PhD in Autonomous Systems and Connectivity - Resilient Electric Vehicle Dispatch Considering Dynamic Environment Uncertainties

Start date: 01 October 2025

In collaboration with ScottishPower Energy Networks (SPEN), this scholarship is offered as part of the TransiT hub, which support researchers to carry out studies in digital twining for transport.

This research aims to utilise reinforcement learning (RL) algorithms for optimising electric vehicle (EV) schedules of charging and discharging (vehicle to grid), focusing on mitigating grid instability caused by uncoordinated EV charging and discharging operations, as well as providing grid operators with insights into potential EV demand and generation profiles based on an optimised operation. Using deep reinforcement learning algorithms, such as Deep Deterministic Policy Gradient (DDPG) or Proximal Policy Optimization (PPO), the system will automatically adapt its dispatch strategies by interacting with real-time grid and environmental data, enabling flexible, data-driven decision-making without requiring detailed models of individual user behaviour. This approach allows for scalable, intelligent coordination between EVs and the grid, facilitating peak shaving, load balancing, and enhanced renewable utilisation, while ensuring a safe operation of EV charging and discharging (i.e. cognisance of cutouts and service cables ratings). The project will cover residential EVs as well as commercial EVs (including electric HGVs). The expected key outputs of the project are a working methodology for the optimal coordination of EV charging and discharging, as well as the profiles of the resulting demand needs and generation (vehicle to grid) to help inform the timing, magnitude and location of potential grid constraints.

How to Apply: Please refer to the following website for details on how to apply:

http://www.gla.ac.uk/research/opportunities/howtoapplyforaresearchdegree/.