Online adaptive humidity monitoring method for proton exchange membrane fuel cell based on fuzzy C-means clustering and online sequence extreme learning machine. (10th October 2022)

• Record Type: Journal Article
• Authors: Dong, Qichao; Zhou, Yafu; Lian, Jing; Li, Linhui
• Is Part Of: Electrochimica acta. Volume 429(2022)
• Publication Date: 2022-10-10
• Subjects: PEMFC aging -- Adaptive humidity monitoring -- OS-ELM -- FCM -- Improved EIS impedance model; Electrochemistry -- Periodicals; Electrochemistry, Industrial -- Periodicals; 541.37
• DOI: 10.1016/j.electacta.2022.141059
• Languages: English
• ISSNs: 0013-4686

Abstract

Abstract: A method to enable online adaptive monitoring for proton exchange membrane fuel cell (PEMFC) humidity under different aging stages based on fuzzy C-mean clustering (FCM) and online sequential extreme learning machine (OS-ELM) is presented. The improved Randles EIS impedance model is served as the electrochemical analysis tool to determine the humidity characterization parameters under different humidity conditions, while the aging experimental data, revealing the aging characteristics, is explained for the evolution process of humidity characterization parameters by establishing the links between the EIS identified parameters and the aging times. Theoretical considerations, assumptions and methodology applications are discussed. The method is demonstrated by the resistance triplet characterization data of three typical aging stages, and a stable humidity condition recognition accuracy above 99% is shown to estimate the reliability of the developed model. In the aging experiment, the prediction accuracy of the traditional static model drops to 47.8% at aging time T =1016 h, however, the prediction accuracy of the proposed method can still retain 99.7%, and a better prediction performance is demonstrated cover the whole life cycle of PEMFC, which contributes to accelerating the commercialization process.