This PhD will develop machine learning and optimisation methods to improve climate modelling and integrate climate models with energy system models. The project will leverage GPU programming and provide decision-support tools for sustainable planning under climate uncertainty, contributing directly to Europe’s energy transition and net-zero targets.
Climate change poses unprecedented challenges to energy system planning and operation. Long-term uncertainties in climate patterns affect renewable energy availability, energy demand, and system resilience. This project aims to bridge climate modelling and energy system modelling through advanced machine learning (ML) and optimisation.
The PhD student will develop ML techniques to extract key features from climate models and construct scenario representations that can be integrated into long-term energy system planning models. Optimisation algorithms will then be designed to handle these large-scale, stochastic problems efficiently, balancing cost, resilience, and sustainability. For example, optimal decisions could account for potential new energy technologies and robustness to extreme climate scenarios. Particular emphasis will be placed on using ML methods to generate computationally tractable, yet informative scenarios, and algorithms to solve the underlying energy models accounting for climate-driven uncertainty at both short-term and long-term horizons. Algorithms will be designed to be scalable (e.g., GPU parallelisable) at both steps.
The student will work at the interface of applied mathematics, computer science, and climate/energy policy, joining a growing network of researchers at Southampton and Imperial College. There will be opportunities to collaborate internationally and contribute to cutting-edge research on sustainable energy transitions.
The project offers the chance to contribute to the global net-zero agenda by providing innovative computational tools for climate-resilient energy system planning.
Please find more information about the application here: https://www.mfccdt.ac.uk/how-to-apply/.
Please direct any questions to the supervisor Dr Hongyu Zhang, hongyu.zhang@soton.ac.uk.