A Comprehensive Analysis of UK Electricity Consumption Using the OSI Model for Data Communication and Predictive Analytics

Elizhabeth Sebastian, Neethu (2024) A Comprehensive Analysis of UK Electricity Consumption Using the OSI Model for Data Communication and Predictive Analytics. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (4MB) | Preview
Preview	PDF (Configuration Manual) Download (740kB) | Preview

Abstract

Demand forecasting for electricity is important in energy planning and management, for the stability of the electric power grid and the incorporation of renewable resources. This paper aims to establish the accuracy of applying sophisticated forecasting methods that can estimate UK electricity consumption using historical data National Grid ESO for the period 2009 to 2024. Since a large share of renewable energy is expected to be integrated into energy systems, it is paramount to precisely forecast the electricity demand. Thus, the present study employs ARIMA, LSTM, and hybrid models to predict the UK electricity demand based on the dataset of 2009–2024. The work employs a structured-data processing pipeline derived from the OSI model designed to improve and organize the predictive modeling process. The transformations involve normalization, lag features, and seasonality encoding obtained from national demand, wind and/or solar generation, and interconnector flows. The findings highlighted here establish that the effectiveness of the ARIMA model and the LSTM model for the small-scale wind speed data: ARIMA is powerful in capturing linear and periodic trends while LSTM supersedes in capturing other types of trends and long-term oscillations. Other improvements include the extension of convolutional layers with recurrent networks using both local dependencies and temporal patterns. This model simplifies the sending process of data and enhances the efficacy of model training through the advancement in processing. These conclusions provide practical recommendations for grid operators and policymakers to improve grid control, decrease the dependency on conventional resources, and better incorporate renewable energy. Future work should extend to dynamic integration of data, external factors like weather and detailed study of hybrid models to enhance their performance and flexibility. This study is a clear example of how machine learning and structured data framework can revolutionise energy forecasting techniques.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Cosgrave, Noel UNSPECIFIED
Uncontrolled Keywords:	Electricity Demand Forecasting; LSTM (Long Short-Term Memory); Machine Learning; Time-Series Analysis; OSI Model (Open Systems Interconnection)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electricity Supply H Social Sciences > HC Economic History and Conditions > Natural resources > Power resources > Energy consumption 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:	02 Sep 2025 10:27
Last Modified:	02 Sep 2025 10:27
URI:	https://norma.ncirl.ie/id/eprint/8696

Actions (login required)

View Item