SOURCE-LINKED INTELLIGENCE

A temporal graph neural network and reinforcement learning based secure and efficient V2G system

CORDIS · observation · Publication date unknown

critical flexibility but also exposes the power system to new cybersecurity risks. The TRUST-V2G project will develop a secure, privacy-preserving, and efficient V2G framework that combines advanced artificial intelligence with real-world validation in European living labs. The research introduces three integrated innovations: (i) Federated Privacy-Preserving Intelligence to detect cyber threats across heterogeneous V2G networks while guaranteeing data protection; (ii) Adaptive Temporal Graph Neural Networks to model evolving attack patterns and anticipate cascading risks; and (iii) Multi-Agent Reinforcement Learning-based Defense Orchestration to optimize grid resilience, safety, and performance under adversarial conditions. TRUST-V2G will be validated at Linköping University Vehicle Lab with over 20 operational V2G systems, advancing technology readiness from TRL 3 to 5. This project will enhance the scientific skills and innovation capability, expand research horizons, and establish research collaborations with both the academic & non-academic sectors of the applicant. The fello

Read original source ↗ Open in workspace

recordType

award

status

SIGNED

region

EU

value

252180

unit

EUR