Project Abstract

This project is driven by the growing sustainability demands and the increasing need to reduce carbon emissions and optimise energy use. Amidst these challenges, this research aims to address critical gaps in how AI-driven approaches can significantly lower operational carbon intensity and costs for buildings. The work is positioned uniquely at the intersection of machine learning and energy management, the core proposition is to develop intelligent models that leverage historical data from renewable sources. These models enable predictive strategies to minimise energy costs and environmental impacts simultaneously. The study employs an integrated approach combining time-series forecasting, optimisation algorithms, and supervised machine learning techniques. These techniques are applied to benchmark datasets like the ASHRAE Great Energy Predictor III energy data. The results demonstrate substantial potential for cost savings and carbon intensity reduction. They also highlight clear patterns in renewable resource utilisation and uncover actionable insights into optimal energy management practices. Ultimately, this research contributes novel methodologies and robust predictive models. These expand knowledge on practical implementations of AI for sustainable energy management. They also offer clear pathways for significantly reducing both environmental footprints and operational costs in real-world applications.

Keywords: Artificial Intelligence Driven Sustainability, Machine Learning Energy Optimisation, Sustainable Energy Modelling

 Conference Details

Session: A

Location: Sir Stanley Clarke Auditorium at 11:00 13:00

Markers: Deepak Sahoo, Fernando Maestre Avila

Course: BSc Computer Science 3yr FT

Future Plans: I’m looking for work