Accurate and Efficient Short-Term Solar Power Forecasting Across Climatic Regions

Danda, Nihal Raj Reddy (2025) Accurate and Efficient Short-Term Solar Power Forecasting Across Climatic Regions. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (631kB) | Preview

Abstract

Accurate short-term solar power forecasting is essential for ensuring grid stability, optimizing renewable energy integration, and enabling intelligent energy management across diverse geographic and climatic regions. While traditional and deep learning-based models have improved prediction accuracy, they often struggle with balancing computational efficiency and generalizability. This study proposes a hybrid Gated Recurrent Unit (GRU) model enhanced with an attention mechanism to address these limitations. The GRU component captures sequential dependencies in solar power data with reduced computational overhead, while the attention layer selectively emphasizes critical time steps, improving accuracy and interpretability. Experimental evaluations across multiple countries India, Brazil, Norway, and Australia demonstrate the model’s superior performance, achieving the lowest MAE and RMSE compared to standalone ANN and GRU models. The results validate the GRU-attention model’s ability to offer high accuracy, lightweight architecture, and adaptability across regions with varied climatic conditions, making it an effective and scalable tool for real-time solar forecasting in modern energy systems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Nagahamulla, Harshani UNSPECIFIED
Subjects:	G Geography. Anthropology. Recreation > GE Environmental Sciences > Environment H Social Sciences > HC Economic History and Conditions > Natural resources H Social Sciences > HC Economic History and Conditions > Natural resources > Power resources
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	24 Aug 2026 12:11
Last Modified:	24 Aug 2026 12:11
URI:	https://norma.ncirl.ie/id/eprint/9603

Actions (login required)

View Item