Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks
This paper proposes a nonlinear dynamic recurrent neural network (DRNN) prognostics method for predicting the performance degradation trend and estimating the remaining useful life (RUL) of a proton exchange membrane fuel cell (PEMFC). The conducted DRNN prognostic methods are based on a nonlinear a...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-11-01
|
| Series: | Energy Reports |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352484722012987 |
| _version_ | 1797901960734572544 |
|---|---|
| author | Te-Jen Chang Shan-Jen Cheng Chang-Hung Hsu Jr-Ming Miao Shih-Feng Chen |
| author_facet | Te-Jen Chang Shan-Jen Cheng Chang-Hung Hsu Jr-Ming Miao Shih-Feng Chen |
| author_sort | Te-Jen Chang |
| collection | DOAJ |
| description | This paper proposes a nonlinear dynamic recurrent neural network (DRNN) prognostics method for predicting the performance degradation trend and estimating the remaining useful life (RUL) of a proton exchange membrane fuel cell (PEMFC). The conducted DRNN prognostic methods are based on a nonlinear autoregressive neural network (NARNN) model and nonlinear autoregressive with exogenous inputs neural network (NARXNN) model, with the goal of taking actions to extend the life of the PEMFC. To effectively remove data noise or spikes and provide smooth reconstruction, a locally weighted regression (LWR) method is used for pre-processing. The DRNN model parameters for the time lags are processed by an autocorrelation function (ACF). For comparing the predictions within the different DRNN models, the data sets are divided into three different prognostic configurations for predicting and estimating the remaining life of the PEMFC. The results show that the NARXNN model has the best predictions for the degradation trend and accurate remaining useful life, i.e. better than the NARNN model. Finally, the prognostic capability of the proposed DRNN method is compared with those in the literature. |
| first_indexed | 2024-04-10T09:10:15Z |
| format | Article |
| id | doaj.art-bb425c133fc74c0db6df1c904b513a40 |
| institution | Directory Open Access Journal |
| issn | 2352-4847 |
| language | English |
| last_indexed | 2024-04-10T09:10:15Z |
| publishDate | 2022-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Energy Reports |
| spelling | doaj.art-bb425c133fc74c0db6df1c904b513a402023-02-21T05:12:23ZengElsevierEnergy Reports2352-48472022-11-01894419452Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networksTe-Jen Chang0Shan-Jen Cheng1Chang-Hung Hsu2Jr-Ming Miao3Shih-Feng Chen4Department of Electrical and Electronic Engineering, Chung Cheng Institute of Technology, National Defense University, Taoyuan 335, Taiwan, ROCDepartment of Mechanical Engineering, Asia Eastern University of Science and Technology, New Taipei 220, Taiwan, ROC; Corresponding author.Department of Mechanical Engineering, Asia Eastern University of Science and Technology, New Taipei 220, Taiwan, ROCDepartment of Biomechatronics Engineering, National Pingtung University of Science and Technology, Pingtung 91201, Taiwan, ROCDepartment of Mechanical Engineering, Lunghwa University of Science and Technology, Taoyuan 333, Taiwan, ROCThis paper proposes a nonlinear dynamic recurrent neural network (DRNN) prognostics method for predicting the performance degradation trend and estimating the remaining useful life (RUL) of a proton exchange membrane fuel cell (PEMFC). The conducted DRNN prognostic methods are based on a nonlinear autoregressive neural network (NARNN) model and nonlinear autoregressive with exogenous inputs neural network (NARXNN) model, with the goal of taking actions to extend the life of the PEMFC. To effectively remove data noise or spikes and provide smooth reconstruction, a locally weighted regression (LWR) method is used for pre-processing. The DRNN model parameters for the time lags are processed by an autocorrelation function (ACF). For comparing the predictions within the different DRNN models, the data sets are divided into three different prognostic configurations for predicting and estimating the remaining life of the PEMFC. The results show that the NARXNN model has the best predictions for the degradation trend and accurate remaining useful life, i.e. better than the NARNN model. Finally, the prognostic capability of the proposed DRNN method is compared with those in the literature.http://www.sciencedirect.com/science/article/pii/S2352484722012987PrognosticsProton exchange membrane fuel cellDynamic recurrent neural networksNonlinear autoregressive neural networkNonlinear autoregressive with exogenous inputs neural network |
| spellingShingle | Te-Jen Chang Shan-Jen Cheng Chang-Hung Hsu Jr-Ming Miao Shih-Feng Chen Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks Energy Reports Prognostics Proton exchange membrane fuel cell Dynamic recurrent neural networks Nonlinear autoregressive neural network Nonlinear autoregressive with exogenous inputs neural network |
| title | Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks |
| title_full | Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks |
| title_fullStr | Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks |
| title_full_unstemmed | Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks |
| title_short | Prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks |
| title_sort | prognostics for remaining useful life estimation in proton exchange membrane fuel cell by dynamic recurrent neural networks |
| topic | Prognostics Proton exchange membrane fuel cell Dynamic recurrent neural networks Nonlinear autoregressive neural network Nonlinear autoregressive with exogenous inputs neural network |
| url | http://www.sciencedirect.com/science/article/pii/S2352484722012987 |
| work_keys_str_mv | AT tejenchang prognosticsforremainingusefullifeestimationinprotonexchangemembranefuelcellbydynamicrecurrentneuralnetworks AT shanjencheng prognosticsforremainingusefullifeestimationinprotonexchangemembranefuelcellbydynamicrecurrentneuralnetworks AT changhunghsu prognosticsforremainingusefullifeestimationinprotonexchangemembranefuelcellbydynamicrecurrentneuralnetworks AT jrmingmiao prognosticsforremainingusefullifeestimationinprotonexchangemembranefuelcellbydynamicrecurrentneuralnetworks AT shihfengchen prognosticsforremainingusefullifeestimationinprotonexchangemembranefuelcellbydynamicrecurrentneuralnetworks |