Fault-tolerant method of open-cathode PEMFC based on adaptive strong tracking Kalman filter combined with Hampel algorithm

Abstract

Control systems of open-cathode proton exchange membrane fuel cell (PEMFC) rely heavily on the stability of sensors, especially temperature control. Few targeted solutions have been developed for problems such as sensor data anomalies caused by sudden current and voltage changes during current pulses. This can cause the tracking trajectory in the temperature control system to deviate from the correct target trajectory and can easily lead to thermal runaway and severe damage to the stack. In this paper, a temperature fault-tolerant method based on Adaptive Strong Tracking Kalman Filter (ASTKF) is proposed, considering a series of problems such as the erroneous control caused by sensor failure and the strong nonlinearity of the system caused by external current loading. Combined with Hampel algorithm, the stability of the control system is improved effectively. The experimental data is used to verify that the sudden-change data and slow-change data caused by current pulses during system operation are well processed. Finally, the practical application scheme of the proposed fault-tolerant method is given. The proposed method can improve the accuracy, stability and real-time performance of the control system. The noise reduction and anti-interference ability of sensor data are significantly improved.