This paper deals with a pattern-recognition-based diagnosis approach, which aim is to estimate the Fuel Cell (FC) operating time, and consequently its remaining duration life. With the method proposed, both static and dynamic information extracted from the stack (i.e. polarization curve records and Electrochemical Impedance Spectroscopy (EIS) measurements) can be used. The complete diagnosis method consists of several steps. First, features are extracted from EIS measurements and polarization curves independently. This enables us to simplify the extracted information without losing relevant information, and to remove noise. For the polarization curves, an empiric model is exploited to ensure the feature extraction. For the impedance spectra, both expert knowledge and parametric modeling are used to extract features. In particular, a latent regression model is used to split automatically the imaginary part of the spectra into several segments that are approximated by polynomials. The next step of the method consists in selecting the most relevant features from the whole set of extracted features. This helps us to estimate the operating time, while adjusting the complexity of the model. The final step of the approach is a linear regression that uses the selected subset of features to estimate the FC operating time. The performances of the proposed approach are evaluated on a dataset made up of EIS measurements and polarization curves extracted from two FC lifetime tests. A mean error of about 2 h over a global operating duration of 1000 h can be obtained. Moreover, the portability of the method is shown by considering another FC ageing test conducted on a different FC stack type
One of the limiting factors for the spreading of the fuel cell technology is the durability and researches to extend their lifetime are being done world-widely. We present here a pattern recognition approach aiming to estimate fuel cell operating time based on electrochemical impedance spectroscopy measurements. It consists in first extracting features from the impedance spectrum. For that purpose, two approaches have been investigated. In the first one, particular points of the spectrum are empirically extracted as features. In the second approach, a parametric modelling is performed to extract features from both the real and the imaginary parts of the impedance spectrum. In particular, a latent regression model is used to automatically split the spectrum into several segments that are approximated by polynomials. The number of segments is adjusted taking account the a priori knowledge about the physical behaviour of fuel cell components. Then, a linear regression model using different subsets of extracted features is employed for the estimation of fuel cell operating time. The performances of the proposed approach are evaluated on experimental data set to show its feasibility. Being able to estimate the fuel cell operating time, and consequently its remaining duration life, these results could lead to interesting perspectives for predictive maintenance policy of fuel cells.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.