Short-Term  Photovoltaic (PV) Energy Prediction Using Machine Learning Approach

The efficient prediction of short-term photovoltaic (PV) energy output is a pressing challenge in the renewable energy sector. Accurate PV energy forecasts are pivotal for optimizing grid integration, minimizing energy wastage, and reducing operational costs in solar power plants. This study addresses these challenges by leveraging machine learning (ML) techniques and comparing the performance of three ML models, namely linear regression, random forest, and gradient boosting, with the objective of identifying the most effective model for short-term PV energy prediction during the given timeframe. The study utilizes 5 MWp PV power plant data collected in Melaka, Malaysia, over a daily period from 1st September 2013 until 31st January 2014 providing a robust dataset for training and testing the models. The primary evaluation metrics used in this analysis are the root mean squared error (RMSE), R-squared (R2) score, and the mean absolute percentage error (MAPE). The findings reveal that the gradient boosting (GB) model outperforms both linear regression (LR) and random forest (RF) in terms of predictive accuracy; RMSE (1380.13), R-squared (R2) (0.8), and MAPE (4.3%). This suggests that GB is the most suitable ML model for accurate short-term PV energy prediction in the context of Melaka’s PV power plant data.