A fuel cell bus power-thermal collaborative optimization method based on reinforcement learning
By establishing dynamics, energy flow, and temperature rise models for fuel cell buses, and combining them with the reinforcement learning TD3 algorithm to optimize fuel cell output power and thermal management, a deep coupling optimization of power distribution and thermal safety of fuel cell buses is achieved, solving the problem of improving the overall performance of fuel cell buses under dynamic operating conditions.
Patent Information
- Application Number
- CN202610323920.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing fuel cell buses have decoupled control between their energy management and thermal management systems, making it difficult to coordinate power distribution and thermal safety under dynamic operating conditions. This leads to accelerated degradation of the lifespan of key components and thermal safety risks, thus limiting the overall performance of the vehicle.
A dynamics, energy flow, aging, and temperature rise model of a fuel cell bus was established. The reinforcement learning TD3 algorithm was used to optimize the fuel cell output power and thermal management. Through multi-agent collaborative optimization of coolant flow and radiator airflow, precise control of fuel cell temperature was achieved.
It significantly improves the overall energy efficiency of fuel cell buses, extends the lifespan of key components, and ensures the thermal reliability of the system, demonstrating outstanding engineering application value.
Smart Images
Figure CN122413894A_ABST