Hierarchical fault diagnostic method for a polymer electrolyte fuel cell system

Hydrogen Leakage Diagnosis for Proton Exchange Membrane Fuel Cell Systems: Methods and Suggestions on Its Application in Fuel Cell Vehicles

“…The residuals, between the measured outputs and the outputs estimated by an observer of the faultless model, were used to calculate the fault sensitivities and the Euclidean distances corresponding to different faults, so as to realize the diagnosis of hydrogen leakage from the PEMFC hydrogen supply system. Li et al [37] proposed a hierarchical fault diagnostic method for PEMFC system. Firstly, faults were diagnosed at the subsystem level by the residuals between the model outputs and the measured outputs.…”

“…In addition, the differences between data-driven fault diagnosis and isolation system and hybrid fault diagnosis and isolation system are also analyzed in [48]. It can be seen from [35][36][37][38][39][40][41][42][43][44][45][46][47][48] that the model-based methods and data-driven methods can diagnose the gas leakage through the operating state data of the fuel cell system collected by the sensors. Therefore, they can be used to diagnose hydrogen leakage during the fuel cell system testing or FCVs running.…”

“…In the existing research, there are many diagnosis methods, either for the hydrogen leakage from the hydrogen supply system or for the hydrogen leakage in the stack. As shown in Figure 1, these methods used different strategies to diagnose hydrogen leakage, mainly including environmental hydrogen concentration diagnosis [24][25][26][27][28][29][30], hydrogen pressure decay diagnosis [9,10,[32][33][34]100,[112][113][114], gas leakage diagnosis by model-based methods or data-driven methods [36][37][38][39][40][41][42][43][44][45][46][47][48][49][50]115,123], cross-current diagnosis [10,[100][101][102][103], and diagnosis by stack (or cells) output voltage [10,100,[104][105][106][107][108]…”

Optimization of fuel cell thermal management system based on back propagation neural network

“…From 300-1600 seconds, the cooling water temperature is in the sound stage, and the inlet and outlet cooling water temperature of the electric reactor changes in the same trend, which indicates that the coupling effect between the pump and the radiator fan still exists. When the cooling water flow follows the current and power control strategy at the same time, the changing trend of inlet and outlet cooling water temperature and cooling water temperature difference of the power reactor is the most gentle and close to the control target value, and the oscillation phenomenon is weak at the maximum power [10]. The relative error value between the simulation data and the control target of the inlet and outlet cooling water temperature and temperature difference of the fuel cell reactor represents the fluctuation of the inlet and outlet cooling water temperature and temperature difference of the electric reactor.…”

Data-Driven Fault Diagnosis for Automotive PEMFC Systems Based on the Steady-State Identification

“…Through accurate fault diagnosis, relevant personnel can take corresponding measures to prevent more serious faults. At present, many scholars have done a lot of research on fault diagnosis methods of the PEMFC system [1,[3][4][5][6][7][8][9][10][11][12][13][14]. The proposed methods can be mainly categorized into two types [4]-model-based methods [3,[5][6][7][8][9][10] and data-driven methods [1,4,[11][12][13][14].…”