A Review and Analysis of Forecasting of Photovoltaic Power Generation Using Machine Learning

Abstract

Solar technologies are currently in demand due to the clean energy source. It is the most abundant renewable source of energy and can be used in a variety of ways including electricity generation. Suitable photovoltaic maintenance management plans are required to ensure proper levels of reliability minimizing the operating costs. For this reason, it is crucial to assess and forecast the performance of a photovoltaic power system. This paper provides a methodological review for the forecasting of photovoltaic power generation using Machine Learning. Different machine learning algorithms such as support vector machine, logistic regression, decision trees, random forest, convolutional neural networks, multilayer perceptron, etc. have been covered for detailed review and analysis. This paper explores the opportunities and dimensions through the analysis of the photovoltaic power generation computational framework. The assessment and analysis were performed based on different parameters, covering the parametric evaluation of learning approaches. It also covers the new challenges and prospective solutions with detailed discussion and analysis. Finally, future suggestions have been provided.