeling, B., Ren, H., Harman, R. & Aspuru-Guzik, A. A mixed quantum chemistry/machine learning approach for the fast and accurate prediction of biochemical redox potentials and its large-scale application to 315000 redox reactions. ACS Cent. Sci. 5 , 1199–1210 (2019).

Sarkar, A. et al. High entropy oxides for reversible energy storage. Nat. Commun. 9 , 3400 (2018).

Choudhury, S. et al. Stabilizing polymer electrolytes in high-voltage lithium batteries. Nat. Commun. 10 , 3091 (2019).

Sanchez-Lengeling, B. & Aspuru-Guzik, A. Inverse molecular design using machine learning: generative models for matter engineering. Science 361 , 360–365 (2018).

Ng, M.-F., Zhao, J., Yan, Q., Conduit, G. J. & Seh, Z. W. Predicting the state of charge and health of batteries using data-driven machine learning. Nat. Mach. Intell. 2 , 161–170 (2020).

Steinmann, S. N. & Seh, Z. W. Understanding electrified interfaces. Nat. Rev. Mater. 6 , 289–291 (2021).

Kandasamy, N., Badrinarayanan, R., Kanamarlapudi, V., Tseng, K. & Soong, B.-H. Performance analysis of machine-learning approaches for modeling the charging/discharging profiles of stationary battery systems with non-uniform cell aging. Batteries 3 , 18 (2017).

Wei, Q., Liu, D. & Shi, G. A novel dual iterative Q-learning method for optimal battery management in smart residential environments. IEEE Trans. Ind. Electron. 62 , 2509–2518 (2015).

Murphey, Y. L. et al. Intelligent hybrid vehicle power control — Part II: online intelligent energy management. IEEE Trans. Vehicular Technol. 62 , 69–79 (2013).

Seh, Z. W. et al. Combining theory and experiment in electrocatalysis: insights into materials design. Science 355 , eaad4998 (2017).

Staffell, I. et al. The role of hydrogen and fuel cells in the global energy system. Energy Environ. Sci. 12 , 463–491 (2019).

Montoya, J. H. H. et al. Materials for solar fuels and chemicals. Nat. Mater. 16 , 70–81 (2017).

Pérez-Ramírez, J. & López, N. Strategies to break linear scaling relationships. Nat. Catal. 2 , 971–976 (2019).

Shi, C., Hansen, H. A., Lausche, A. C. & Norskov, J. K. Trends in electrochemical CO 2 reduction activity for open and close-packed metal surfaces. Phys. Chem. Chem. Phys. 16 , 4720–4727 (2014).

Calle-Vallejo, F., Loffreda, D., Koper, M. T. M. & Sautet, P. Introducing structural sensitivity into adsorption-energy scaling relations by means of coordination numbers. Nat. Chem. 7 , 403–410 (2015).

Ulissi, Z. W. et al. Machine-learning methods enable exhaustive searches for active bimetallic facets and reveal active site motifs for CO 2 reduction. ACS Catal. 7 , 6600–6608 (2017).

Nørskov, J. K., Studt, F., Abild-Pedersen, F. & Bligaard, T. Activity and selectivity maps. In Fundamental Concepts in Heterogeneous Catalysis 97–113 (John Wiley, 2014).

Garijo del Río, E., Mortensen, J. J. & Jacobsen, K. W. Local Bayesian optimizer for atomic structures. Phys. Rev. B 100 , 104103 (2019).

Jørgensen, M. S., Larsen, U. F., Jacobsen, K. W. & Hammer, B. Exploration versus exploitation in global atomistic structure optimization. J. Phys. Chem. A 122 , 1504–1509 (2018).

Jacobsen, T. L., Jørgensen, M. S. & Hammer, B. On-the-fly machine learning of atomic potential in density functional theory structure optimization. Phys. Rev. Lett. 120 , 026102 (2018).

Peterson, A. A. Acceleration of saddle-point searches with machine learning. J. Chem. Phys. 145 , 074106 (2016).

Ulissi, Z. W., Medford, A. J., Bligaard, T. & Nørskov, J. K. To address surface reaction network complexity using scaling relations machine learning and DFT calculations. Nat. Commun. 8 , 14621 (2017).

Huang, Y., Chen, Y., Cheng, T., Wang, L.-W. & Goddard, W. A. Identification of the selective sites for electrochemical reduction of CO to C 2+ products on copper nanoparticles by combining reactive force fields, density functional theory, and machine learning. ACS Energy Lett. 3 , 2983–2988 (2018).

Chen, Y., Huang, Y., Cheng, T. & Goddard, W. A. Identifying active sites for CO 2 reduction on dealloyed gold surfaces by combining machine learning with multiscale simulations. J. Am. Chem. Soc. 141 , 11651–11657 (2019).

Lai, Y., Jones, R. J. R., Wang, Y., Zhou, L. & Gregoire, J. M. Scanning electrochemical flow cell with online mass spectroscopy for accelerated screening of carbon dioxide reduction electrocatalysts. ACS Comb. Sci. 21 , 692–704 (2019).

Lai, Y. et al. The sensitivity of Cu for electrochemical carbon dioxide reduction to hydrocarbons as revealed by high throughput experiments. J. Mater. Chem. A 7 , 26785–26790 (2019).

Tran, K. & Ulissi, Z. W. Active learning across intermetallics to guide discovery of electrocatalysts for CO 2 reduction and H 2 evolution. Nat. Catal. 1 , 696–703 (2018).

Zhong, M. et al. Accelerated discovery of CO 2 electrocatalysts using active machine learning. Nature 581 , 178–183 (2020).

Tshitoyan, V. et al. Unsupervised word embeddings capture latent knowledge from materials science literature. Nature 571 , 95–98 (2019).

Yao, Y. et al. Carbothermal shock synthesis of high-entropy-alloy nanoparticles. Science 359 , 1489–1494 (2018).

Yao, Y. et al. High-throughput, combinatorial synthesis of multimetallic nanoclusters. Proc. Natl Acad. Sci. USA 117 , 6316–6322 (2020).

Timoshenko, J. & Roldan Cuenya, B. In situ/operando electrocatalyst characterization by X-ray absorption spectroscopy. Chem. Rev. 121 , 882–961 (2021).

Zheng, C., Chen, C., Chen, Y. & Ong, S. P. Random forest models for accurate identification of coordination environments from X-ray absorption near-edge structure. Patterns 1 , 100013–100023 (2020).

Torrisi, S. B. et al. Random forest machine learning models for interpretable X-ray absorption near-edge structure spectrum-property relationships. npj Comput. Mater. 6 , 109 (2020).

Timoshenko, J. et al. Neural network approach for characterizing structural transformations by X-ray absorption fine structure spectroscopy. Phys. Rev. Lett. 120 , 225502 (2018).

Marcella, N. et al. Neural network assisted analysis of bimetallic nanocatalysts using X-ray absorption near edge structure spectroscopy. Phys. Chem. Chem. Phys. 22 , 18902–18910 (2020).

Chen, K., Laghrouche, S. & Djerdir, A. Degradation model of proton exchange membrane fuel cell based on a novel hybrid method. Appl. Energy 252 , 113439–113447 (2019).

Ma, R. et al. Data-driven proton exchange membrane fuel cell degradation predication through deep learning method. Appl. Energy 231 , 102–115 (2018).

Jeppesen, C. et al. Fault detection and isolation of high temperature proton exchange membrane fuel cell stack under the influence of degradation. J. Power Sources 359 , 37–47 (2017).

Liu, J. et al. Sequence fault diagnosis for PEMFC water management subsystem using deep learning with t-SNE. IEEE Access. 7 , 92009–92019 (2019).

Ansari, M. A., Rizvi, S. M. A. & Khan, S. Optimization of electrochemical performance of a solid oxide fuel cell using artificial neural network. in 2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT) 4230–4234 (IEEE, 2016).

Zhang, C. et al. Modelling of solid oxide electrolyser cell using extreme learning machine. Electrochim. Acta 251 , 137–144 (2017).

Esche, E., Weigert, J., Budiarto, T., Hoffmann, C. & Repke, J.-U. Optimization under uncertainty based on a data-driven model for a chloralkali electrolyzer cell. Computer-aided Chem. Eng. 46 , 577–582 (2019).

Siddaiah, R. & Saini, R. P. A review on planning, configurations, modeling and optimization techniques of hybrid renewable energy systems for off grid applications. Renew. Sustain. Energy Rev. 58 , 376–396 (2016).

Yin, L., Yu, T., Zhang, X. & Yang, B. Relaxed deep learning for real-time economic generation dispatch and control with unified time scale. Energy 149 , 11–23 (2018).

Yu, T., Wang, H. Z., Zhou, B., Chan, K. W. & Tang, J. Multi-agent correlated equilibrium Q ( λ ) learning for coordinated smart generation control of interconnected power grids. IEEE Trans. Power Syst. 30 , 1669–1679 (2015).

Marino, D. L., Amarasinghe, K. & Manic, M. Building energy load forecasting using deep neural networks. in IECON Proceedings (Industrial Electronics Conference) 7046–7051 (IECON, 2016).

Ryu, S., Noh, J. & Kim, H. Deep neural network based demand side short term load forecasting. in 2016 IEEE International Conference on Smart Grid Communications (SmartGridComm 2016) 308–313 (IEEE, 2016).

Mocanu, E., Nguyen, P. H., Kling, W. L. & Gibescu, M. Unsupervised energy prediction in a Smart Grid context using reinforcement cross-building transfer learning. Energy Build. 116 , 646–655 (2016).

Lund, P. D., Lindgren, J., Mikkola, J. & Salpakari, J. Review of energy system flexibility measures to enable high levels of variable renewable electricity. Renew. Sustain. Energy Rev. 45 , 785–807 (2015).

Kim, B. G., Zhang, Y., Van Der Schaar, M. & Lee, J. W. Dynamic pricing and energy consumption scheduling with reinforcement learning. IEEE Trans. Smart Grid 7 , 2187–2198 (2016).

Dusparic, I., Taylor, A., Marinescu, A., Cahill, V. & Clarke, S. Maximizing renewable energy use with decentralized residential demand response. in 2015 IEEE 1st International Smart Cities Conference (ISC2 2015 ) 1–6 (IEEE, 2015).

Dusparic, I., Harris, C., Marinescu, A., Cahill, V. & Clarke, S. Multi-agent residential demand response based on load forecasting. in 2013 1st IEEE Conference on Technologies for Sustainability (SusTech 2013) 90–96 (IEEE, 2013).

Bengio, Y., Courville, A. & Vincent, P. Representation learning: a review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence 35 , 1798–1828 (2013).

Duvenaud, D. et al. Convolutional networks on graphs for learning molecular fingerprints. in Advances In Neural Information Processing Systems 2224–2232 (NIPS, 2015).

Gómez-Bombarelli, R. et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Cent. Sci. 4 , 268–276 (2018).

Krenn, M., Hase, F., Nigam, A., Friederich, P. & Aspuru-Guzik, A. Self-referencing embedded strings (SELFIES): a 100% robust molecular string representation. Mach. Learn. Sci. Technol. 1 , 045024–045031 (2020).

Jin, W., Barzilay, R. & Jaakkola, T. Junction tree variational autoencoder for molecular graph generation. Mach. Learn. 5 , 3632–3648 (2018).

You, J., Liu, B., Ying, R., Pande, V. & Leskovec, J. Graph convolutional policy network for goal-directed molecular graph generation. Adv. Neural Inf. Process. Syst. 31 , 6412–6422 (2018).

Liu, Q., Allamanis, M., Brockschmidt, M. & Gaunt, A. L. Constrained graph variational autoencoders for molecule design. in Proceedings of the 32nd International Conference on Neural Information Processing Systems (NIPS’18) 7806–7815 (Curran Associates Inc., 2018).

Bartók, A. P., Kondor, R. & Csányi, G. On representing chemical environments. Phys. Rev. B 87 , 184115 (2013).

Ward, L. et al. Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations. Phys. Rev. B 96 , 024104 (2017).

Isayev, O. et al. Universal fragment descriptors for predicting properties of inorganic crystals. Nat. Commun. 8 , 15679 (2017).

Ziletti, A., Kumar, D., Scheffler, M. & Ghiringhelli, L. M. Insightful classification of crystal structures using deep learning. Nat. Commun. 9 , 2775 (2018).

Ryan, K., Lengyel, J. & Shatruk, M. Crystal structure prediction via deep learning. J. Am. Chem. Soc. 140 , 10158–10168 (2018).

Park, C. W. & Wolverton, C. Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery. Phys. Rev. Mater. 4 , 063801 (2020).

Liu, X. et al. Self-Supervised Learning: Generative or Contrastive (IEEE, 2020).

Ruder, S. An overview of multi-task learning in deep neural networks. Preprint at https://doi.org/10.48550/arXiv.1706.05098 (2017).

Hospedales, T., Antoniou, A., Micaelli, P. & Storkey, A. Meta-learning in neural networks: a survey. IEEE Transactions on Pattern Analysis & Machine Intelligence 44 , 5149–5169 (2020).

Vaswani, A. et al. Attention is all you need. in Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS’17) 6000–6010 (Curran Associates Inc., 2017).

Veličković, P. et al. Graph attention networks. Preprint at https://doi.org/10.48550/arXiv.1710.10903 (2017).

Battaglia, P. W. et al. Relational inductive biases, deep learning, and graph networks. Preprint at https://doi.org/10.48550/arXiv.1806.01261 (2018).

Satorras, V. G., Hoogeboom, E. & Welling, M. E(n) equivariant graph neural networks. Preprint at https://doi.org/10.48550/arXiv.2102.09844 (2021).

Jain, A. et al. Commentary: The Materials Project: a materials genome approach to accelerating materials innovation. Apl. Mater. 1 , 011002–011012 (2013).

Laasner, R. et al. MatD3: a database and online presentation package for research data supporting materials discovery, design, and dissemination. J. Open Source Softw. 5 , 1945–1947 (2020).

Coley, C. W. et al. A robotic platform for flow synthesis of organic compounds informed by AI planning. Science 365 , 557 (2019).

Hachmann, J. et al. The harvard clean energy project: large-scale computational screening and design of organic photovoltaics on the world community grid. J. Phys. Chem. Lett. 2 , 2241–2251 (2011).

Bıyık, E., Wang, K., Anari, N. & Sadigh, D. Batch active learning using determinantal point processes. Preprint at https://doi.org/10.48550/arXiv.1906.07975 (2019).

Hoffmann, J. et al. Machine learning in a data-limited regime: augmenting experiments with synthetic data uncovers order in crumpled sheets. Sci. Adv. 5 , eaau6792 (2019).

Liu, J. Z. et al. Simple and principled uncertainty estimation with deterministic deep learning via distance awareness. Adv. Neural Inf. Process Syst. 33 , 7498–7512 (2020).

Lakshminarayanan, B., Pritzel, A. & Blundell, C. Simple and scalable predictive uncertainty estimation using deep ensembles. In Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS’17) 6405–6416 (Curran Associates Inc., 2017).

D’Amour, A. et al. Underspecification presents challenges for credibility in modern machine learning. Preprint at https://doi.org/10.48550/arXiv.2011.03395 (2020).

Barredo Arrieta, A. et al. Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion. 58 , 82–115 (2020).

Ribeiro, M. T., Singh, S. & Guestrin, C. “Why should I trust you?” Explaining the predictions of any classifier. in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 1135–1144 (2016).

Lundberg, S. & Lee, S.-I. An unexpected unity among methods for interpreting model predictions. Preprint at https://arxiv.org/abs/1611.07478 (2016).

Bach, S. et al. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE 10 , e0130140 (2015).

Sun, W. et al. The thermodynamic scale of inorganic crystalline metastability. Sci. Adv. 2 , e1600225 (2016).

Aykol, M., Dwaraknath, S. S., Sun, W. & Persson, K. A. Thermodynamic limit for synthesis of metastable inorganic materials. Sci. Adv. 4 , eaaq0148 (2018).

Wei, J. N., Duvenaud, D. & Aspuru-Guzik, A. Neural networks for the prediction of organic chemistry reactions. ACS Cent. Sci. 2 , 725–732 (2016).

Coley, C. W. et al. A graph-convolutional neural network model for the prediction of chemical reactivity. Chem. Sci. 10 , 370–377 (2019).

Nagabhushana, G. P., Shivaramaiah, R. & Navrotsky, A. Direct calorimetric verification of thermodynamic instability of lead halide hybrid perovskites. Proc. Natl Acad. Sci. USA 113 , 7717–7721 (2016).

Sanna, S. et al. Enhancement of the chemical stability in confined δ-Bi 2 O 3 . Nat. Mater. 14 , 500–504 (2015).

Podolyan, Y., Walters, M. A. & Karypis, G. Assessing synthetic accessibility of chemical compounds using machine learning methods. J. Chem. Inf. Model. 50 , 979–991 (2010).

Coley, C. W., Rogers, L., Green, W. H. & Jensen, K. F. SCScore: synthetic complexity learned from a reaction corpus. J. Chem. Inf. Model. 58 , 252–261 (2018).

Kim, E. et al. Inorganic materials synthesis planning with literature-trained neural networks. J. Chem. Inf. Model. 60 , 1194–1201 (2020).

Huo, H. et al. Semi-supervised machine-learning classification of materials synthesis procedures. npj Comput. Mater. 5 , 62 (2019).

Tian, C., Lin, F. & Doeff, M. M. Electrochemical characteristics of layered transition metal oxide cathode materials for lithium ion batteries: surface, bulk behavior, and thermal properties. Acc. Chem. Res. 51 , 89–96 (2018).

Guilminot, E., Corcella, A., Charlot, F., Maillard, F. & Chatenet, M. Detection of Pt z + ions and Pt nanoparticles inside the membrane of a used PEMFC. J. Electrochem. Soc. 154 , B96 (2007).

Pender, J. P. et al. Electrode degradation in lithium-ion batteries. ACS Nano 14 , 1243–1295 (2020).

Li, Y. et al. Atomic structure of sensitive battery materials and interfaces revealed by cryo-electron microscopy. Science 358 , 506–510 (2017).

Wang, H. Numerical modeling of non-planar hydraulic fracture propagation in brittle and ductile rocks using XFEM with cohesive zone method. J. Pet. Sci. Eng. 135 , 127–140 (2015).

Hsu, Y.-C., Yu, C.-H. & Buehler, M. J. Using deep learning to predict fracture patterns in crystalline solids. Matter 3 , 197–211 (2020).

Wuest, T., Weimer, D., Irgens, C. & Thoben, K. D. Machine learning in manufacturing: advantages, challenges, and applications. Prod. Manuf. Res. 4 , 23–45 (2016).

De Jong, P., Sánchez, A. S., Esquerre, K., Kalid, R. A. & Torres, E. A. Solar and wind energy production in relation to the electricity load curve and hydroelectricity in the northeast region of Brazil. Renew. Sustain. Energy Rev. 23 , 526–535 (2013).

Zolfani, S. H. & Saparauskas, J. New application of SWARA method in prioritizing sustainability assessment indicators of energy system. Eng. Econ. 24 , 408–414 (2013).

Tao, F., Zhang, M., Liu, Y. & Nee, A. Y. C. Digital twin driven prognostics and health management for complex equipment. CIRP Ann. 67 , 169–172 (2018).

Yun, S., Park, J. H. & Kim, W. T. Data-centric middleware based digital twin platform for dependable cyber-physical systems. in International Conference on Ubiquitous and Future Networks (ICUFN) 922–926 (2017).

Boretti, A. Integration of solar thermal and photovoltaic, wind, and battery energy storage through AI in NEOM city. Energy AI 3 , 100038–100045 (2021).

Ghoddusi, H., Creamer, G. G. & Rafizadeh, N. Machine learning in energy economics and finance: a review. Energy Econ. 81 , 709–727 (2019).

Asensio, O. I., Mi, X. & Dharur, S. Using machine learning techniques to aid environmental policy analysis: a teaching case regarding big data and electric vehicle charging infrastructure. Case Stud. Environ. 4 , 961302 (2020).

Zheng, S., Trott, A., Srinivasa, S., Parkes, D. C. & Socher, R. The AI economist: taxation policy design via two-level deep multiagent reinforcement learning. Sci. Adv. 8 , eabk2607 (2022).

Sun, S. et al. A data fusion approach to optimize compositional stability of halide perovskites. Matter 4 , 1305–1322 (2021).

Sun, W. et al. Machine learning — assisted molecular design and efficiency prediction for high-performance organic photovoltaic materials. Sci. Adv. 5 , eaay4275 (2019).

Sun, S. et al. Accelerated development of perovskite-inspired materials via high-throughput synthesis and machine-learning diagnosis. Joule 3 , 1437–1451 (2019).

Kirman, J. et al. Machine-learning-accelerated perovskite crystallization. Matter 2 , 938–947 (2020).

Langner, S. et al. Beyond ternary OPV: high-throughput experimentation and self-driving laboratories optimize multicomponent systems. Adv. Mater. 32 , 1907801 (2020).

Hartono, N. T. P. et al. How machine learning can help select capping layers to suppress perovskite degradation. Nat. Commun. 11 , 4172 (2020).

Odabaşı, Ç. & Yıldırım, R. Performance analysis of perovskite solar cells in 2013–2018 using machine-learning tools. Nano Energy 56 , 770–791 (2019).

Fenning, D. P. et al. Darwin at high temperature: advancing solar cell material design using defect kinetics simulations and evolutionary optimization. Adv. Energy Mater. 4 , 1400459 (2014).

Allam, O., Cho, B. W., Kim, K. C. & Jang, S. S. Application of DFT-based machine learning for developing molecular electrode materials in Li-ion batteries. RSC Adv. 8 , 39414–39420 (2018).

Okamoto, Y. & Kubo, Y. Ab initio calculations of the redox potentials of additives for lithium-ion batteries and their prediction through machine learning. ACS Omega 3 , 7868–7874 (2018).

Takagishi, Y., Yamanaka, T. & Yamaue, T. Machine learning approaches for designing mesoscale structure of Li-ion battery electrodes. Batteries 5 , 54 (2019).

Tan, Y., Liu, W. & Qiu, Q. Adaptive power management using reinforcement learning. in IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers (ICCAD ) 461–467 (IEEE, 2009).

Ermon, S., Xue, Y., Gomes, C. & Selman, B. Learning policies for battery usage optimization in electric vehicles. Mach. Learn. 92 , 177–194 (2013).

Schmidt, J., Marques, M. R. G., Botti, S. & Marques, M. A. L. Recent advances and applications of machine learning in solid-state materials science. npj Comput. Mater. 5 , 83 (2019).

Pilania, G. Machine learning in materials science: from explainable predictions to autonomous design. Comput. Mater. Sci. 193 , 110360 (2021).

Butler, K. T., Davies, D. W., Cartwright, H., Isayev, O. & Walsh, A. Machine learning for molecular and materials science. Nature 559 , 547–555 (2018).

Kaufmann, K. et al. Crystal symmetry determination in electron diffraction using machine learning. Science 367 , 564–568 (2020).

Chen, C., Zuo, Y., Ye, W., Li, X. & Ong, S. P. Learning properties of ordered and disordered materials from multi-fidelity data. Nat. Comput. Sci. 1 , 46–53 (2021).

Liu, M., Yan, K., Oztekin, B. & Ji, S. GraphEBM: molecular graph generation with energy-based models. Preprint at https://doi.org/10.48550/arXiv.2102.00546 (2021).

Segler, M. H. S., Preuss, M. & Waller, M. P. Planning chemical syntheses with deep neural networks and symbolic AI. Nature 555 , 604–610 (2018).

Granda, J. M., Donina, L., Dragone, V., Long, D.-L. & Cronin, L. Controlling an organic synthesis robot with machine learning to search for new reactivity. Nature 559 , 377–381 (2018).

Epps, R. W. et al. Artificial chemist: an autonomous quantum dot synthesis bot. Adv. Mater. 32 , 2001626 (2020).

MacLeod, B. P. et al. Self-driving laboratory for accelerated discovery of thin-film materials. Sci. Adv. 6 , eaaz8867 (2020).

Li, Z. et al. Robot-accelerated perovskite investigation and discovery. Chem. Mater. 32 , 5650–5663 (2020).

Aguiar, J. A., Gong, M. L., Unocic, R. R., Tasdizen, T. & Miller, B. D. Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning. Sci. Adv. 5 , eaaw1949 (2019).

Download references

Acknowledgements

Z.Y. and A.A.-G. were supported as part of the Nanoporous Materials Genome Center by the US Department of Energy, Office of Science, Office of Basic Energy Sciences under award number DE-FG02-17ER16362 and the US Department of Energy, Office of Science — Chicago under award number DE-SC0019300. A.J. was financially supported by Huawei Technologies Canada and the Natural Sciences and Engineering Research Council (NSERC). L.M.M.-M. thanks the support of the Defense Advanced Research Projects Agency under the Accelerated Molecular Discovery Program under cooperative agreement number HR00111920027 dated 1 August 2019. Y.W. acknowledges funding support from the Singapore National Research Foundation under its Green Buildings Innovation Cluster (GBIC award number NRF2015ENC-GBICRD001-012) administered by the Building and Construction Authority, its Green Data Centre Research (GDCR award number NRF2015ENC-GDCR01001-003) administered by the Info-communications Media Development Authority, and its Energy Programme (EP award number NRF2017EWT-EP003-023) administered by the Energy Market Authority of Singapore. A.A.-G. is a Canadian Institute for Advanced Research (CIFAR) Lebovic Fellow. E.H.S. acknowledges funding by the Ontario Ministry of Colleges and Universities (grant ORF-RE08-034), the Natural Sciences and Engineering Research Council (NSERC) of Canada (grant RGPIN-2017-06477), the Canadian Institute for Advanced Research (CIFAR) (grant FS20-154 APPT.2378) and the University of Toronto Connaught Fund (grant GC 2012-13). Z.W.S. acknowledges funding by the Singapore National Research Foundation (NRF-NRFF2017-04).

Author information

These authors contributed equally: Zhenpeng Yao, Yanwei Lum, Andrew Johnston.

Authors and Affiliations

Shanghai Key Laboratory of Hydrogen Science & Center of Hydrogen Science, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, China

Zhenpeng Yao

Chemical Physics Theory Group, Department of Chemistry and Department of Computer Science, University of Toronto, Toronto, Ontario, Canada

Zhenpeng Yao, Luis Martin Mejia-Mendoza & Alán Aspuru-Guzik

Innovation Center for Future Materials, Zhangjiang Institute for Advanced Study, Shanghai Jiao Tong University, Shanghai, China

State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, China

Institute of Materials Research and Engineering, Agency for Science, Technology and Research (A*STAR), Innovis, Singapore, Singapore

Yanwei Lum & Zhi Wei Seh

Department of Electrical and Computer Engineering, University of Toronto, Toronto, Ontario, Canada

Yanwei Lum, Andrew Johnston & Edward H. Sargent

School of Computer Science and Engineering, Nanyang Technological University, Singapore, Singapore

Xin Zhou & Yonggang Wen

Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada

Alán Aspuru-Guzik

You can also search for this author in PubMed   Google Scholar

Contributions

Z.Y., Y.L. and A.J. contributed equally to this work. All authors contributed to the writing and editing of the manuscript.

Corresponding authors

Correspondence to Alán Aspuru-Guzik , Edward H. Sargent or Zhi Wei Seh .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature Reviews Materials thanks Shijing Sun and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Related links

MatD3 database: https://github.com/HybriD3-database/MatD3

Materials Project: https://materialsproject.org/

Machine learning techniques that can query a user interactively to modify its current strategy (that is, label an input).

(AI). Theory and development of computer systems that exhibit intelligence.

A system for adjusting the power output of multiple generators at different power plants, in response to changes in the load.

A technology development pipeline that incorporates automation to go from idea to realization of technology. ‘Closed’ refers to the concept that the system improves with experience and iterations.

Process of increasing the amount of data through adding slightly modified copies or newly created synthetic data from existing data.

A generative graphical model, or alternatively a class of deep neural network, composed of multiple layers of latent variables, with connections between the layers but not between units within each layer.

(DL). Machine learning subfield that is based on neural networks with representation learning.

The ability to adapt to new, unseen data, drawn from the same distribution as the one used to create the model.

Machine learning techniques that learn to model the data distribution of a dataset and sample new data points.

Degree to which a human can understand a model’s decision. Interpretability can be used to build trust and credibility.

A design method where new materials and compounds are ‘reverse-engineered’ simply by inputting a set of desired properties and characteristics and then using an optimization algorithm to generate a predicted solution.

A special kind of recurrent neural networks that are capable of selectively remembering patterns for a long duration of time.

(ML). Field within artificial intelligence that deals with learning algorithms, which improve automatically through experience (data).

A computerized system composed of multiple interacting intelligent agents.

The combination of ridge regression (a method of estimating the coefficients of multiple-regression models in scenarios where the independent variables are highly correlated) with multiple kernel techniques.

Models that involve the analysis of multiple, simultaneous physical phenomena, which can include heat transfer, fluid flow, deformation, electromagnetics, acoustics and mass transport.

The field of solving problems that have important features at multiple scales of time and/or space.

A neural network is composed of parameterized and optimizable transformations.

A class of artificial neural networks where connections between nodes form a directed or undirected graph along a temporal sequence.

Process of incorporating additional information into the model to constrain its solution space.

Machine learning techniques that make a sequence of decisions to maximize a reward.

Features used in a representation learning model, which transforms inputs into new features for a task.

Technique for solving problems in the planning of chemical synthesis.

A robotic equipment automated chemical synthesis plan.

Design process composed of several stages where materials are iteratively filtered and ranked to arrive to a few top candidates.

Machine learning techniques that involve the usage of labelled data.

Machine learning techniques that adapt a learned representation or strategy from one dataset to another.

Process of evaluating the statistical confidence of model.

Machine learning techniques that learn patterns from unlabelled data.

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Cite this article.

Yao, Z., Lum, Y., Johnston, A. et al. Machine learning for a sustainable energy future. Nat Rev Mater 8 , 202–215 (2023). https://doi.org/10.1038/s41578-022-00490-5

Download citation

Accepted : 14 September 2022

Published : 18 October 2022

Issue Date : March 2023

DOI : https://doi.org/10.1038/s41578-022-00490-5

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Artificial intelligence-based methods for renewable power system operation.

• Yuanzheng Li
• Yizhou Ding
• Zhigang Zeng

Nature Reviews Electrical Engineering (2024)

Nature-inspired interfacial engineering for energy harvesting

• Baoping Zhang
• Wanghuai Xu
• Zuankai Wang

Non-destructive characterization techniques for battery performance and life-cycle assessment

• Charlotte Gervillié-Mouravieff
• Wurigumula Bao
• Ying Shirley Meng

Attention towards chemistry agnostic and explainable battery lifetime prediction

• Fuzhan Rahmanian
• Robert M. Lee
• Helge Sören Stein

npj Computational Materials (2024)

Bringing digital synthesis to Mars

• Mark D. Symes
• Leroy Cronin

Nature Synthesis (2024)

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

Renewable Energy

Renewable energy sources are growing quickly and will play a vital role in tackling climate change..

Since the Industrial Revolution, the energy mix of most countries across the world has become dominated by fossil fuels. This has major implications for the global climate, as well as for human health. Three-quarters of global greenhouse gas emissions result from the burning of fossil fuels for energy. Fossil fuels are responsible for large amounts of local air pollution – a health problem that leads to at least 5 million premature deaths each year.

To reduce CO 2 emissions and local air pollution, the world needs to rapidly shift towards low-carbon sources of energy – nuclear and renewable technologies.

Renewable energy will play a key role in decarbonizing our energy systems in the coming decades. But how rapidly is our production of renewable energy changing? What technologies look most promising in transforming our energy mix?

In this article we look at the data on renewable energy technologies across the world; what share of energy they account for today, and how quickly this is changing.

Renewable energy generation

How much of our primary energy comes from renewables.

We often hear about the rapid growth of renewable technologies in media reports. But how much of an impact has this growth had on our energy systems?

In this interactive chart, we see the share of primary energy consumption that came from renewable technologies – the combination of hydropower, solar, wind, geothermal, wave, tidal, and modern biofuels. Traditional biomass – which can be an important energy source in lower-income settings is not included.

Note that this data is based on primary energy calculated by the 'substitution method' which attempts to correct for the inefficiencies in fossil fuel production. It does this by converting non-fossil fuel sources to their 'input equivalents': the amount of primary energy that would be required to produce the same amount of energy if it came from fossil fuels.

Approximately one-seventh of the world's primary energy is now sourced from renewable technologies.

Note that this is based on renewable energy's share in the energy mix. Energy consumption represents the sum of electricity, transport, and heating. We look at the electricity mix later in this article.

Breakdown of renewables in the energy mix

In the section above we looked at what share renewable technologies collectively accounted for in the energy mix.

In the charts shown here, we look at the breakdown of renewable technologies by their components – hydropower, solar, wind, and others.

The first chart shows this as a stacked area chart, which allows us to more readily see the breakdown of the renewable mix and the relative contribution of each. The second chart is shown as a line chart, allowing us to see more clearly how each source is changing over time.

Globally we see that hydropower is by far the largest modern renewable source. However, we also see wind and solar power both growing rapidly.

Renewables in the electricity mix

How much of our electricity comes from renewables.

In the sections above we looked at the role of renewables in the total energy mix . This includes not only electricity but also transport and heating. Electricity forms only one component of energy consumption.

Since transport and heating tend to be harder to decarbonize – they are more reliant on oil and gas – renewables tend to have a higher share in the electricity mix versus the total energy mix.

This interactive chart shows the share of electricity that comes from renewable technologies.

Globally, almost one-third of our electricity comes from renewables.

Hydropower generation

Hydroelectric power has been one of our oldest and largest sources of low-carbon energy. Hydroelectric generation at scale dates back more than a century, and is still our largest renewable source – excluding traditional biomass, it still accounts for approximately half of renewable generation.

However, the scale of hydroelectric power generation varies significantly across the world. This interactive chart shows its contribution by country.

Share of primary energy that comes from hydropower

This interactive chart shows the share of primary energy that comes from hydropower.

Share of electricity that comes from hydropower

This interactive chart shows the share of electricity that comes from hydropower.

Wind energy

Wind energy generation.

This interactive chart shows the amount of energy generated from wind each year. This includes both onshore and offshore wind farms.

Wind generation at scale – compared to hydropower, for example – is a relatively modern renewable energy source but is growing quickly in many countries across the world.

Installed wind capacity

The previous section looked at the energy output from wind farms across the world. Energy output is a function of power (installed capacity) multiplied by the time of generation.

Energy generation is therefore a function of how much wind capacity is installed. This interactive chart shows installed wind capacity – including both onshore and offshore – across the world.

Share of primary energy that comes from wind

This interactive chart shows the share of primary energy that comes from wind.

Share of electricity that comes from wind

This interactive chart shows the share of electricity that comes from wind.

Solar energy

Solar energy generation.

This interactive chart shows the amount of energy generated from solar power each year.

Solar generation at scale – compared to hydropower, for example – is a relatively modern renewable energy source but is growing quickly in many countries across the world.

Installed solar capacity

The previous section looked at the energy output from solar across the world. Energy output is a function of power (installed capacity) multiplied by the time of generation.

Energy generation is therefore a function of how much solar capacity is installed. This interactive chart shows installed solar capacity across the world.

Share of primary energy that comes from solar

This interactive chart shows the share of primary energy that comes from solar power.

Share of electricity that comes from solar

This interactive chart shows the share of electricity that comes from solar power.

Biofuel production

Traditional biomass – the burning of charcoal, organic wastes, and crop residues – was an important energy source for a long period of human history. It remains an important source in lower-income settings today. However, high-quality estimates of energy consumption from these sources are difficult to find. The Energy Institute Statistical Review of World Energy – our main data source on energy – only publishes data on commercially traded energy, so traditional biomass is not included.

However, modern biofuels are included in this energy data. Bioethanol and biodiesel – fuel made from crops such as corn, sugarcane, hemp, and cassava – are now a key transport fuel in many countries.

This interactive chart shows modern biofuel production across the world.

Installed geothermal capacity

This interactive chart shows the installed capacity of geothermal energy across the world.

Cite this work

Our articles and data visualizations rely on work from many different people and organizations. When citing this article, please also cite the underlying data sources. This article can be cited as:

BibTeX citation

Reuse this work freely

All visualizations, data, and code produced by Our World in Data are completely open access under the Creative Commons BY license . You have the permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited.

The data produced by third parties and made available by Our World in Data is subject to the license terms from the original third-party authors. We will always indicate the original source of the data in our documentation, so you should always check the license of any such third-party data before use and redistribution.

All of our charts can be embedded in any site.

Our World in Data is free and accessible for everyone.

Help us do this work by making a donation.

Renewable and Alternative Energy

Latest Nelson Issue Brief focuses on the benefits, barriers, and considerations of adopting alternative energy sources.

Imagine a future where our energy comes from the sun, wind, and flowing water, rather than from burning fossil fuels. Fortunately, that vision isn’t too far off; research continues to show the environmental, economic, societal, and health benefits that will come from shifting our energy systems to renewable sources like solar, wind, hydropower, nuclear, and biofuels.

However, the path to widespread adoption of alternative energy isn’t without its challenges. We need vast mineral resources, answers to tough land-use questions, and solutions to creating better energy facilities and storage systems.

The Nelson Institute Issue Brief is a quarterly publication that shares new research from across the UW–Madison campus on key issues of environmental concern. This edition focuses on renewable and alternative energy in Wisconsin and beyond. Learn how we can use satellite data to support energy applications, engage with the community to address local energy needs, and better understand the economic, health and societal benefits of grid decarbonization.

Featured in This Issue

• Satellite Data for Renewable Energy : UW–Madison’s Tracey Holloway, Michael Kamp, Brad Pierce, and Morgan Edwards (and Paul Stackhouse, NASA Langley Research Center)
• What are Wisconsinites’ Perspectives on Community Energy Management and the Clean Energy Transition? Mikhaila Calice
• A Holistic Framework of the Benefits of Utility-Scale Solar : Katie Mummah, Sofia Taylor, Manman Ding, and Lew Blank (and Cristina Crespo Montañés, UC–Berkeley; Eric O’Shaughnessy, Lawrence Berkeley National Laboratory)
• Health Benefits for Electricity Decarbonization Pathways in Wisconsin : Nova Tebbe, Jonathan Patz, Michael Ferris

Read this issue as a PDF or browse past editions

Renewable Energy

Renewable energy comes from sources that will not be used up in our lifetimes, such as the sun and wind.

Earth Science, Experiential Learning, Engineering, Geology

Wind Turbines in a Sheep Pasture

Wind turbines use the power of wind to generate energy. This is just one source of renewable energy.

Photograph by Jesus Keller/ Shutterstock

The wind, the sun, and Earth are sources of  renewable energy . These energy sources naturally renew, or replenish themselves.

Wind, sunlight, and the planet have energy that transforms in ways we can see and feel. We can see and feel evidence of the transfer of energy from the sun to Earth in the sunlight shining on the ground and the warmth we feel when sunlight shines on our skin. We can see and feel evidence of the transfer of energy in wind’s ability to pull kites higher into the sky and shake the leaves on trees. We can see and feel evidence of the transfer of energy in the geothermal energy of steam vents and geysers .

People have created different ways to capture the energy from these renewable sources.

Solar Energy

Solar energy can be captured “actively” or “passively.”

Active solar energy uses special technology to capture the sun’s rays. The two main types of equipment are photovoltaic cells (also called PV cells or solar cells) and mirrors that focus sunlight in a specific spot. These active solar technologies use sunlight to generate electricity , which we use to power lights, heating systems, computers, and televisions.

Passive solar energy does not use any equipment. Instead, it gets energy from the way sunlight naturally changes throughout the day. For example, people can build houses so their windows face the path of the sun. This means the house will get more heat from the sun. It will take less energy from other sources to heat the house.

Other examples of passive solar technology are green roofs , cool roofs, and radiant barriers . Green roofs are completely covered with plants. Plants can get rid of pollutants in rainwater and air. They help make the local environment cleaner.

Cool roofs are painted white to better reflect sunlight. Radiant barriers are made of a reflective covering, such as aluminum. They both reflect the sun’s heat instead of absorbing it. All these types of roofs help lower the amount of energy needed to cool the building.

Advantages and Disadvantages There are many advantages to using solar energy. PV cells last for a long time, about 20 years.

However, there are reasons why solar power cannot be used as the only power source in a community. It can be expensive to install PV cells or build a building using passive solar technology.

Sunshine can also be hard to predict. It can be blocked by clouds, and the sun doesn’t shine at night. Different parts of Earth receive different amounts of sunlight based on location, the time of year, and the time of day.

Wind Energy

People have been harnessing the wind’s energy for a long, long time. Five-thousand years ago, ancient Egyptians made boats powered by the wind. In 200 B.C.E., people used windmills to grind grain in the Middle East and pump water in China.

Today, we capture the wind’s energy with wind turbines . A turbine is similar to a windmill; it has a very tall tower with two or three propeller-like blades at the top. These blades are turned by the wind. The blades turn a generator (located inside the tower), which creates electricity.

Groups of wind turbines are known as wind farms . Wind farms can be found near farmland, in narrow mountain passes, and even in the ocean, where there are steadier and stronger winds. Wind turbines anchored in the ocean are called “ offshore wind farms.”

Wind farms create electricity for nearby homes, schools, and other buildings.

Advantages and Disadvantages Wind energy can be very efficient . In places like the Midwest in the United States and along coasts, steady winds can provide cheap, reliable electricity.

Another great advantage of wind power is that it is a “clean” form of energy. Wind turbines do not burn fuel or emit any pollutants into the air.

Wind is not always a steady source of energy, however. Wind speed changes constantly, depending on the time of day, weather , and geographic location. Currently, it cannot be used to provide electricity for all our power needs.

Wind turbines can also be dangerous for bats and birds. These animals cannot always judge how fast the blades are moving and crash into them.

Geothermal Energy

Deep beneath the surface is Earth’s core . The center of Earth is extremely hot—thought to be over 6,000 °C (about 10,800 °F). The heat is constantly moving toward the surface.

We can see some of Earth’s heat when it bubbles to the surface. Geothermal energy can melt underground rocks into magma and cause the magma to bubble to the surface as lava . Geothermal energy can also heat underground sources of water and force it to spew out from the surface. This stream of water is called a geyser.

However, most of Earth’s heat stays underground and makes its way out very, very slowly.

We can access underground geothermal heat in different ways. One way of using geothermal energy is with “geothermal heat pumps.” A pipe of water loops between a building and holes dug deep underground. The water is warmed by the geothermal energy underground and brings the warmth aboveground to the building. Geothermal heat pumps can be used to heat houses, sidewalks, and even parking lots.

Another way to use geothermal energy is with steam. In some areas of the world, there is underground steam that naturally rises to the surface. The steam can be piped straight to a power plant. However, in other parts of the world, the ground is dry. Water must be injected underground to create steam. When the steam comes to the surface, it is used to turn a generator and create electricity.

In Iceland, there are large reservoirs of underground water. Almost 90 percent of people in Iceland use geothermal as an energy source to heat their homes and businesses.

Advantages and Disadvantages An advantage of geothermal energy is that it is clean. It does not require any fuel or emit any harmful pollutants into the air.

Geothermal energy is only avaiable in certain parts of the world. Another disadvantage of using geothermal energy is that in areas of the world where there is only dry heat underground, large quantities of freshwater are used to make steam. There may not be a lot of freshwater. People need water for drinking, cooking, and bathing.

Biomass Energy

Biomass is any material that comes from plants or microorganisms that were recently living. Plants create energy from the sun through photosynthesis . This energy is stored in the plants even after they die.

Trees, branches, scraps of bark, and recycled paper are common sources of biomass energy. Manure, garbage, and crops , such as corn, soy, and sugar cane, can also be used as biomass feedstocks .

We get energy from biomass by burning it. Wood chips, manure, and garbage are dried out and compressed into squares called “briquettes.” These briquettes are so dry that they do not absorb water. They can be stored and burned to create heat or generate electricity.

Biomass can also be converted into biofuel . Biofuels are mixed with regular gasoline and can be used to power cars and trucks. Biofuels release less harmful pollutants than pure gasoline.

Advantages and Disadvantages A major advantage of biomass is that it can be stored and then used when it is needed.

Growing crops for biofuels, however, requires large amounts of land and pesticides . Land could be used for food instead of biofuels. Some pesticides could pollute the air and water.

Biomass energy can also be a nonrenewable energy source. Biomass energy relies on biomass feedstocks—plants that are processed and burned to create electricity. Biomass feedstocks can include crops, such as corn or soy, as well as wood. If people do not replant biomass feedstocks as fast as they use them, biomass energy becomes a non-renewable energy source.

Hydroelectric Energy

Hydroelectric energy is made by flowing water. Most hydroelectric power plants are located on large dams , which control the flow of a river.

Dams block the river and create an artificial lake, or reservoir. A controlled amount of water is forced through tunnels in the dam. As water flows through the tunnels, it turns huge turbines and generates electricity.

Advantages and Disadvantages Hydroelectric energy is fairly inexpensive to harness. Dams do not need to be complex, and the resources to build them are not difficult to obtain. Rivers flow all over the world, so the energy source is available to millions of people.

Hydroelectric energy is also fairly reliable. Engineers control the flow of water through the dam, so the flow does not depend on the weather (the way solar and wind energies do).

However, hydroelectric power plants are damaging to the environment. When a river is dammed, it creates a large lake behind the dam. This lake (sometimes called a reservoir) drowns the original river habitat deep underwater. Sometimes, people build dams that can drown entire towns underwater. The people who live in the town or village must move to a new area.

Hydroelectric power plants don’t work for a very long time: Some can only supply power for 20 or 30 years. Silt , or dirt from a riverbed, builds up behind the dam and slows the flow of water.

Other Renewable Energy Sources

Scientists and engineers are constantly working to harness other renewable energy sources. Three of the most promising are tidal energy , wave energy , and algal (or algae) fuel.

Tidal energy harnesses the power of ocean tides to generate electricity. Some tidal energy projects use the moving tides to turn the blades of a turbine. Other projects use small dams to continually fill reservoirs at high tide and slowly release the water (and turn turbines) at low tide.

Wave energy harnesses waves from the ocean, lakes, or rivers. Some wave energy projects use the same equipment that tidal energy projects do—dams and standing turbines. Other wave energy projects float directly on waves. The water’s constant movement over and through these floating pieces of equipment turns turbines and creates electricity.

Algal fuel is a type of biomass energy that uses the unique chemicals in seaweed to create a clean and renewable biofuel. Algal fuel does not need the acres of cropland that other biofuel feedstocks do.

Renewable Nations

These nations (or groups of nations) produce the most energy using renewable resources. Many of them are also the leading producers of nonrenewable energy: China, European Union, United States, Brazil, and Canada

Articles & Profiles

Media credits.

The audio, illustrations, photos, and videos are credited beneath the media asset, except for promotional images, which generally link to another page that contains the media credit. The Rights Holder for media is the person or group credited.

Last Updated

June 21, 2024

User Permissions

For information on user permissions, please read our Terms of Service. If you have questions about how to cite anything on our website in your project or classroom presentation, please contact your teacher. They will best know the preferred format. When you reach out to them, you will need the page title, URL, and the date you accessed the resource.

If a media asset is downloadable, a download button appears in the corner of the media viewer. If no button appears, you cannot download or save the media.

Text on this page is printable and can be used according to our Terms of Service .

Interactives

Any interactives on this page can only be played while you are visiting our website. You cannot download interactives.

Related Resources

• History & Society
• Science & Tech
• Biographies
• Animals & Nature
• Geography & Travel
• Arts & Culture
• Games & Quizzes
• On This Day
• One Good Fact
• New Articles
• Lifestyles & Social Issues
• Philosophy & Religion
• Politics, Law & Government
• World History
• Health & Medicine
• Browse Biographies
• Birds, Reptiles & Other Vertebrates
• Bugs, Mollusks & Other Invertebrates
• Environment
• Fossils & Geologic Time
• Entertainment & Pop Culture
• Sports & Recreation
• Visual Arts
• Demystified
• Image Galleries
• Infographics
• Top Questions
• Britannica Kids
• Saving Earth
• Space Next 50
• Student Center

• What is solar energy?
• How is solar energy collected?

renewable energy

Our editors will review what you’ve submitted and determine whether to revise the article.

• U.S. Energy Information Administration - Energy Kids - Energy Sources - Renewable
• Natural Resources Defense Council - Renewable Energy: The Clean Facts
• Energy.gov - Renewable Energy
• United Nations - What is renewable energy?
• World Nuclear Association - Renewable Energy and Electricity
• alternative energy - Children's Encyclopedia (Ages 8-11)
• alternative energy - Student Encyclopedia (Ages 11 and up)

Recent News

renewable energy , usable energy derived from replenishable sources such as the Sun ( solar energy ), wind ( wind power ), rivers ( hydroelectric power ), hot springs ( geothermal energy ), tides ( tidal power ), and biomass ( biofuels ).

At the beginning of the 21st century, about 80 percent of the world’s energy supply was derived from fossil fuels such as coal , petroleum , and natural gas . Fossil fuels are finite resources; most estimates suggest that the proven reserves of oil are large enough to meet global demand at least until the middle of the 21st century. Fossil fuel combustion has a number of negative environmental consequences. Fossil-fueled power plants emit air pollutants such as sulfur dioxide , particulate matter , nitrogen oxides, and toxic chemicals (heavy metals: mercury , chromium , and arsenic ), and mobile sources, such as fossil-fueled vehicles, emit nitrogen oxides, carbon monoxide , and particulate matter. Exposure to these pollutants can cause heart disease , asthma , and other human health problems. In addition, emissions from fossil fuel combustion are responsible for acid rain , which has led to the acidification of many lakes and consequent damage to aquatic life, leaf damage in many forests, and the production of smog in or near many urban areas. Furthermore, the burning of fossil fuels releases carbon dioxide (CO 2 ), one of the main greenhouse gases that cause global warming .

In contrast, renewable energy sources accounted for nearly 20 percent of global energy consumption at the beginning of the 21st century, largely from traditional uses of biomass such as wood for heating and cooking . By 2015 about 16 percent of the world’s total electricity came from large hydroelectric power plants, whereas other types of renewable energy (such as solar, wind, and geothermal) accounted for 6 percent of total electricity generation. Some energy analysts consider nuclear power to be a form of renewable energy because of its low carbon emissions; nuclear power generated 10.6 percent of the world’s electricity in 2015.

Growth in wind power exceeded 20 percent and photovoltaics grew at 30 percent annually in the 1990s, and renewable energy technologies continued to expand throughout the early 21st century. Between 2001 and 2017 world total installed wind power capacity increased by a factor of 22, growing from 23,900 to 539,581 megawatts. Photovoltaic capacity also expanded, increasing by 50 percent in 2016 alone. The European Union (EU), which produced an estimated 6.38 percent of its energy from renewable sources in 2005, adopted a goal in 2007 to raise that figure to 20 percent by 2020. By 2016 some 17 percent of the EU’s energy came from renewable sources. The goal also included plans to cut emissions of carbon dioxide by 20 percent and to ensure that 10 percent of all fuel consumption comes from biofuels . The EU was well on its way to achieving those targets by 2017. Between 1990 and 2016 the countries of the EU reduced carbon emissions by 23 percent and increased biofuel production to 5.5 percent of all fuels consumed in the region. In the United States numerous states have responded to concerns over climate change and reliance on imported fossil fuels by setting goals to increase renewable energy over time. For example, California required its major utility companies to produce 20 percent of their electricity from renewable sources by 2010, and by the end of that year California utilities were within 1 percent of the goal. In 2008 California increased this requirement to 33 percent by 2020, and in 2017 the state further increased its renewable-use target to 50 percent by 2030.