Online Diagnosis of PEM Fuel Cell by Fuzzy C-Means Clustering

• Published in: Reference Module in Earth Systems and Environmental Sciences
• Main Authors: Chanal, Damien, Yousfi Steiner, Nadia, Petrone, Raffaele, Chamagne, Didier, Péra, Marie-Cécile
• Format: Book Chapter
• Language: English
• Published: Elsevier Inc 2013
• Subjects: Clustering; Diagnosis; Electrochemical impedance spectroscopy (EIS); Fuel cells; Fuzzy C-means; Non model-based; Poisoning; Proton exchange membrane (PEM); Starvation; Water management; Non model-based; Starvation; Water management; Fuzzy C-means; Fuel cells; Proton exchange membrane (PEM); Poisoning; Diagnosis; Clustering; Electrochemical impedance spectroscopy (EIS)
• ISBN: 9780128010914; 0128010916; 0124095488; 9780124095489
• DOI: 10.1016/B978-0-12-819723-3.00099-8

To improve the performances and lifetime of Low Temperature Proton Exchange Membrane Fuel Cell (LT-PEMFC), a good monitoring is necessary to detect and recover faults. In this article an online diagnosis method able to detect the most frequently reported faults in LT-PEMFC, namely flooding, drying, starvation and poisoning is proposed. This method is based on the use of the Electrochemical Impedance Spectroscopy (EIS) as a monitoring tool and a fuzzy C-means classifier to automatically identify the faults at an early stage. To validate the proposed methodology and emphasize its genericity, the algorithm has been tested on data extracted from two different stack technologies: one fed with hydrogen and pure oxygen and another one with hydrogen and air. The experimental tests were carried out by two different laboratories using different test benches.