AI-Driven Optimisation for Energy-Efficient Smart Buildings

Artificial Intelligence is revolutionising how we approach climate challenges in the built environment. This fully funded PhD opportunity will equip you to develop a hybrid AI system that optimises the performance of smart building technologies — specifically HVAC and refrigeration systems — improving energy efficiency without compromising comfort, reliability, or safety. In collaboration with LoweConex, a leading software and analytics provider for connected building assets, this project combines machine learning, physics-based models, and expert domain knowledge to deliver real-time optimisation that is explainable, scalable, and impactful. With access to one of the UK's largest IoT energy datasets, this is a unique opportunity to contribute to the development of AI systems that directly support organisations in achieving Net Zero carbon goals. Research Objectives: • Identify the key drivers influencing the operation and energy consumption of connected building assets. • Develop a decision-making framework that integrates: - Advanced machine learning methods (e.g. reinforcement learning), - Physics-based models of system behaviour, - Expert and regulatory knowledge, - Multi-modal datasets including telemetry, weather, maintenance logs, and potentially video. • Enable real-time optimisation of energy use, even with delayed data (e.g. day+1 MPAN data).