Solar-Net: Leveraging Transformers for Enhanced Solar Power Prediction

Shaik, Akif Roshan (2024) Solar-Net: Leveraging Transformers for Enhanced Solar Power Prediction. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Accurate forecasting of electricity generation from solar systems is very critical for efficient grid balancing and integration of renewable energy sources. This study explores the application of machine learning (ML) and deep learning (DL techniques to predict solar power generation using various environmental data. Specifically, it evaluates the performance of various traditional ML and DL algorithms and proposed Transformer-based model, Solar-Net. The dataset used for experiments in this study has 21 independent variables including temperature, humidity, radiation, wind speed and the solar power output which is the dependent variable expressed in kilowatts. The results have shown that Solar-Net Transformer model had higher predictive accuracy than the traditional ML models with an R² score of 81.54%, the lowest MSE of 0.1906 and the lowest MAE of 0.2899. For Transformer-based models with attention, the study demonstrates the possibility of using this approach to increase the accuracy of forecasting of solar power generation based on time series data. Future work will include the consideration of more sources of information, improving the models’ accuracy in real-time predictions, and extending the applicability of the models for different geographic areas and conditions. The findings of this research will be useful in the existing literature on solar energy forecasting as well as the use of deep learning in renewable energy systems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Milosavljevic, Vladimir UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning H Social Sciences > HC Economic History and Conditions > Natural resources
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	04 Sep 2025 14:16
Last Modified:	04 Sep 2025 14:16
URI:	https://norma.ncirl.ie/id/eprint/8797

Actions (login required)

View Item