Time Series Prediction of Photovoltaic Energy Using Machine Learning Approach-a Review

Rising worldwide dependency on renewable power requires exact forecast systems emphasizing photovoltaic (PV) energy prediction. Time series prediction remains essential for both power supply management and electricity grid optimization as well as improved resource distribution. The research paper examines PV energy output forecasting using machine learning (ML) technologies while analyzing traditional statistical models and advanced ML algorithm applications. This paper discusses regression-based approaches alongside ensemble learning methods while providing deep learning model analysis of Long-Short-Term Menory Networks(LSTM), Recurrent Neural Networks (RNN), alongside CNN-LSTM hybrid models. This paper conducts a model comparison analysis usingMean Absolute percentage Error(MAPE), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) performance metrics. This investigation creates a strong base for researchers and practitioners to build dependable and extensible ML forecasting models for photovoltaic energy which helps develop sustainable energy solutions.