Long-Term Performance Prediction of PEMFC Based on LASSO-ESN

In recent years, with wide application of proton exchange membrane fuel cell (PEMFC) in vehicles and portable applications, researches regarding PEMFC lifetime behavior associated prognostic techniques receive more interest. this article, a least absolute shrinkage selection operator-echo state network (LASSO-ESN)-based strategy is proposed for the optimization input parameters long-term prediction. analysis, ESN selected to predict iteratively, while are optimized using LASSO. With LASSO, contribution prediction can be evaluated, those minimum weight eliminated iteratively during From findings, most accurate predictions corresponding determined. Furthermore, effectiveness investigated data at different operating conditions. Results demonstrate that strategy, conditions determined, provided.