Proton Exchange Membrane Fuel Cell Prognostics Using Genetic Algorithm and Extreme Learning Machine

Abstract

The prognostics can predict the degradation of proton exchange membrane fuel cell (PEMFC) to formulate a reasonable maintenance plan for improving its lifetime and performance. In this paper, the voltage degradation for PEMFC at different conditions is predicted, by using a novel prognostics method based on genetic algorithm (GA) and extreme learning machine (ELM). The novel prognostics method considers the effects of the PEMFC load current, relative humidity, hydrogen pressure, and temperature on the degradation for PEMFC. Firstly, the voltage degradation prediction model for PEMFC is built by the ELM. Then, the GA is adopted to determine the optimal parameter of the proposed degradation prediction model. Finally, the voltage degradation prediction of the proposed prognostics method is validated, using data derived from the PEMFC in actual postal fuel cell electric vehicle (PFCEV) at real conditions and PEMFC at dynamic load current. The experimental results show that the proposed prognostics method can obtain good voltage degradation prediction for PEMFC in PFCEV at real conditions. The proposed method achieves better voltage degradation prediction for PEMFC at dynamic load current than other traditional data‐based prognostics methods.

Performance prediction of fuel cells using long short‐term memory recurrent neural network

“…Compared to model‐driven methods, data‐driven methods use experimental data obtained from the operations of PEMFC to predict the performance of PEMFC, without requiring understanding of physicochemical mechanism of fuel cell. A variety of different data‐driven models, such as artificial neural network (ANN), support vector machine (SVM), and Adaptive Neuro‐Fuzzy Inference System (ANFIS) have been employed to PEMFC modeling 9‐24 . Saengrung et al 12 constructed two ANNs including the back‐propagation (BP) and radial basis function (RBF) networks to predict the performance of a commercial PEMFC system.…”

“…Pan et al 20 proposed a prediction method based on improved SVM to predict the utilization of a catalyst of fuel cells. Chen et al 21 developed a prognostics method using genetic algorithm (GA) and extreme learning machine (ELM) to predict the voltage degradation of fuel cell. The operation conditions including fuel cell temperature, relative humidity, hydrogen pressure, and current are considered in the method.…”

Proton exchange membrane fuel cell degradation model based on catalyst transformation theory

“…Currently, the high cost and short durability of the PEMFC system are the main barriers to commercial applications [5]. For durability problem of the PEMFC system, the prognostics and health management (PHM) has played a vital role in recent years [6], as it can predict and prevent the failure before it happens [7]. The prediction of the remaining useful life (RUL) is the main aim of the prognosis of the PEMFC system.…”

Data-Driven Prognostics of the SOFC System Based on Dynamic Neural Network Models

“…The PHM of the SOFC system uses real measurement data to predict the health degradation trend and estimate the residual of life or remaining useful life (RUL) to effectively prolong the lifespan of the SOFC system. In the process of PHM, prognostics technology plays an important role in predicting future conditions, and for RUL to be functional [6].…”