Diagnosis for PEMFC Systems: A Data-Driven Approach With the Capabilities of Online Adaptation and Novel Fault Detection

Zhongliang Li (1) , Rachid Outbib (1) , Stefan Giurgea (2) , Daniel Hissel (2)

1 LSIS - Laboratoire des Sciences de l'Information et des Systèmes
2 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)

In this paper, a data-driven strategy is proposed for polymer electrolyte membrane fuel cell system diagnosis. In the strategy, features are first extracted from the individual cell voltages using Fisher discriminant analysis. Then, a classification method named spherical-shaped multiple-class support vector machine is used to classify the extracted features into various classes related to health states. Using the diagnostic decision rules, the potential novel failure mode can be also detected. Moreover, an online adaptation method is proposed for the diagnosis approach to maintain the diagnostic performance. Finally, the experimental data from a 40-cell stack are proposed to verify the approach relevance. Index Terms-Classification, data-driven diagnosis, feature extraction, novel fault detection, online adaptation, polymer electrolyte membrane fuel cell (PEMFC) systems.

Mots clés: classification; online adaptation; novel fault detection; feature extraction; PEMFC systems; data-driven diagnosis

Domaines: Automatique / Robotique; Energie électrique

Zhongliang Li, Rachid Outbib, Stefan Giurgea, Daniel Hissel. Diagnosis for PEMFC Systems: A Data-Driven Approach With the Capabilities of Online Adaptation and Novel Fault Detection. IEEE Transactions on Industrial Electronics, 2015, 62 (8), pp.5164-5174. ⟨10.1109/TIE.2015.2418324⟩. ⟨hal-02004105⟩