Forecasting of Solar Photovoltaic Power Output using machine learning

Abstract

A precise estimation of solar energy is necessary for a higher integration of renewable energy for better functioning of the current power system. Data-driven algorithms may be used to enhance solar energy estimates as a result of the availability of data at previously unheard-of granularities. The Linear Regression algorithm is provided in this work as the core model for the enhanced, broadly applicable stackable ensemble method. It is used to aggregate the outcomes from the core models, improving forecasting accuracy for solar PV power. In this paper comparison of two power plants has been done and it is found that one of them have better conversion accuracy than the other.