Research On Operating Conditions Identification Of Proton Exchange Membrane Fuel Cell Stacks Based On AC Impedance

Proton exchange membrane fuel cell has promising application prospects in energy and transportation fields due to its high-power conversion efficiency and fast response.The reliability and durability of fuel cell systems are the bottlenecks that restrict their large-scale commercial application.The difficulty in improving the reliability and durability of the system lies in the lack of effective means to accurately identify the operating conditions(e.g.,hydration,flooding,air starvation and other abnormal operating conditions)in the stack,and the health management method based on the operating condition identification.Therefore,this thesis investigates the steady-state and dynamic response characteristics of an engineering-grade stack using AC impedance online measurement technology,and proposes an approach for online identification of the operating states based on local impedance information.This operating state identification method has been applied to the closed-loop management of fuel cells to avoid various abnormal operating states in real time and improve the reliability and durability of the system.The specific research of this thesis is as follows.1.Characterization of Electrochemical Impedance Spectra(EIS)of PEMFCs.At first,the correspondence between the operating conditions and polarization impedance has been determined by comparing the principle that operating conditions affect the operating conditions and the polarization impedance.Then,experiments have been carried out to identify the operating conditions based on the EIS.the feasibility of the EIS to identify the operating conditions has been verified from both qualitative and quantitative perspectives of the results.Finally,the shortcomings of the EIS-based states identification have been analyzed from two aspects,i.e.,the time and calculation costs.Based on analysis of shortcomings,the necessity of the impedance information-simplified state identification method is introduced.2.Research on the operating condition identification approach based on local impedance information.Firstly,according to the desired characteristics of the identification method,the peak impedance phase angle that can distinguish different operating states has been extracted from the EIS.Secondly,from both the equivalent circuit theory and the experimental data analysis,it is verified that there is a strong correlation between the peak impedance phase angle and the operating states.The peak impedance phase angle has been proved feasible to identify the operating states.Thirdly,the dynamic experiments of PEMFCs under different conditions are carried out using the peak impedance phase angle as the feedback of the operating state.Finally,based on the dynamic characteristics results,the peak impedance phase angle output model of the stack is designed with the support vector machine regression prediction method,and the output characteristics of this model has been investigated.3.Research on the closed-loop health management based on peak impedance phase angle.A closed-loop health management idea considering the regulation time cost and regulation sensitivity is proposed in order to face the practical application requirements of power systems.The time cost and regulation sensitivity parameters are obtained with the help of dynamic experimental data.The controller design is carried out using the multi-objective water circulation algorithm to solve the optimization target of minimal time cost and regulation amount.Based on the proposed peak impedance phase angle output model,a group of health management experiments have been carried out.The results show that the proposed method is effective in terms of the comprehensive optimization of the time cost and regulation amount.