A Multi-Sine Excitation Signal Optimization Strategy for EIS Measurement of High-Power Fuel Cells

Electrochemical impedance spectroscopy (EIS) is widely used for fuel cell fault diagnosis. However, traditional EIS measurements take a long time and are difficult to use for real-time diagnosis. Using multi-sine composite signals as the excitation source for fuel cell EIS measurements can shorten the measurement time, but the problem of large signal peaks is also introduced. Moreover, for high-power fuel cell systems, the smallest possible excitation amplitude is needed to reduce power fluctuations, but too small an excitation signal amplitude leads to a lower signal-to-noise ratio (SNR) and poor noise immunity. To tackle this challenge, the strategy proposed in this paper is to maximize the amplitude of each individual frequency component while minimizing the peak value of the composite signal. Firstly, the boundary condition is determined as the peak value of the composite signal does not exceed 1% of the DC current, after that, the amplitude of the individual frequency is changed and then the phase is optimized by genetic algorithm respectively. The multi-sine composite excitation signal closest to the boundary condition is selected as the final optimized signal. Finally, the EIS measurement system is built and the experimental test is carried out on a 7.5mΩ standard resistor. Impedance measurement simulations are also performed on a fuel cell equivalent circuit model. The results show that the proposed optimized multi-sine signal has higher measurement accuracy, better stability for multiple measurements, and improved noise immunity compared with the unoptimized signal.