Regression analysis on specific yield of thin film solar panel

Azmi, Muhammad ‘Arash (2024) Regression analysis on specific yield of thin film solar panel. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

Text (Full Text) Regression analysis on specific yield of thin film solar panel.pdf - Submitted Version Download (2MB)

Abstract

This project involves conducting a comprehensive regression analysis on the specific yield of thin film solar panels. Utilizing a weather data that represents five independent variables, that are; global irradiance, tilt irradiance, average temperature, average relative humidity, and average windspeed. The study aims to develop a predictive model for the specific yield of thin film solar panels. There are five single regression models and four multiple regression models developed in this study for predicting specific yield. The models' accuracy is measured using mean absolute error (MAE), mean squared error (MSE), and mean percentage error (MPE). The regression model suited the data well, with MAE ranging from 0.56 to 1.01, MSE ranging from 0.71 to 2.13, and MPE ranging from 5.85% to 43.44%. Through regression techniques, the relationships between the independent variables and specific yield are explored, providing valuable insights into the performance of thin film solar technology under varying conditions. The regression model with relative humidity and tilt irradiance is the most accurate multiple regression model for prediction of specific yield by this solar panel at the FTKE, UTeM. The findings of this study are to understand factors influencing specific yield production in thin-film solar panels, contributing to the optimization of their efficiency and overall sustainability.

Actions (login required)

View Item