Watch Again | IEEE Power and Energy Seminar – Can Machine Learning Help Secure Power Systems? by Dr Panagiotis Papadopoulos, University of Manchester

This event was co-hosted by the PES Student Branch Chapter at The University of Manchester and all other PES Student Branch Chapters in UK and Ireland, and was sponsored by The University of Manchester.

Synopsis

Electrical power systems are undergoing unprecedented changes mainly driven by decarbonisation targets and climate change as well as other technical, economic and social reasons. This leads to the integration of new technologies such as renewable generation, electric vehicles, HVDC links, etc. These devices are mostly connected via power electronics with very different dynamic behaviour, leading to increasing complexity of power system dynamics. In addition, uncertainty is also increasing due to intermittent nature of renewable generation but also because of how society will use energy on the way to decarbonization (e.g. electrification of transport or possibly heating). At the same time, advanced measurement and communication infrastructure is being integrated in modern power systems. Such technologies, especially couple with machine learning, offer opportunities for advanced situational awareness, decision support tools and automated control methods.

This presentation highlights the challenges faced in future power systems with high penetration of converter connected units, in terms of their dynamic behaviour, and discusses methods and tools to tackle them, inspired to a large extent by machine learning. Such methods can enable system operators to consider detailed dynamics in cases where the computational effort needed is otherwise prohibitive. Two main aspects are discussed related to: i) how do we model and characterise the complex and uncertain dynamic behaviour in power systems with high converter penetration and ii) how machine learning can enable fast and informative dynamic security assessment with focus on how to build trust in such methods going beyond the notion that machine learning is simply a powerful black-box predictor.

Probabilistic stability assessment methods to quantify the impact and characterise new types of interactions of power systems with converter connected units at both transmission and distribution level are also discussed. Machine learning based methods to enable fast stability assessment in operational and planning timescales are presented with a focus on explainability for improved decision support, graph-based methods that take into consideration dynamics, and physics informed reinforcement learning. Methods enabling the ability to account for detailed dynamics in optimisation are also be presented.