Two parameters identification for polarization curve fitting of PEMFC based on genetic algorithm

Experimental and numerical studies are the main methods to obtain the polarization curves of proton exchange membrane fuel cell (PEMFC) stack for the application of balance of plant (BOP) selection, system matching, and controller design. A novel methodology of polarization curve fitting for PEMFC is proposed based on genetic algorithm, and a rated power of 62 kW fuel cell stack is used for identification experiments. The exchange current density and charge transfer coefficient are selected as optimization variables. The results show that the two optimization variables can effectively replace the four empirical parameters in the calculation of activation overpotential and demonstrate good agreement between the proposed model and experimental data within 3% error. In addition, a group of polarization curves is obtained by considering the relationship between the exchange current density and reactant concentration, activation energy, and temperature, which is of great significance in the engineering application of PEMFC for quick and accurate control. Furthermore, a dynamic model is also used to investigate the transient response of PEMFC, and the simulation results show a transition region of voltage drop at step current densities, coinciding with the experimental data and indicating the reliability of the method for polarization curve fitting.

Automated summary

A novel methodology for fitting polarization curves of PEMFC is proposed based on genetic algorithm, and the results show good agreement with experimental data. Furthermore, a dynamic model is used to investigate the transient response of PEMFC, and the simulation results match well with experimental data, indicating the reliability of the method for polarization curve fitting.