IEEE Power and Energy Webinar – Trustworthy Machine Learning for Power System Situational Awareness by Dr Tabia Ahmad, University of Manchester

Abstract

The electric power system is witnessing significant transformations towards an integrated, active, and ubiquitously sensed cyber-physical system.

An abundance of data from phasor measurement units (PMU), point on wave (POW) devices and digital disturbance recorders offers tremendous opportunities as well as scientific challenges for situational awareness of power systems. Also, to mitigate climate change, power systems worldwide are increasingly moving towards more and more renewable energy (RE) generation which might lead to complex and uncertain operations.

Building on theoretical foundations, this talk aimed to provide an overview of machine learning (ML) and data analytic tools for better monitoring of converter interfaced RE integrated power systems, especially in the event of catastrophic failures. The key highlights of this talk included revisiting the nature of field power system measurement data and discussing how this may affect the accuracy of downstream analytics such as power system dynamic stability.

The potential of ML leveraging spatio-temporal data for specific problem of detecting cascading failures due to transient response of power systems in RE integrated power systems will also be discussed. The talk concludes on how trust can be built in ML based techniques to support the operation of future power systems.

About the Speaker

Tabia Ahmad is working as a post-doctoral researcher with the Power and Energy division, EEE department of University of Manchester and has previous postdoctoral experience at the Institute of Energy and Environment, EEE, University of Strathclyde, Glasgow (2021-2023) working on, addressing the complexity of future power systems’ dynamic behaviour.

She earned her PhD in electric power systems (2021) from the Indian Institute of Technology Delhi, India with a Distinction in Doctoral Thesis Award and POSOCO Award for best Doctoral Thesis relevant to power transmission utility. Prior to that, she received a BEng in Electrical Engineering and the MEng in Instrumentation and Control Engineering (with University Gold Medal) from AMU, India, in 2014 and 2016, respectively.

Her research interests include power system dynamics, trustworthy ML for power systems, WAMS based analytics and signal processing techniques in power systems. She is also passionate about addressing climate change through her research on decarbonised power systems and engagement with the Climate Change AI Initiative.