| Literature DB >> 35715677 |
Thompson Stephan1, Fadi Al-Turjman2, Monica Ravishankar3, Punitha Stephan4.
Abstract
India is severely affected by the COVID-19 pandemic and is facing an unprecedented public health emergency. While the country's immediate measures focus on combating the coronavirus spread, it is important to investigate the impacts of the current crisis on India's renewable energy transition and air quality. India's economic slowdown is mainly compounded by the collapse of global oil prices and the erosion of global energy demand. A clean energy transition is a key step in enabling the integration of energy and climate. Millions in India are affected owing to fossil fuel pollution and the increasing climate heating that has led to inconceivable health impacts. This paper attempts to study the impact of COVID-19 on India's climate and renewable energy transitions through machine learning algorithms. India is observing a massive collapse in energy demand during the lockdown as its coal generation is suffering the worst part of the ongoing pandemic. During this current COVID-19 crisis, the renewable energy sector benefits from its competitive cost and the Indian government's must-run status to run generators based on renewable energy sources. In contrast to fossil fuel-based power plants, renewable energy sources are not exposed to the same supply chain disruptions in this current pandemic situation. India has the definite potential to surprise the global community and contribute to cost-effective decarbonization. Moreover, the country has a good chance of building more flexibility into the renewable energy sector to avoid an unstable future.Entities:
Keywords: Air quality; COVID-19; Climate change; Energy transition; Machine learning; Renewable energy
Year: 2022 PMID: 35715677 PMCID: PMC9205654 DOI: 10.1007/s11356-022-20997-2
Source DB: PubMed Journal: Environ Sci Pollut Res Int ISSN: 0944-1344 Impact factor: 5.190
Factors related to the energy transition [11]
| Percentage of the global average energy transition index (ETI) score in 2020 | 55.10% |
| Number of countries improving their ETI score since 2015 | 94 |
| Number of countries making steady progress each year since 2015 | 11 |
| Percentage increase in the average ETI score of countries in the top quartile since 2015 | <1% |
| Percentage of global population using as much energy as the remaining 80% | 20% |
| Expected percentage of decline in the coal power generation globally in 2019 | 3% |
| Percentage of young people considering the energy transition speed to be either stagnant or too slow | 70% |
Fig. 1ETI 2020 framework World Economic Forum (2020)
Fig. 2The ETI ranking in 2020 for G20 countries
Fig. 3Year-on-year changes in the power generation during COVID-19 lockdown by considering the timeframe in 2019 and 2020
Fig. 4(a). Comparison of energy requirement between 2019 and 2020 (b). Comparison of energy availability between 2019 and 2020 (c). Comparison of peak power demand between 2019 and 2020 (d). Comparison of peak power demand met between 2019 and 2020 (e). PLF in the country (Coal & Lignite based) from 2009–2010 to 2020–2021
Comparison of energy and power between April 2019 and April 2020 in India
| Region | Drop (%) in April 2020 as compared with April 2019 | |||
|---|---|---|---|---|
| Energy (MU) | Power (MW) | |||
| Requirement | Availability | Peak demand | Peak met | |
| Northern India | 26.16 | 26.56 | 13.71 | 13.83 |
| Western India | 22.02 | 22.02 | 22.26 | 22.24 |
| Southern India | 19.4 | 19.4 | 13.11 | 12.9 |
| Eastern India | 23.9 | 23.67 | 25.97 | 25.85 |
| North Eastern India | 17.48 | 18.28 | 13.1 | 12.48 |
| All India | 22.57 | 22.66 | 24.86 | 24.9 |
Fig. 5(a). Concentration levels of the major air pollutants in 2020 (b). India’s annual fossil fuel emissions
Forecasting off-grid applications
| Year | Street Lighting | Home Light | Solar Lantern | Solar Pumps | Stand Alone Power Plants (kW) |
|---|---|---|---|---|---|
| 2016 | 387632 | 1285877 | 996931 | 61834 | 150001.44 |
| 2017 | 464103 | 1407209 | 996931 | 114878 | 176847.36 |
| 2018 | 605249 | 1647821 | 3074002 | 171228 | 185900.07 |
| 2019 | 659218 | 1715214 | 5823800 | 237120 | 212054.17 |
| 2020 | 15029 | 1721343 | 7529365 | 256156 | 214565 |
| 2021 | 830373 | 1723479 | 7948219 | 286830 | 216407.67 |
| 2022 | 587247 | 1938529 | 10132680 | 353791 | 241842.77 |
Fig. 6(a). Predicted number of off-grid solar applications (b). Predicted Stand Alone Power Plants (kW)
Fig. 7PM2.5 concentration analysis in different parts of Bengaluru City during Phase 1 lockdown due to COVID-19
Climate factors affecting air pollutant concentrations
| Climate Features | PM2.5 | PM10 | NO2 |
|---|---|---|---|
| T: Average Temperature ( | 0.03664602 | 0.03316525 | 0.04767059 |
| TM:Maximum temperature ( | 0.01763994 | 0.03148776 | 0.01931601 |
| Tm: Minimum temperature ( | 0.09971398 | 0.0960497 | 0.21996907 |
| H: Average relative humidity (%) | 0.0800367 | 0.08955058 | 0.09324564 |
| VM: Maximum sustained wind speed (Km/h) | 0.12050093 | 0.10674678 | 0.09099328 |
| VV: Average visibility (Km) | 0.21300068 | 0.22666861 | 0.28990238 |
| V: Average wind speed (Km/h) | 0.43246175 | 0.41633132 | 0.23890303 |
Fig. 8Correlation values among different features
Fig. 9Prediction of most important features
Pollutants prediction test results with regression algorithms
| Models | PM2.5 | PM10 | NO2 | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RMSE | RMSE | RMSE | |||||||||||||
| AdaBoostRegressor (Al-Ash et al. | 0.42 | 0.48 | 24.27 | 0.03 | 74.47 | 0.49 | 0.54 | 13.75 | 0.03 | 45.02 | 0.73 | 0.76 | 3.24 | 0.03 | 8.2 |
| BaggingRegressor (Kadiyala and Kumar | 0.46 | 0.51 | 23.49 | 0.03 | 74.06 | 0.47 | 0.52 | 14.12 | 0.03 | 44.8 | 0.78 | 0.8 | 2.95 | 0.04 | 8.42 |
| BayesianRidge (Yang and Yang | 0.49 | 0.54 | 22.83 | 0.01 | 74.07 | 0.47 | 0.52 | 14.09 | 0.01 | 45.04 | 0.77 | 0.79 | 3.03 | 0.01 | 8.27 |
| DecisionTreeRegressor (Gupta et al. | 0.46 | 0.51 | 23.4 | 0.01 | 74.2 | 0.49 | 0.54 | 13.82 | 0.01 | 45.15 | 0.8 | 0.82 | 2.8 | 0.01 | 8.29 |
| DummyRegressor (Kwan-to | −0.11 | −0.01 | 33.75 | 0.01 | 73.69 | −0.1 | 0 | 20.3 | 0.01 | 44.12 | −0.11 | −0.01 | 6.61 | 0.01 | 7.95 |
| ElasticNet (Hans | 0.45 | 0.51 | 23.62 | 0.01 | 74.89 | 0.45 | 0.5 | 14.4 | 0.01 | 45.12 | 0.62 | 0.66 | 3.84 | 0.01 | 8.34 |
| ElasticNetCV (Qaraad et al. | 0.49 | 0.54 | 22.81 | 0.07 | 74.53 | 0.48 | 0.53 | 13.98 | 0.07 | 45.04 | 0.77 | 0.79 | 3.04 | 0.08 | 8.29 |
| ExtraTreesRegressor (Reza and Haque | 0.46 | 0.51 | 23.4 | 0.11 | 74.20 | 0.49 | 0.54 | 13.82 | 0.11 | 45.14 | 0.8 | 0.82 | 2.8 | 0.11 | 8.28 |
| GammaRegressor (Tao et al. | 0.45 | 0.5 | 23.72 | 0.01 | 74.86 | 0.44 | 0.49 | 14.48 | 0.01 | 45.17 | 0.59 | 0.63 | 4.03 | 0.01 | 8.25 |
| GaussianProcessRegressor | |||||||||||||||
| (Stamenkovic et al. | 0.46 | 0.51 | 23.4 | 0.03 | 74.20 | 0.49 | 0.54 | 13.82 | 0.03 | 45.14 | 0.8 | 0.82 | 2.8 | 0.02 | 8.28 |
| GradientBoostingRegressor (Li et al. | 0.46 | 0.51 | 23.4 | 0.1 | 74.20 | 0.49 | 0.54 | 13.82 | 0.08 | 45.14 | 0.8 | 0.82 | 2.8 | 0.11 | 8.28 |
| HistGradientBoostingRegressor | |||||||||||||||
| (Hirasen et al. | 0.46 | 0.51 | 23.39 | 0.15 | 74.20 | 0.49 | 0.54 | 13.77 | 0.19 | 45.17 | 0.8 | 0.82 | 2.8 | 0.14 | 8.29 |
| HuberRegressor (Malki et al. | 0.58 | 0.62 | 20.67 | 0.03 | 73.01 | 0.58 | 0.62 | 12.56 | 0.03 | 44.46 | 0.78 | 0.8 | 2.94 | 0.03 | 8.21 |
| KernelRidge (Vovk | −4.31 | −3.81 | 73.67 | 0.04 | 73.41 | −4.45 | −3.94 | 45.13 | 0.02 | 44.56 | −0.94 | −0.76 | 8.74 | 0.02 | 8.18 |
| KNeighborsRegressor (Patel et al. | 0.58 | 0.62 | 20.63 | 0.01 | 79.94 | 0.59 | 0.62 | 12.43 | 0.02 | 46.36 | 0.84 | 0.85 | 2.54 | 0.01 | 8.86 |
| Lars (Miche et al. | 0.48 | 0.53 | 23.06 | 0.03 | 73.14 | −4.05 | −3.58 | 43.45 | 0.02 | 45.79 | 0.8 | 0.82 | 2.81 | 0.02 | 8.29 |
| LarsCV (Manjunathan et al. | 0.47 | 0.52 | 23.32 | 0.08 | 74.35 | 0.46 | 0.51 | 14.26 | 0.08 | 44.88 | 0.76 | 0.78 | 3.1 | 0.03 | 8.30 |
| Lasso (Park and Casella | 0.46 | 0.51 | 23.41 | 0.02 | 74.28 | 0.46 | 0.51 | 14.26 | 0.02 | 45.10 | 0.69 | 0.72 | 3.5 | 0.03 | 8.35 |
| LassoCV (Muir and Zhan | 0.54 | 0.58 | 21.75 | 0.15 | 73.86 | 0.55 | 0.59 | 12.95 | 0.15 | 44.90 | 0.78 | 0.8 | 2.97 | 0.11 | 8.28 |
| LassoLars (Usai et al. | 0.06 | 0.15 | 30.99 | 0.01 | 73.58 | −0.1 | 0 | 20.3 | 0.01 | 44.11 | −0.11 | −0.01 | 6.61 | 0.01 | 7.95 |
| LassoLarsCV (Leyva et al. | 0.54 | 0.59 | 21.62 | 0.04 | 73.64 | 0.57 | 0.61 | 12.73 | 0.04 | 44.80 | 0.78 | 0.8 | 2.96 | 0.03 | 8.26 |
| LassoLarsIC (Rao et al. | 0.54 | 0.58 | 21.7 | 0.02 | 73.64 | 0.56 | 0.61 | 12.76 | 0.02 | 44.80 | 0.75 | 0.78 | 3.12 | 0.02 | 8.31 |
| LinearRegression (Montgomery et al. | 0.54 | 0.59 | 21.62 | 0.01 | 73.64 | 0.57 | 0.61 | 12.73 | 0.01 | 44.80 | 0.78 | 0.8 | 2.91 | 0.01 | 8.26 |
| LinearSVR (Ho and Lin | 0.46 | 0.51 | 23.56 | 0.01 | 73.84 | 0.45 | 0.5 | 14.35 | 0.01 | 50.69 | 0.77 | 0.79 | 3.02 | 0.02 | 11.98 |
| MLPRegressor (Gabralla and Abraham | −0.17 | −0.06 | 34.6 | 0.71 | 72.96 | 0.23 | 0.3 | 16.98 | 0.71 | 44.23 | 0.73 | 0.75 | 3.26 | 0.71 | 7.96 |
| NuSVR (Faraone | 0.35 | 0.41 | 25.73 | 0.03 | 71.57 | 0.44 | 0.49 | 14.47 | 0.05 | 44.49 | 0.8 | 0.82 | 2.83 | 0.05 | 7.72 |
| OrthogonalMatchingPursuit (Wang et al. | 0.33 | 0.39 | 26.25 | 0.04 | 73.12 | 0.35 | 0.42 | 15.52 | 0.04 | 44.79 | 0.43 | 0.49 | 4.73 | 0.02 | 8.17 |
| OrthogonalMatchingPursuitCV | |||||||||||||||
| (Kallummil and Kalyani | 0.47 | 0.52 | 23.3 | 0.05 | 73.96 | 0.45 | 0.5 | 14.31 | 0.03 | 44.74 | 0.76 | 0.78 | 3.09 | 0.02 | 8.35 |
| PassiveAggressiveRegressor (Malki et al. | 0.5 | 0.54 | 22.7 | 0.02 | 101.01 | 0.2 | 0.27 | 17.3 | 0.01 | 35.17 | 0.69 | 0.72 | 3.48 | 0.02 | 4.21 |
| PoissonRegressor (Ranzato and Szummer | 0.5 | 0.55 | 22.64 | 0.02 | 73.94 | 0.47 | 0.51 | 14.14 | 0.02 | 45.12 | 0.71 | 0.74 | 3.38 | 0.02 | 8.32 |
| RandomForestRegressor (Graw et al. | 0.47 | 0.52 | 23.36 | 0.19 | 74.12 | 0.49 | 0.53 | 13.86 | 0.18 | 45.12 | 0.8 | 0.82 | 2.82 | 0.18 | 8.28 |
| RANSACRegressor (Aziz et al. | 0.46 | 0.51 | 23.51 | 0.08 | 71.43 | 0.41 | 0.47 | 14.8 | 0.08 | 40.97 | 0.75 | 0.78 | 3.11 | 0.07 | 6.85 |
| Ridge (McDonald | 0.52 | 0.56 | 22.25 | 0.01 | 73.68 | 0.51 | 0.55 | 13.59 | 0.02 | 44.81 | 0.77 | 0.79 | 3.02 | 0.01 | 8.26 |
| RidgeCV (Padhi and Padhy | 0.54 | 0.58 | 21.73 | 0.01 | 73.64 | 0.55 | 0.59 | 12.92 | 0.01 | 44.81 | 0.78 | 0.8 | 2.94 | 0.01 | 8.26 |
| SVR (Abo-Khalil and Lee | 0.37 | 0.43 | 25.37 | 0.02 | 65.29 | 0.47 | 0.52 | 14.1 | 0.02 | 39.79 | 0.79 | 0.81 | 2.86 | 0.02 | 6.45 |
| TransformedTargetRegressor (Brownlee | 0.54 | 0.59 | 21.62 | 0.02 | 73.64 | 0.57 | 0.61 | 12.73 | 0.01 | 44.8 | 0.78 | 0.8 | 2.91 | 0.01 | 8.26 |
| TweedieRegressor (Russo et al. | 0.44 | 0.49 | 24.01 | 0.01 | 75.1 | 0.44 | 0.49 | 14.52 | 0.02 | 45.12 | 0.61 | 0.65 | 3.9 | 0.02 | 8.34 |