A Health-Driven Integrated Optimization of Energy Management and Component Sizing for Fuel Cell Hybrid Electric Vehicles

Optimized algorithms for component sizing and energy management are pivotal in reducing energy consumption, operating cost and extending fuel cell (FC) lifespan in FC hybrid electric vehicles (FCHEVs). These interdependent challenges are of significant research interest. Therefore, this paper proposes an integrated approach to optimize energy management control and system design for an FCHEV, featuring a battery-supercapacitor hybrid energy storage system (HESS), by employing the non-dominated sorting genetic algorithm III (NSGA-III). The optimization process targets maximizing driving range and FC lifespan, while concurrently minimizing hydrogen consumption, HESS size, and overall lifetime costs. The health-driven fuzzy logic-based energy management strategy (HFBEMS) is proposed to allocates power demand between the FC and HESS effectively by considering FCHEV operating cost optimization and the thermal constraints for battery safety. The optimized FCHEV configuration achieved a hydrogen consumption rate of 0.58 kg/100 km and driving range of 411.4 km for US06 test driving conditions, while maintaining an optimal trade-off among all objectives. Furthermore, this optimized HFBEMS proved resilient under real-world driving conditions like US06 and UDDS, while ensures compliance with operational constraints, including FC power ramp rate limits, SOC limits, and thermal safety constraints, minimizing FC and battery degradation. A comparative analysis of various EMS revealed that the proposed HFBEMS exhibit notable superiority in minimizing hydrogen fuel consumption, achieving reductions of approximately 38.30% and 25.46%, when compared to the conventional PI control and frequency-decoupling fuzzy-based control, respectively. Consequently, an optimized HFBEMS, due to its rule-based structure and minimal execution complexity, is suitable for real-time integration within a vehicle, and outperforms conventional methods by optimally balancing key performance criteria.