Statistical Analysis of PV Cell Power Generation and Influence of Weather on Power Generation

Abstract

The need for renewable energy is growing as a result of climate change, global warming and the subsequent increase in natural disasters brought on by the use of carbon-emitting resources. An earlier examination of potential future energy pathways demonstrates that it is theoretically feasible to simultaneously improve energy security, air quality, and access while preventing disastrous climate change. Land, energy, and water are some of our most precious resources, but how and how much we use them also affects the climate. At the national level, renewable energy prospects would be very diverse. The rise in global economic activities results in higher demand in electricity, as its one of the main sources of energy. This leads to search for renewable energy sources. Solar cells are one of the technological innovations that directly convert light energy into electricity through the photovoltaic effect, creating electrical charges that are free to travel through semiconductors. But due to uncertainty in the weather and the influence of weather on power generation makes the integration of PV cell into existing grid a difficult part. To operate the electricity distribution system efficiently one should know both the demand and supply of distribution center. To solve this issue, a statistical model that forecasts the power generation at the PV cell plant and aids in grid operation, as well as information about the impact of weather on power generation, is required. In addition to using traditional time series models fundamental machine learning methods like Linear Regression, Random Forest Model, XG Boost, and Support Vector Regression (SVR) models are also trained and applied for prediction.