A Comparative Analysis of Machine Learning Techniques for Solar Power Forecasting

Sonawane, Karan Manohar (2023) A Comparative Analysis of Machine Learning Techniques for Solar Power Forecasting. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (960kB) | Preview
Preview	PDF (Configuration Manual) Download (186kB) | Preview

Abstract

Solar power generation relies heavily on unpredictable environmental factors, making accurate forecasting essential for grid integration. This analysis evaluates machine learning techniques for forecasting photovoltaic (PV) power output using multivariate weather data from 12 northern hemisphere sites, without solar irradiance data that can have significant measurement errors. The methods examined include XGBoost Regression, Random Forest Regression, LightGBM, and an ensemble Stacking Regressor combining the models. The results demonstrate the Stacking Regressor achieves the best performance with an RMSE of 0.1370 and R-squared of 0.637 on the test set by leveraging the strengths of the component models. The data preprocessing, feature engineering, and ensemble architecture optimization were critical to maximizing accuracy. Overall, the research provides empirical evidence that advanced machine learning, specifically a tailored stacking ensemble approach, can effectively forecast solar PV power utilizing meteorological data as inputs without needing irradiation values that require additional equipment. The framework demonstrates significant promise for integrating solar resources, with opportunities to enhance flexibility across locations through supplementary datasets.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Horn, Christian UNSPECIFIED
Uncontrolled Keywords:	Solar power forecasting; Photovoltaic power prediction Machine Learning; Ensemble Learning
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TD Environmental technology. Sanitary engineering T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electricity Supply Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	23 May 2025 10:48
Last Modified:	23 May 2025 10:48
URI:	https://norma.ncirl.ie/id/eprint/7618

Actions (login required)

View Item