Talks & engagement

Talks

Selected talks, panels, conference presentations, and workshops connecting research with practice.

2026

The Internet meets the Electricity Grid: Technical Standards and Societal Challenges

Role: Panelist | Session: Internetification of the Electric Grid | Date: 16 July 2026 | Location: University of Passau, Germany

I joined experts from the electricity-grid and Internet-protocol communities to discuss how far the electric grid should be “Internetified”, the technical boundaries of this convergence, and the standards and coordination mechanisms needed for sustainable, trustworthy cross-infrastructure operation.

Yihong Zhou presenting research on grid-intelligent AI data centres at London Climate Action Week
AI Data Centres: Power Burden or Future Grid Flexibility Backbone?

Role: Speaker and Panelist | Event: London Climate Action Week | Location: The ZERO Institute, University of Oxford

I presented our latest research on data-centre grid intelligence and compute flexibility in a panel chaired by Prof. Thomas Morstyn. The discussion brought together Vince Colby (Pure Data Centres Group), Frederik Drewes (Loadwise), and Ethan Tiao (Emerald AI), connecting academic research with data-centre operations and flexibility demonstrations.

The Q&A covered grid-balancing cost allocation, data-centre water use, flexible grid connections, and how compute flexibility could help bring resilient digital infrastructure to developing regions.

Yihong Zhou presenting supervised reinforcement learning for distributed energy resources at PSCC 2026
XXIV Power Systems Computation Conference (PSCC 2026)

Role: Presenter | Date: 8–12 June 2026 | Location: Limassol, Cyprus

I presented our work on Supervised Reinforcement Learning for the Coordination of Distributed Energy Resources, connecting supervised pre-training and reinforcement-learning fine-tuning for more efficient and robust DER coordination. The conference also provided a great opportunity to exchange ideas with the international power-systems computation community.

FleXEdge project workshop at the University of Oxford
FleXEdge Project Workshop (Kick-off)

Role: Participant | Location: University of Oxford

We organized the first FleXEdge workshop at Oxford and welcomed collaborators from Imperial College London, the University of Edinburgh (EPCC), and the Luxembourg Institute of Science and Technology (LIST).

Panel on electricity markets and AI at the IEEE PES International Meeting
Panel 4.5: Reinventing Future Electricity Markets with AI (IEEE PES International Meeting)

Role: Panelist | Location: Hong Kong

Invited by Prof. Hongye Guo (Tsinghua University), I shared our recent work on scalable and safety-aware decision-making tools for future electricity markets.

2025

Research presentation at the University of Exeter
GW4 Exeter Workshop

Role: Session Chair and Speaker | Location: University of Exeter, UK

Invited by Dr. Dawei Qiu, I presented our recent research on scalable and uncertainty-aware flexibility dispatch in power systems.

SAGEflex workshop at the University of Oxford
Oxford Workshop on Safeguarded AI Agents

Role: Organizer and Chair | Location: Saïd Business School, University of Oxford

I organized and chaired this workshop on AI, power systems, and safety, with participants from SSEN, SPEN, UKPN, Octopus Energy, the University of Oxford, and the University of Exeter.

Academic talk at North China Electric Power University
Seminar: Towards Real-World, Computationally Affordable Reformulation of Uncertainty-Aware Power System Optimisation

Role: Speaker | Location: North China Electric Power University, Beijing

Invited by Prof. Zhaohao Ding, I delivered a seminar at my alma mater on practical formulations for uncertainty-aware optimization in power systems.

2024

Presentation at the Power Systems Computation Conference 2024
Power Systems Computation Conference (PSCC)

Role: Presenter | Location: Paris

Presented our accepted paper, "A Novel Surrogate Polytope Method for Day-Ahead Virtual Power Plant Scheduling with Joint Probabilistic Constraints."