Forecasting Specific Yield of Thin Film Solar Panel in Malacca Via Regression Analysis

Abstract

This study employs multiple regression analysis to investigate the relationship between environmental variables and the specific yield of solar PV panels installed in Malacca, Malaysia. The objective of this research is to develop a predictive statistical model for specific yield and to quantify the influence of environmental parameters. Over a period of 365 days data from PV solar panels and several independent variables that are relative humidity, tilt irradiance, global irradiance, temperature and wind speed were collected. The dataset was pre-processed and analysed using descriptive statistics and multiple linear regression. Single variable models and multiple variables models were developed to identify the best-fit model for forecasting specific yield of thin film solar panel. The accuracy of the developed models was evaluated using statistical measures such as the coefficient of determination, Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). The results from error calculation and graphical analysis reveal that the multiple regression model demonstrating a high predictive accuracy as compared to single variable model. The analysis quantifies the relative impact of each variable, providing valuable insights into which factors are most critical for optimizing PV performance in Malacca, Malaysia.