Electricity control room algorithms for a decarbonised system

Electricity control room algorithms for a decarbonised system

The operation of energy networks requires the processing of enormous amounts of information to ensure a balance between the energy being produced across the energy grid, and the energy demands of the population. The complexity of this challenge has been made all the difficult with the shoft electric vehivle use impacting demand, and with the growing utilisation of green energy sources which may have less predictable production patterns. Within the national grids control rooms, this places specific challenges on the engineers who are making decisions related to switching on and off parts of the grid under increased uncertainty, under time pressure, and with large economic and enviornmental stakes.

The goal of this project is to explore the development of new AI-enabled tools to help decision-makers in control rooms and the preceding planning steps. For this goal, we need to improve techniques in stochastic decision-making, optimisation under uncertainty, the use of high-power computing in energy network planning and operation. We also need these techniques to be embedded in new tools to help decision-makers understand and interpret results in real-time so that decisions can be made quickly and with confidence.

Project partners:

Lars Schewe

Chris Dent

Kings College London

The Alan Turing Institute

National Grid ESO