Solar PV Power Generation Prediction Using Machine Learning Algorithms at Different Tilt Angles

One of the main contributors to the warming of the planet is the carbon dioxide that these fossil fuels release into the atmosphere. To tackle this worrying problem, the country should use easily accessible renewable energy sources including solar, wind, and hydropower. We can restore ecological harmony and mitigate climate change by moving to renewable energy sources. Photovoltaic (PV) systems that are connected to the grid rely on accurate solar power forecasts, which become even more important in the face of environmental variations. Predicting the power output is essential for scheduling, grid energy management, and safe grid operation. Machine learning approaches have recently been effectively utilized in predicting the power output of the systems due to their robust capabilities in non-linear regression. This study examines the prediction of power generation for specific location Noida, Uttar Pradesh India at different Tilt angle 5,15,30 and 45 degree of PV Module from a 10kWp solar PV system using four prominent machine learning techniques for different tilt angle are Neural Networks (NN), Support Vector Machine (SVM), Ensemble method (EN) and Linear Regression (LR). The analysis is based on historical data obtained from a publicly accessible NREL (US National Renewable Energy Laboratory) renewable energy repository and PV power utilized to generate the Power output at different tilt angle. In order to accurately anticipate solar power generation, the study evaluates many performance indicators, including R-Squared, Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). It then finds the best machine learning approach and tilt angle.