An improved neural network model for predicting the remaining useful life of proton exchange membrane fuel cells

Sun, Xilei; Xie, Mingke; Fu, Jianqin; Zhou, Feng; Liu, Jingping

doi:10.1016/j.ijhydene.2023.03.219

Cited by 18 publications

(6 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These outcomes are measured using the RMSE metric to highlight how well the models reduce the gap between the expected and targeted RUL. The results of the model proposed in the 2014 PHM Data Challenge Dataset are contrasted with various competing models such as Fusion [48], 1 input-ESN [49], 2 input-ESN [49], 3 input-ESN [49], SAE-DNN [50], ML-DNN [50], and RCLMA [51]. The superior efficacy of the system is illustrated through the presentation of findings, as evidenced by the RMSE values detailed in Table 5.…”

Section: Comparison Resultsmentioning

confidence: 99%

Enhancing Prognostics of PEM Fuel Cells with a Dual-Attention LSTM Network for Remaining Useful Life Estimation: A Deep Learning Model

Darwish

2024

Sustain. Mach. Intell. J.

View full text Add to dashboard Cite

The Proton Exchange Membrane Fuel Cell (PEMFC) presents itself as a viable and effective technology to consider for transportation purposes. An essential aspect in the realm of electric vehicles is the crucial evaluation of the deterioration of the PEMFC stack. This paper proposed a data-driven deep learning framework that combines a Long Short-Term Memory (LSTM), self-attention, and scaled dot-product attention mechanism in order to enhance the precision of Remaining Useful Life (RUL) prediction for Proton Exchange Membrane Fuel Cells (PEMFCs). The LSTM enables the model to comprehend intricate temporal patterns and produce more abstract data representations. The self-attention mechanism is utilized to detect correlations among various time points. The scaled dot-product attention mechanism is employed to focus on the most crucial features. The effectiveness of the proposed model was evaluated by utilizing the 2014 PHM Data Challenge Dataset and comparing it with different Deep Learning models to showcase its efficacy. The findings from the experiments suggest that DA-LSTM proves to be a reliable choice for predicting the RUL of PEMFCs, as it demonstrated superior performance compared to all other models examined.

show abstract

Section: Comparison Resultsmentioning

confidence: 99%

Enhancing Prognostics of PEM Fuel Cells with a Dual-Attention LSTM Network for Remaining Useful Life Estimation: A Deep Learning Model

Darwish

2024

Sustain. Mach. Intell. J.

View full text Add to dashboard Cite

show abstract

“…In order to evaluate the model performance, the selected evaluation metrics used were mean absolute error (MAE)and root mean square error (RMSE). In addition, this study introduced a key performance evaluation indicator, percentage error (%Er FT ), determined by calculating the percentage difference between the real remaining useful life (RUL act ) and the predicted remaining useful life (RUL pre ) [21]. This indicator directly reflects the degree of variation between the estimated and actual values.…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…These thresholds represent the RUL when device performance drops to varying degrees, providing a comprehensive evaluation of the predictive performance of the model under different health states. By calculating %Er FT under these different fault thresholds, we can more accurately understand the strengths and weaknesses of the model in predicting device health conditions [21].…”

Section: Evaluation Metricsmentioning

confidence: 99%

TabNet: Locally Interpretable Estimation and Prediction for Advanced Proton Exchange Membrane Fuel Cell Health Management

Zhang,

Jin,

Liang

et al. 2024

Electronics

View full text Add to dashboard Cite

In the pursuit of advanced Predictive Health Management (PHM) for Proton Exchange Membrane Fuel Cells (PEMFCs), conventional data-driven models encounter considerable barriers due to data reconstruction resulting in poor data quality, and the complexity of models leading to insufficient interpretability. In addressing these challenges, this research introduces TabNet, a model aimed at augmenting predictive interpretability, and integrates it with an innovative data preprocessing technique to enhance the predictive performance of PEMFC health management. In traditional data processing approaches, reconstruction methods are employed on the original dataset, significantly reducing its size and consequently diminishing the accuracy of model predictions. To overcome this challenge, the Segmented Random Sampling Correction (SRSC) methodology proposed herein effectively eliminates noise from the original dataset whilst maintaining its effectiveness. Notably, as the majority of deep learning models operate as black boxes, it becomes challenging to identify the exact factors affecting the Remaining Useful Life (RUL) of PEMFCs, which is clearly disadvantageous for the health management of PEMFCs. Nonetheless, TabNet offers insights into the decision-making process for predicting the RUL of PEMFCs, for instance, identifying which experimental parameters significantly influence the prediction outcomes. Specifically, TabNet’s distinctive design employs sequential attention to choose features for reasoning at each decision-making step, not only enhancing the accuracy of RUL predictions in PEMFC but also offering interpretability of the results. Furthermore, this study utilized Gaussian augmentation techniques to boost the model’s generalization capability across varying operational conditions. Through pertinent case studies, the efficacy of this integrated framework, merging data processing with the TabNet architecture, was validated. This work not only evidences that the effective data processing and strategic deployment of TabNet can markedly elevate model performance but also, via a visual analysis of the parameters’ impact, provides crucial insights for the future health management of PEMFCs.

show abstract

“…4.2.6. ML in the Field of CL Machine learning approaches have been used in CL for feature extraction [244], optimization [58,147,245,246], predicting performance [247], and degradation of CL on PEMFC [156,[248][249][250][251][252]. Wang et al [244] implemented deep learning super-resolution and multi-label segmentation to process the images from X-ray micro-computed tomography, followed by LBM with multi-relaxation time (MRT) for water management modeling.…”

Section: Issues Related To State-of-art CL Modelingmentioning

confidence: 99%

“…Among the compositions, the volume fraction of dry ionomers has ben proven to be the most sensitive parameter. Data-driven ML has also been employed to predict performance reductions resulting from PEMFC degradation [248,251,252,254], although it cannot be clearly pinpointed whether it is due to membrane, CL, or GDL. Pt loss and reorganization are critical factors resulting from the high temperature, humidity, and load cycling [255].…”

Section: Issues Related To State-of-art CL Modelingmentioning

confidence: 99%

Towards Reliable Prediction of Performance for Polymer Electrolyte Membrane Fuel Cells via Machine Learning-Integrated Hybrid Numerical Simulations

Kaiser,

Ahn,

Kim

et al. 2024

Processes

View full text Add to dashboard Cite

For mitigating global warming, polymer electrolyte membrane fuel cells have become promising, clean, and sustainable alternatives to existing energy sources. To increase the energy density and efficiency of polymer electrolyte membrane fuel cells (PEMFC), a comprehensive numerical modeling approach that can adequately predict the multiphysics and performance relative to the actual test such as an acceptable depiction of the electrochemistry, mass/species transfer, thermal management, and water generation/transportation is required. However, existing models suffer from reliability issues due to their dependency on several assumptions made for the sake of modeling simplification, as well as poor choices and approximations in material characterization and electrochemical parameters. In this regard, data-driven machine learning models could provide the missing and more appropriate parameters in conventional computational fluid dynamics models. The purpose of the present overview is to explore the state of the art in computational fluid dynamics of individual components of the modeling of PEMFC, their issues and limitations, and how they can be significantly improved by hybrid modeling techniques integrating with machine learning approaches. Furthermore, a detailed future direction of the proposed solution related to PEMFC and its impact on the transportation sector is discussed.

show abstract

An improved neural network model for predicting the remaining useful life of proton exchange membrane fuel cells

Cited by 18 publications

References 40 publications

Enhancing Prognostics of PEM Fuel Cells with a Dual-Attention LSTM Network for Remaining Useful Life Estimation: A Deep Learning Model

Enhancing Prognostics of PEM Fuel Cells with a Dual-Attention LSTM Network for Remaining Useful Life Estimation: A Deep Learning Model

TabNet: Locally Interpretable Estimation and Prediction for Advanced Proton Exchange Membrane Fuel Cell Health Management

Towards Reliable Prediction of Performance for Polymer Electrolyte Membrane Fuel Cells via Machine Learning-Integrated Hybrid Numerical Simulations

Contact Info

Product

Resources

About