| abstract |
The invention discloses a mode identification fault diagnosis method of a proton exchange membrane fuel cell system. , hydrogen inlet pressure, air inlet pressure, compressor outlet pressure and other diagnostic variables, for the obtained diagnostic variables in the normal state and fault state, perform data normalization processing, establish an initialization sample set, and use PFCM‑OABC‑ The SVM combined pattern recognition algorithm performs fault diagnosis of the proton exchange membrane fuel cell system and determines the state of the proton exchange membrane fuel cell system. The present invention uses the PFCM‑OABC‑SVM combined pattern recognition algorithm to diagnose the fault of the proton exchange membrane fuel cell system, Realize data filtering, obtain optimal target parameters, accurately identify faults, and effectively ensure the normal operation of the system. |