Incorporating Demand and Supply Unit Metadata into Multi-Agent Networks for Optimising National Grid Dispatch Strategies

Lead Research Organisation: University College London
Department Name: Bartlett Sch of Env, Energy & Resources


With the promotion of sustainability and carbon mitigation as a backdrop, this PhD will explore the applications of Agent-Based Modelling and Simulation (ABMS) for assessing the impact of policy, market forces, and technological change on the UK's electrical transmission network, by representing entities such as utilities, generators, and consumers as Agents that learn from the actions of themselves and others through both supervised and unsupervised techniques. Central to this framework will be Liebreich's [2] concept of applying deeper meaning to quantised units of supply and demand in the form of metadata, describing currently 'market smeared' information such as supply sources' response time, cleanliness, and reliability or demand sinks' criticality, predictability, and transmission loss. Studying how Agents interact using this enhanced language to pursue their goals can be directly applied to helping utilities and the National Grid efficiently operate in an increasingly decentralised and
interconnected network; as noted by Snape et al. [4] and Bergman et al. [5], ABMS is particularly suited to modelling such systems in transition. With a testbed trained on authentic TNO dispatch data, we will then evaluate the effectiveness of hypothetical policies and market stimuli, such as subsidy variation and utility quotas, with the aim of securely increasing renewables capacity while minimising spinning reserve, developing these conclusions into actionable scenarios akin to the Shell 2050 Pathways.


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/R512400/1 01/10/2017 31/03/2022
1926550 Studentship EP/R512400/1 25/09/2017 15/09/2021 Connor Galbraith