Multi-Stack Fuel Cell System Stacks Allocation Optimization Based on Genetic Algorithms

High-powered and modularity is the trend for fuel cell systems. Similar to the evolution from single-cylinder to multi-cylinder in conventional internal combustion engines, fuel cell systems shall also follow this developing process. Compared to single-stack fuel cell systems, multi-stack fuel cell systems (MFCS) can enhance the system maximum output power and improve the system performance. To achieve modular design and improve the performance of high-powered MFCS, a MFCS stacks allocation optimization algorithm based on genetic algorithms is proposed in this paper. First, remaining useful life (RUL) and efficiency are choosing as an integrated optimization index, the decision model for MFCS stacks allocation is developed. Then, a heavy-duty commercial vehicle was used as an example to match the vehicle power train parameters. The genetic algorithm is used to solve the global optimal stacks allocation scheme for the vehicle in a specific application scenario. The individual fitness values in the genetic algorithm are solved by the SQP algorithm. The optimized results are analyzed and beneficial conclusions are obtained. Weights of efficiency and RUL and the number of stacks has impacts on the determination of optimal stacks allocation scheme. The overall system efficiency will increase with the efficiency factor weight increase, and the overall RUL will decrease with the efficiency factor weight increase. If the focus is on improving the overall economy of the MFCS, the efficiency/RUL weights shall be taken as 1/0. Finally, the impact of different stacks allocation schemes on the efficiency and RUL of the MFCS is compared.