PV-Power Forecasting using Machine Learning Techniques

Abstract

Solar energy forecasting plays a pivotal role in the efficient utilization of renewable energy resources for sustainable power generation. This study delves into the domain of solar-power forecasting, employing a comprehensive analysis of machine learning models. The primary objective is to evaluate and compare the performance of Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Multi-Layer Perceptron (MLP), and Linear Regression (LR) models in predicting solar energy production. Through a comprehensive evaluation of individual model performance, the study provides nuanced insights into the strengths and limitations of each forecasting approach. Results indicate that the Multy-Layer Perceptron (MLP) model excels in accuracy, exhibiting low root mean square error (RMSE) and high correlation among the parameters. The Gated Recurrent Unit (GRU) model demonstrates competitive performance, while the Recurrent Neural Network model showcases strengths in multiple metrics. Additionally, MLP and GRU models display superior predictive accuracy, emphasizing their efficacy in solar energy forecasting.