AI and Agent-based Control

Artificial Intelligence and Agent-based Control for Improving Energy Network Resilience to Threats (2026)

What you get

This self-funded PhD project is available to UK and Overseas applicants who are able to self-fund or identify their own sources of funding. You will be supported when applying to any external sources.

The University of Sussex believes that the diversity of its staff and student community is fundamental to creative thinking, pedagogic innovation, intellectual challenge, and the interdisciplinary approach to research and learning. We celebrate and promote diversity, equality and inclusion amongst our staff and students. As such, we welcome applicants from all backgrounds.

Type of award

PhD

PhD project

How can we leverage artificial intelligence to tackle modern serious threats to energy infrastructure that leave millions without power?

This PhD project aims to investigate the use of Artificial Intelligence (AI) tools, including machine learning (ML) and agent-based control, for predicting, managing and improving the resilience of energy networks to disruption.

AI tools will be used to predict the likelihood and impact of cascading failures. Cascading failures can lead to widespread electrical blackouts, typically characterised as High-Impact Low Probability (HILP) events, potentially leaving millions of people without energy, water or communications, risking lives, and costing £ billions. Prior knowledge of the occurrence of such HILP events can enhance the response of infrastructure operators, thus limiting their impact.

You will build on prior research that has been done by the supervisor’s team on leveraging machine learning to predict large-scale blackouts, including the Network Theory Resilience Metric (NTRM) toolkit (https://github.com/sskazakos/NTRM).

What you will do

• Develop a prototype toolkit, which can be used to assess the resilience of energy networks and link with industrial systems to extract data and advise on the response interventions.
• Work with datasets from energy networks, wherever possible.
• Build advanced simulation models utilising machine learning and agent-based control techniques.
• Collaborate with researchers and industry stakeholders.
• Publish in high-impact journals and conferences.

Skills you will develop

• Energy network and complex systems modelling
• Artificial intelligence methods applied to infrastructure
• Resilience and risk analysis for critical systems
• Experience with real-world datasets
• You will benefit from our researcher development training programme, to enable you to develop your skills as a researcher and ensure you have what it takes to be successful in your future career.

Applicants (or their funder, if one is identified) will be liable for the appropriate fees, plus a one-off Research & Training Support Grant (RTSG) of £2,000, to support their research. Home PhD student fees are set at the level recommended by United Kingdom Research and Innovation (UKRI) annually, rising in line with inflation. Overseas fees are subject to an annual increase - see details on our tuition fees page.

Supervisor: Dr Spyros Skarvelis-Kazakos (s.skarvelis-kazakos@sussex.ac.uk) – my research expertise covers many aspects of smart grids and infrastructure resilience. Research interests include energy network resilience and reliability, real-time simulations of power system cascading failures, critical infrastructure interdependencies, graph theory, impact of climate change and pandemics on energy networks, adaptation and mitigation of those impacts, intelligent control / aggregation of Distributed Energy Resources, multiple energy carriers / integrated energy systems, micro-grids, Virtual Power Plants, energy storage, multi-agent systems, electric vehicles. Google Scholar profile: https://scholar.google.co.uk/citations?user=F4u6xmEAAAAJ&hl=en&oi=ao

Further information on this approach can be found on the website of the Critical Infrastructure Resilience Network (CIReN): https://www.sussex.ac.uk/research/centres/critical-infrastructure-resilience-network/publications

Eligibility

Full time or part time, open to UK and international applicants. Ideal candidates will have a background in electrical engineering, physics, mathematics, or data science, as well as interest in complex networks or infrastructure systems. Some programming experience and motivation to work at the interface of engineering, data and resilience are desirable. Eligible candidates will normally have an upper second-class (2:1) undergraduate honours degree (or equivalent qualification) in a related field.

Please ensure you application includes each of the following:

• A research proposal.
• Your CV.
• Degree certificates and transcripts.
• 2 references, including a minimum of 1 from any institution studied at within the last 5 years.
• If your first language is not English you will need to demonstrate that you meet the University’s English language requirements.

Deadline

How to apply

• Apply online for a full time PhD in Engineering using our step-by-step guide
• Please clearly state on your application that you are applying for the “Artificial intelligence and agent-based control for improving energy network resilience to threats” project under the supervision of Dr Spyros Skarvelis-Kazakos.

Contact us

If you have practical questions about the progress of your on-line application, contact FoSEM-PGR@sussex.ac.uk

For academic questions please contact Dr Spyros Skarvelis-Kazakos (s.skarvelis-kazakos@sussex.ac.uk) 

Timetable

Application Deadline: 01 January 2027

Interview Date: TBC

Entry Date: Academic year 2026/27 (Between September 2026 and April 2027)