Performance degradation analysis and fault prognostics of solid oxide fuel cells using the data-driven method

Zhang, Xiaochen; He, Zhenyu; Zhan, Zhongliang; Han, Te

doi:10.1016/j.ijhydene.2021.01.126

Cited by 18 publications

(4 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generally, EMD sifting is iteratively implemented and stopped under the condition that a monotonic residual is obtained. The residual reflects the trend characteristics in the original signal [15]. The residual contributes to revealing the physical characteristics of the signal as it is the low frequency or null component [14].…”

Section: Health Indicator Extraction Modelmentioning

confidence: 99%

A hybrid method for remaining useful life prediction of fuel cells under variable loads

Wang

Outbib

et al. 2022

2022 10th International Conference on Systems and Control (ICSC)

View full text Add to dashboard Cite

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

Section: Health Indicator Extraction Modelmentioning

confidence: 99%

A hybrid method for remaining useful life prediction of fuel cells under variable loads

Wang

Outbib

et al. 2022

2022 10th International Conference on Systems and Control (ICSC)

View full text Add to dashboard Cite

show abstract

“…To tackle this challenge, data-driven methods are usually used to learn and intelligently provide valuable information from the current online sampling data and the large amount of historical offline data stored in the system [20], This avoids overdependence on the complex decay mechanism of the PEMFC [21]. The data-driven approaches can be utilized to monitor the health status of the PEMFC system by learning and training data [22][23][24][25][26]. Data-driven approaches include echo state network (ESN), long-and short-term memory network (LSTM), adaptive neuro-fuzzy inference system (ANFIS), nonlinear autoregressive exogenous (NARX), relevance vector machine (RVM), Gaussian process regression (GPR), extreme learning machine (ELM), and Digital twin.…”

Section: Data-driven Approachesmentioning

confidence: 99%

A Review of Life Prediction Methods for PEMFCs in Electric Vehicles

Tang

Yao

et al. 2022

Sustainability

View full text Add to dashboard Cite

The proton-exchange membrane fuel cell (PEMFC) has the advantage of high energy conversion efficiency, environmental friendliness, and zero carbon emissions. Therefore, as an attractive alternative energy, it is widely used in vehicles. Due to its high nonlinearity, strong time variation, and complex failure mechanisms, it is extremely difficult to predict PEMFC life in electric vehicles. The uncertainty of life predictions for the PEMFC limits its wide application. Since it is particularly important to accurately carry out PEMFC life predictions, significant research efforts are directed toward tackling this issue by adopting effective methods. In this paper, a number of PEMFC life prediction methods for electric vehicles are reviewed and summarized. The goal of this review is to render feasible and potential solutions for dealing with PEMFC life issues considering dynamic vehicle conditions. Based on this review, the reader can also easily understand the research status of PEMFC life prediction methods and this review lays a theoretical foundation for future research.

show abstract

“…Subotić ( 9) tailored three degradation mechanisms on a large-scale planer anode-supported SOFC, fuel starvation, air starvation and carbon deposition, and demonstrated the potential to identify the failure modes at their preliminary stage by applying the THDA method. Malafronte (10) embedded THDA in a 6 kW SOFC system to indicate stack fuel utilization with a frequency of 0.02 Hz and an amplitude of 2 A sinusoidal current corresponding to 6~15% of the DC current. Although the feasibility of THDA applied in SOFC has been verified preliminarily, the numbers of related studies are limited.…”

Section: Introductionmentioning

confidence: 99%

A Novel Online Diagnosis Method for Tubular Solid Oxide Fuel Cells Using Total Harmonic Distortion Analysis

Fan¹,

Wang²,

Wang³

et al. 2023

ECS Trans.

View full text Add to dashboard Cite

Fast and accurate online prognostic technology is urgently needed for the long-term and stable operation of solid oxide fuel cell (SOFCs). Total harmonic distortion analysis (THDA) is a potential candidate that obtains the dynamic information of an electrochemical system by analyzing high-order signals and has been successfully utilized in PEMFC fault diagnosis. In this study, the THDA method is utilized in a tubular SOFC under fuel and air starvation conditions at different hydrogen and air flow rates. The total harmonic distortion spectra are obtained in the frequency range of 0.1~100 Hz. The effects of different excitation current amplitudes on the results accuracy are further investigated. As a consequence, the excitation amplitude range, frequencies with high sensitivity and appropriate threshold are chosen to diagnose different faults efficiently.

show abstract

Performance degradation analysis and fault prognostics of solid oxide fuel cells using the data-driven method

Cited by 18 publications

References 45 publications

A hybrid method for remaining useful life prediction of fuel cells under variable loads

A hybrid method for remaining useful life prediction of fuel cells under variable loads

A Review of Life Prediction Methods for PEMFCs in Electric Vehicles

A Novel Online Diagnosis Method for Tubular Solid Oxide Fuel Cells Using Total Harmonic Distortion Analysis

Contact Info

Product

Resources

About