Short term forecasting of photovoltaic module using machine learning

Abstract

The objective of this study is to analysis and observe the performance of the photovoltaic (PV) modules in different environmental conditions by applying machine learning algorithm . There were two PV Modules , one is cleaned and other one is dusty . Real-time data from each sensor is effectively collected from November 2019 to February 2020, and prediction has been done on 2 different days from march month of 2020 from the weather station situated in Gabtoli. In this study short term performance analysis has been done with different error calculation. Result shows that, the performance depends on the volume of training dataset. In this study two artificial neural network models has been used to train and test the data of PV module output and assess the short term performance.