Invited Speaker丨特邀报告

Prof. Duo Li, Huazhong University of Science and Technology

Duo Li, Professor, National Young Talent, Humboldt Fellow, IEEE Senior Member. He received his Bachelor’s, Master’s, and Doctoral degrees from Huazhong University of Science and Technology, The University of Queensland, and The University of Auckland, respectively. His previous appointments include positions at the University of Cambridge, Newcastle University and the German Aerospace Center (DLR). His research focuses on Intelligent Transportation Systems (ITS), Traffic Sensing and Control, Transportation Resilience, and Urban Computing. He serves as an Associate Editor for leading journals, including the IEEE Transactions on Neural Networks and Learning Systems, Transportation Research Record, and the IEEE Open Journal of Intelligent Transportation Systems. He is also Peer Review College member for the UK Royal Academy of Engineering (RAEng) and the UK Engineering and Physical Sciences Research Council (EPSRC), Member of TRB Highway Traffic Monitoring Committee, Fellow of World Economic Forum (WEF) Sustainable Tourism Council, and Fellow of the Higher Education Academy (FHEA), UK. He has been PI/Co-I in over 30 research projects funded by China, the UK, Germany, and wider European programmes. He has published over 50 papers in top-tier peer-reviewed journals, with an h-index of 24.

Speech Topic: Cyberattack-Resilient EV Charging Demand Forecasting

Abstract: The Electric Vehicle (EV) market is experiencing unprecedented growth. Accurate prediction of EV charging demand is essential for transportation system operations, such as real-time traffic management, route optimization, and station utilization planning. However, Cyber threats can compromise the accuracy of charging demand predictions, leading to significant disruptions in transportation services, e.g., suboptimal station management, unexpected congestion at charging facilities, and degraded service quality for EV users. This study introduces Generative Multi-task Self-supervised Learning for Prediction (GenS2-P), a cyberattack-resilient framework designed to ensure reliable charging demand predictions under adversarial conditions. GenS2-P incorporates a Denoising/Reconstruction AutoEncoder (DRAE) and a spatio-temporal prediction model to tackle the dual challenges of data poisoning and DoS attacks. By leveraging generative self-supervised learning and multi-task learning, GenS2-P effectively extracts spatio-temporal patterns to denoise and reconstruct data corrupted by cyberattacks. Experimental evaluations using real-world EV charging data demonstrate that GenS2-P significantly reduces prediction errors and mitigates cyberattack-induced disruptions. This improved prediction reliability enables more effective charging infrastructure management and supports robust transportation system operations even under adverse conditions.