Machine Learning Approaches for Large-Scale Optimisation in Power Systems

As power systems become increasingly complex—driven by higher renewable penetration, growing uncertainty, and decentralised assets—traditional optimisation methods such as mixed-integer programming can struggle to deliver solutions fast enough for real-time or large-scale applications. Data-driven surrogate models offer faster alternatives but often compromise on accuracy, particularly in respecting physical constraints and maintaining reliability across diverse operating conditions. This project aims to develop a machine learning framework that integrates specialised models to tackle different components of large-scale power system optimisation problems. The primary focus will be on enhancing the scalability and reliability of solutions to key problems such as Optimal Power Flow (OPF), Unit Commitment (UC), and Security-Constrained Optimal Power Flow (SCOPF).

For more information about the project contact Dr Waqquas Bukhsh (waqquas.bukhsh@strath.ac.uk), Lecturer in Advanced Optimisation and Decision Support at the Department of Electronic and Electrical Engineering at the University of Strathclyde.

For a list of the research areas in which ARCHIE-WeSt users are active please click here.