The system operates under time-varying energy supply, dynamic vehicle demand, and operational constraints, forming a complex environment for coordinated decision-making. The modelling framework captures interactions between energy flows, storage states, and service operations within a unified simulation environment.

This research investigates the intelligent operation of an integrated electricity–hydrogen–transportation energy system, where multiple operational tasks must be jointly optimized to ensure efficient, low-carbon, and reliable system performance. The study addresses a multi-task operational problem involving:

• • Energy dispatch — coordinating electricity and hydrogen flows across renewable generation, storage, and conversion units,
  • Vehicle scheduling — managing dynamic charging and refuelling services for electric and hydrogen vehicles.

• • Renewable generation is intermittent and uncertain,
  • Energy storage and conversion processes are dynamically coupled,
  • Vehicle arrivals and energy demands are stochastic,
  • Energy management and service scheduling decisions interact with each other.

Traditional rule-based or single-objective optimization methods struggle to capture these nonlinear and time-dependent interactions. In particular, energy dispatch decisions influence vehicle service performance, while charging demand affects system energy states.

To address the coupled operational challenges, the research develops an intelligent control framework that learns coordinated operational policies for multiple system tasks.

The study explores:

• • Data-driven system modelling of energy and service dynamics
  • Multi-objective optimization of dispatch and scheduling decisions
  • Learning-based decision-making methods for adaptive system operation

A multi-agent learning framework is investigated, where different decision components focus on energy management and vehicle service scheduling, while learning to coordinate system-wide objectives such as cost efficiency, carbon reduction, and service performance. The approach enables adaptive operation under uncertain energy supply and demand conditions without relying on fixed rule-based strategies.

The research aims to:

• • Improve efficiency of integrated multi-energy systems
  • Enhance renewable energy utilization and reduce carbon emissions
  • Improve charging and refuelling service quality
  • Enable intelligent coordination across electricity, hydrogen, and transport infrastructures
  • Support resilient zero-emission mobility systems

The outcomes provide insights for designing future zero-emission energy and transport infrastructures and contribute to the development of resilient urban energy systems and sustainable mobility solutions.

This research investigates the intelligent operation of integrated electricity–hydrogen–transportation energy systems that support electric and hydrogen vehicles. It addresses a coupled multi-task problem involving energy dispatch and vehicle scheduling under uncertain energy supply and dynamic demand. The study explores learning-based coordination strategies to improve system efficiency, reduce carbon emissions, and enhance service performance in future zero-emission energy infrastructures.

You’ll receive the latest updates on CLEETS research and programming.