A hybrid optimization strategy of electrical efficiency about cooling PEMFC combined with ultra-thin vapor chambers

The fan's cooling is critical to the overall safety and efficiency of an open-cathode proton exchange membrane fuel cell (PEMFC). The power of the fan directly determines the temperature, parasitic power, and electrical efficiency. Existing experimental research and simulations are still insufficient for the optimization of the electrical efficiency by adjusting the power of the fan in an open-cathode PEMFC. This paper develops a hybrid optimization strategy based on the team-designed open-cathode PEMFC combined with ultra-thin vapor chambers (UTVC) to search for the optimal fan power at a specific temperature and maximize the stack's electrical efficiency. The dynamic model of the PEMFC-UTVC system is established and validated by experiments by analyzing the heat dissipation structure of the PEMFC-UTVC system and combining the electrochemical principle with the thermodynamic theory. To maximize the electrical efficiency of the PEMFC-UTVC stack, a neural network method is integrated with a meta-heuristic algorithm with mutation. The penalty function method is used to transform the problem constraints throughout the optimization process. The optimization strategy demonstrates the advantages of a powerful optimal capacity, high precision, and high robustness. The results of optimization show that the electrical efficiency of PEMFC-UTVC can be maximized by adjusting the power of the fan. Under the current conditions of 20A, 30A, and 40A, the maximum rates of rise in electrical efficiency are 2.65%, 3.29%, and 4.35%, respectively. The proposed hybrid optimization strategy is expected to provide insights into the control and optimization of the open-cathode PEMFC system.

Automated summary

This paper proposes a hybrid optimization strategy to maximize the electrical efficiency of an open-cathode PEMFC combined with ultra-thin vapor chambers (UTVC). By establishing a dynamic model and integrating a neural network method with a meta-heuristic algorithm, the optimization results show that adjusting the power of the fan can improve the electrical efficiency of the PEMFC-UTVC.