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

Li, Zhongliang; Zheng, Zhixue; Outbib, Rachid

doi:10.1109/tie.2019.2893827

Cited by 86 publications

(43 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The prognostic method can estimate the degradation trend during ∼700 h but it cannot track the dynamic ageing presented by the stack. In [54], the same authors propose to improve their prediction results, without modifying the LPV model and the SoH index, using an ensemble ESN instead of a conventional ESN. The method presents an acceptable (within the prediction boundaries defined by the authors) RUL prediction during the last 350 h of the experimental test.…”

Section: Prognostic Methods Under Alcmentioning

confidence: 99%

Prognostic methods for proton exchange membrane fuel cell under automotive load cycling: a review

Jacome

Hissel

Heiries³

et al. 2020

IET Electrical Systems in Transportation

View full text Add to dashboard Cite

Section: Prognostic Methods Under Alcmentioning

confidence: 99%

Prognostic methods for proton exchange membrane fuel cell under automotive load cycling: a review

Jacome

Hissel

Heiries³

et al. 2020

IET Electrical Systems in Transportation

View full text Add to dashboard Cite

“…When FCs are used in dynamic load scenarios such as electric vehicles, the start-stop process of FCs will cause their stack voltage degradation trend to show large fluctuations in a short period of time [8]. Although LSTMs can selectively forget the unnecessary voltage fluctuations in the long-term change trend of FCs stack voltage during time series processing, the large-scale fluctuations of the shortterm voltage mentioned above are still easy to cause LSTMs to fall into the error of local overfitting.…”

Section: The Prognostic Strategies Of Nsd-lstmsmentioning

confidence: 99%

“…These multi-step models will give n different voltage degradation prediction results. Assuming that the predictions of multiple NSD-LSTMs models follow the standard Gaussian distribution, the confidence interval (CI) is configured as 95% probability interval [8], and the average value of the predicted values is regarded as the final predicted voltage degradation trend. The RUL is estimated by comparing the predicted degradation trend with a predetermined failure threshold.…”

Section:  Training Lstmsmentioning

confidence: 99%

“…This article sets two evaluation criteria, the root-mean-square error (RMSE) and the relative error (RE), where RMSE evaluates the training effect of the LSTMs model in the offline phase, and RE is used in the online phase to evaluate the accuracy of the RUL given by the prognostic strategy. The mathematical description of RMSE and RE can be found in equations (8)…”

Section: Evaluation Criteriamentioning

confidence: 99%

“…Especially, inspired by the recent remarkable advances in deep neural networks (NNs), special attention has been put on the development of data-driven prognostic tools by configuring and training different NNs structures. Li et al [8] used a low-power PEMFCs to conduct long-term experiments in different operating modes, and extracted health indicators (HI) using a sliding adaptive data prediction strategy. The echo state network (ESN) is used to predict the degradation trend and estimate the RUL.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Proton Exchange Membrane Fuel Cells Prognostic Strategy Based on Navigation Sequence Driven Long Short-term Memory Networks

Wang

Outbib

et al. 2020

IECON 2020 the 46th Annual Conference of the IEEE Industrial Electronics Society

Self Cite

View full text Add to dashboard Cite

The prognostic of proton exchange membrane fuel cells (PEMFCs) degradation and the estimation of its remaining useful life (RUL) are effective ways to improve the reliability of the target system and reduce maintenance costs, which is of great significance for the wide commercialization of PEMFCs. Many factors cause the degradation of PEMFCs, and these factors are often difficult to measure accurately. The prognostic method based on long short-term memory networks (LSTMs) has better memory ability for time series and has been demonstrated able to describe the degradation trend of PEMFCs. However, the traditional LSTM prediction algorithm seems to easily fall into the local optimal solution in long-term prediction cases. Overfitting like errors may result in an imprecise or even unstable prognostic. This paper proposes a novel method, named navigation sequence driven LSTMs (NSD-LSTMs), to enhance the accuracy of PEMFCs degradation trend prediction. Two types of PEMFCs aging test data under different load conditions were used to verify the performance of NSD-LSTMs. Experimental results show that, compared with traditional LSTMs, NSD-LSTMs can improve the accuracy of trend prediction. Accurate degradation prognostic can be used to predict RUL and provide guidance for the commercial application of PEMFCs.

show abstract

Enhancing Time Series Predictability via Structure‐Aware Reservoir Computing

Guo,

Guan,

Leng

2024

Advanced Intelligent Systems

View full text Add to dashboard Cite

Accurate prediction of the future evolution of observational time series is a paramount challenge in current data‐driven research. While existing techniques struggle to learn useful representations from the temporal correlations, the high dimensionality in spatial domain is always considered as obstacle, leading to the curse of dimensionality and excessive resource consumption. This work designs a novel structure‐aware reservoir computing aiming at enhancing the predictability of coupled time series, by incorporating their historical dynamics as well as structural information. Paralleled reservoir computers with redesigned mixing inputs based on spatial relationships are implemented to cope with the multiple time series, whose core idea originates from the principle of the celebrated Granger causality. Representative numerical simulations and comparisons demonstrate the superior performance of the approach over the traditional ones. This work provides valuable insights into deeply mining both temporal and spatial information to enhance the representation learning of data in various machine learning techniques.

show abstract

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

Cited by 86 publications

References 22 publications

Prognostic methods for proton exchange membrane fuel cell under automotive load cycling: a review

Prognostic methods for proton exchange membrane fuel cell under automotive load cycling: a review

Proton Exchange Membrane Fuel Cells Prognostic Strategy Based on Navigation Sequence Driven Long Short-term Memory Networks

Enhancing Time Series Predictability via Structure‐Aware Reservoir Computing

Contact Info

Product

Resources

About