Proton exchange membrane fuel cell degradation prediction based on Adaptive Neuro-Fuzzy Inference Systems

Abstract

This paper studies the prediction of the output voltage reduction caused by degradation during nominal operating condition of a PEM fuel cell stack. It proposes a methodology based on Adaptive Neuro-Fuzzy Inference Systems (ANFIS) which use as input the measures of the fuel cell output voltage during operation. The paper presents the architecture of the ANFIS and studies the selection of its parameters. As the output voltage cannot be represented as a periodical signal, the paper proposes to predict its temporal variation which is then used to construct the prediction of the output voltage. The paper also proposes to split this signal in two components: normal operation and external perturbations. The second component cannot be predicted and then it is not used to train the ANFIS. The performance of the prediction is evaluated on the output voltage of two fuel cells during a long term operation (1000 hours). Validation results suggest that the proposed technique is well adapted to predict degradation in fuel cell systems.

Application of the Sensor Selection Approach in Polymer Electrolyte Membrane Fuel Cell Prognostics and Health Management

“…Data-driven approaches are more widely used for fuel cell diagnostics, that is, extracting the features by applying signal processing techniques to the sensor data, and discriminating fuel cell faults with extracted features [9][10][11][12][13]. Compared to fuel cell diagnostics, fewer studies have been devoted to fuel cell prognostics, and among these studies, training data from a fuel cell system is required to generate the input-output relationship of the fuel cell model for the prediction of future performance [14][15][16][17][18].…”

Proton exchange membrane fuel cell behavioral model suitable for prognostics

“…The black box modeling has already been investigated [21,31]. The physical based approach is developed with this first step of modeling.…”

Adaptive Prognostic of Fuel Cells by Implementing Ensemble Echo State Networks in Time-Varying Model Space

“…In the framework, several studies based on machine learning techniques or other data-based techniques are proposed to trace the stack voltage evolution. In [8], an adaptive Neuro-Fuzzy inference approach is proposed to predict the temporal variation of stack voltage. In [9], [10], two alternatives of Neuro Networks (NNs), named extreme learning machine (ELM) and echo state networks (ESN), are applied for the same purpose.…”