Imperial College London

DrYunjieGu

Faculty of EngineeringDepartment of Electrical and Electronic Engineering

Lecturer in Power Systems
 
 
 
//

Contact

 

yunjie.gu Website

 
 
//

Location

 

1119Electrical EngineeringSouth Kensington Campus

//

Summary

 

Overview

My research focuses on the technologies and analyses to facilitate the zero-carbon transition of power systems. One important feature of this transition is the replacement of rotating generators with power-electronic converters as the interfaces between resources (wind, solar, battery, interconnector, nuclear, hydro, .etc) and networks. This structural change results in a paradigm shift in system dynamics, from physical-defined behaviours (with software amendments) of generators, to software-defined behaviours (with physical constraints) of converters. This paradigm shift creates new patterns of instabilities in power systems that are not well understood, which may have catastrophic consequences for society and the economy, like the 2019 power cut in the UK.

The difficulties of stability analysis for power-electronics-dominated power systems are three-fold: (1) the lack of transparent and unified models since the control software that defines these models is proprietory to vendors; (2) the lack of a mathematical structure (like the Hamiltonian structure of conventional power systems) that eases dynamical analysis; and (3) the lack of consistent behaviours under normal and fault conditions due to tight physical constraints. My research aims to overcome these difficulties with a combination of model-driven and data-driven methods. On the model-driven side, I (and my collaborators) explored the first principles of ac transmission that are agnostic to technologies. I illustrated the duality property that unifies the synchronisation behaviour of different apparatuses and controls. I extended the stability categorisation of power systems to accommodate the new behaviours of power-electronics-based resources into the classic stability framework. I explored the intrinsic similarity between power systems and communication systems and used this to demonstrate how an all-power-electronics system can be stabilised without equivalent inertia. On the data-driven side, I developed the grey-box approach to trace the origins of oscillations from data without transparent models. Mathematically, the grey-box approach is abstracted as port-based modal analysis which is a generalisation to the conventional state-based modal analysis via the chain rule.  

Awards


My research is supported by research councils and industrial partners under the following awards:

  • Network Innovation Allowance (Funded by National Grid ESO): Data-Driven Online Monitoring and Early Warning For System Stability (DOME), 2023-2025. CoI
  • Royal Society International Exchange Scheme: Interconnection Converters and Interaction Analysis in DC Grids, 2021-2024. PI
  • EPSRC Innovation Fellowship: Grid Supportive Power Electronics for Power System Security, 2018-2021. PI



Guest Lectures

Power system stability with high penetration of inverter-based resources, CIGRE UK B4 Liaison Meeting, 2023

The grey box approach: tracing the root cause of oscillations in a complex grid with impedance-based models, IEEE Task Force on Frequency-Domain Modeling and Dynamic Analysis of HVDC and FACTS, 2023

New tools for stability analysis of inverter-based power systems, UK National HVDC Centre, 2023

A framework for analysing mechanical-electronic composite power systems, IEEE 2nd Workshop on Power Electronics for Grid Dynamics (PEGD), 2019

A framework for analysing mechanical-electronic composite grid dynamics, IEEE eGrid Workshop, 2019

POWER SYSTEM CONTROL AND STABILITY WITH POWER ELECTRONICS, IEEE UK and Ireland Section, 2018