Energy Load Prediction across Multi-European Countries

Boyini, Kranthi (2024) Energy Load Prediction across Multi-European Countries. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (984kB) | Preview
Preview	PDF (Configuration Manual) Download (530kB) | Preview

Abstract

This work addresses a very important challenge in striking a balance between the accuracy and interpretability of energy load prediction in many European countries-one of the key components in modern energy management systems that are rapidly integrating renewables. This methodically investigated the best model to apply, starting from traditional time series methods up to state-of-the-art machine learning approaches, followed by the selection of the best one, which provided the highest results with a normalized MAE of 0.0158 and MSE of 0.00050. The analysis is supported by comprehensive data from three European countries over five continuous years, 2014-2019, including weather patterns, temporal features, and renewable energy generation. The temperature pattern and temporal features were the main drivers in the energy consumption, but large regional variations existed in the prediction patterns. SHAP helped to explain model decision-making both at a global level through feature importance and at a local level through prediction explanation. It has, therefore, performed a proof-of-concept in the development of state-of-the-art interpretable methods for energy load prediction by providing both theoretical contributions through systematic model comparisons and practical value due to interpretable predictions. Although these results are directly applicable to immediate operational energy practices, there is also identification of very promising future research directions, mainly concerning developing adaptive frameworks for real-time prediction and integration within renewable energy systems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Singh, Jaswinder UNSPECIFIED
Subjects:	D History General and Old World > D History (General) > D901 Europe (General) Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HC Economic History and Conditions > Natural resources > Power resources > Energy consumption H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Energy industries 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:	31 Jul 2025 13:56
Last Modified:	31 Jul 2025 13:56
URI:	https://norma.ncirl.ie/id/eprint/8390

Actions (login required)

View Item