Exit Menu

A data-driven approach for optimal distribution network operation


Project description

The UK’s current power network cannot meet the energy demand of an influx of EVs. The coupling will be aided by the development of vehicle-to-grid charging and Internet of Things platforms, which will enable the smart management of assets to facilitate a low carbon network. However, as these shifts in demand and supply happen, it will become increasingly difficult for already congested areas of the grid to balance the added pressures of widespread EV charging. A solution is to build large-scale energy storage infrastructures to provide energy for the rapid charging demand. The battery cost has plummeted in the past decade due to technological advancement. By providing a buffer to ease the transmission of electricity along congested lines, large-scale battery storage can enable active management of distribution networks to avoid grid overloads and allow smoother transitions to decarbonise both the energy and transport sectors.

This project develops an innovative toolkit to enable active management of distribution networks to maximise the distribution network’s voltage stability and minimise EV charging cost with large-scale battery storage. The purpose is to address the growing demand of fast charging stations for EVs. To account for the uncertainties introduced in the distribution networks with intermittent renewables and charging behaviour, a multi-agent reinforcement learning approach is adopted to identify the optimal operation for the distribution networks.By 2040, it is estimated that there will be 500 million EVs globally and there will be 36 million EVs in the UK. These EVs are radicalising the way citizens transfer, use and interact with energy and will have a great impact on local and national energy networks. There is a pressing need to identify and utilise the relevant energy storage technologies to mitigate wide-scale blackout due to power supply and demand imbalance.

Project duration: 1st February 2021- 30th July 2021

Meet the Principal Investigator(s) for the project

Dr Chun Sing Lai - Dr Chun Sing Lai is a Lecturer and module leader for EE1618 Devices and Circuits, taught at Chongqing University of Posts and Telecommunication (CQUPT). He is an academic member of the Transnational Education (TNE) CQUPT programme for BEng Electronics and Communications Engineering. He is a member of Brunel Interdisciplinary Power Systems (BIPS) Research Centre. From 2018 to 2020, Dr Lai was an EPSRC Research Fellow with the Faculty of Engineering and Physical Sciences, University of Leeds and also a Visiting Research Fellow with the Department of Electrical Engineering, School of Automation, Guangdong University of Technology, China. He is Vice Chair of the IEEE Smart Cities Publications Committee. He organised the workshop on Smart Grid and Smart City, IEEE SMC 2017 in Canada and workshop on Blockchain for Smart Grid, IEEE SMC 2018 in Japan. His current interests are in power system optimisation, energy system modeling, data analytics, and energy economics for renewable energy and storage systems. Since 2020, Dr Lai is an Associate Editor for IET Energy Conversion and Economics. He is a Guest Editor for Frontiers in Energy Research Journal: "Advanced Energy and Active Management for Smart Grids", Topics Board member for Electronics, and Reviewer Board member for Applied Sciences.
A data-driven approach for optimal distribution network operation