A novel method for polymer electrolyte membrane fuel cell fault diagnosis using 2D data

Liu, Zhongyong; Pei, Mengliu; He, Qingbo; Wu, Qiang; Jackson, Lisa M.; Mao, Lei

doi:10.1016/j.jpowsour.2020.228894

Cited by 34 publications

(4 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the measured maps, to reduce the error caused by continuous change of the current during scanning process, the maps were divided into sub‐regions for statistical analysis. For a magnetic field map of dimension m × n , its m ‐dimensional statistical feature vector was obtained by arithmetic averaging over each row of its magnetic field data [ 22 ]

\begin{matrix} x = (x_{1}, x_{2}, …, x_{m}) = (\frac{\sum_{k = 1}^{n} x_{1 k}}{n}, \frac{\sum_{k = 1}^{n} x_{2 k}}{n}, …, \frac{\sum_{k = 1}^{n} x_{m k}}{n}) \end{matrix}

…”

Section: Methodsmentioning

confidence: 99%

In Situ Detection of Lithium‐Ion Battery Pack Capacity Inconsistency Using Magnetic Field Scanning Imaging

et al. 2022

Self Cite

View full text Add to dashboard Cite

generate more heat and pose potential safety issues like thermal runaway. [4b,5] Therefore, detection and minimization of cell inconsistency within the battery pack is the key for guaranteeing the operation safety of lithium-ion battery.At the individual cell level, the most reliable and accurate capacity detection methods is coulomb counting during a complete charge/discharge cycle, which is difficult to implement in practice facing large-scale cells because of time consuming and costly. [6] To improve detection efficiency, most existing studies estimate cell capacity based on its charge/discharge voltage curve. In recent years, incremental capacity-differential voltage (IC-DV) analysis has become a popular method for evaluating battery capacity degradation. [6] For the battery pack with series-connection configuration, the IC-DV method can be implemented, despite the measuring error, because the voltage of an individual cell is available. [7] On the other side, for parallelconnected cells with the same terminal voltage, this method is no longer applicable to monitor individual cells. In the absence of individual cell current detection in commercial battery packs, cell-to-cell capacity inconsistency is difficult to observe and lack controllability. [1a,8] Although current imbalance between cells can be used for indicating battery pack capacity inconsistency, [9] the use of current sensor for each cell is impractical and expensive, hindering this method in practical applications. In addition, other tools of detecting unbalanced current in the pack are still lacking.Recently, several advanced imaging technologies have been proposed and applied for monitoring lithium-ion cell performance. [10] X-ray imaging was employed to characterize structural changes of cell materials by virtue of its deeper transmission depth. [10a-c] Neutron imaging was used to observe battery internal phenomena including Li-ion distribution, dendrite growth, and gas generation via absorbing neutrons by Li and scattering neutrons by H. [10d,e] Ultrasonic imaging was proposed to visualize electrolyte wetting process and estimate battery state of charge (SOC) through parameters such as the time of flight and amplitude of acoustic waves. [10f,g] With aforementioned imaging techniques, internal structure behavior of individual cell can be monitored and analyzed, but due to the limitations of the imaging principle, the applications of these imaging techniques in detecting cell capacity or current are restricted.In situ inside-out magnetic resonance imaging was proposed in 2018 by Ilott et al. for application to detect tiny induced One of the main obstacles for the reliability and safety of a lithium-ion battery pack is the difficulty in guaranteeing its capacity consistency at harsh operating conditions, while the key solution is accurate detection of cell capacity inconsistency within the battery pack without taking it apart for destructive testing. Here, an in situ and nondestructive technology is proposed for this purpose, by imaging the...

show abstract

\begin{matrix} x = (x_{1}, x_{2}, …, x_{m}) = (\frac{\sum_{k = 1}^{n} x_{1 k}}{n}, \frac{\sum_{k = 1}^{n} x_{2 k}}{n}, …, \frac{\sum_{k = 1}^{n} x_{m k}}{n}) \end{matrix}

…”

Section: Methodsmentioning

confidence: 99%

In Situ Detection of Lithium‐Ion Battery Pack Capacity Inconsistency Using Magnetic Field Scanning Imaging

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…In addition, the proposed can extensively boost the robustness and reliability of HVSR loose fault detection. It should be noted that the data preprocessing approaches, i.e., z-score normalization and sliding-window method [4][5] , are used to preprocess the raw data. The main functions of the HVSR-DenseNet model are as follows.…”

Section: The Proposed Hvsr-densenet Fault Diagnosis Modelmentioning

confidence: 99%

Enabling intelligent HVSR loose fault diagnosis based on vibration perception

Zhao,

Huang,

Qian

et al. 2023

J. Phys.: Conf. Ser.

Self Cite

View full text Add to dashboard Cite

The significance of high-voltage shunt reactors (HVSRs) in ensuring the reliability of power grids is widely acknowledged. Nevertheless, the persistent mechanical vibrations experienced by HVSR can loosen vital components such as windings, iron cores, and bolt fasteners. If the mechanical loose fault cannot be detected and mitigated timely, the vibration will be further aggravated, which tends to cause additional detects such as overheating and discharge, even fire disasters. Therefore, it is essential to diagnose loose faults in high-voltage shunt reactors. This study presents a unique data-driven method that utilizes optimal vibration test location selection and a densely connected neural network (DenseNet) for diagnosing loose faults in shunt reactors. Firstly, we integrate the least square method and R-square to explore the fault sensitivity of the vibration signal collected in different locations on the external tank surface, based on which the optimal vibration test points can be determined. Then the vibration data acquired in the selected points are input into the DenseNet model to achieve high-precise loose fault detection and diagnosis for high-voltage shunt reactor. Furthermore, with test data collected from a 35 kV high-voltage shunt reactor, the diagnostic performance of the proposed method is verified and highlighted.

show abstract

“…Another graphical approach for PEMFC cell failure diagnosis is proposed by [24] that analyzes the one-dimensional (1D) voltage data of a single cell and converts it into a two-dimensional (2D) image; optimal characteristics are determined by Fisher Discriminant Analysis (FDA). The K-means clustering method analyzes the data collected from two different single cells in flooding and dehydration failure states.…”

Section: Introductionmentioning

confidence: 99%

“…These techniques are applied to a single cell and are developed under controlled laboratory conditions, where temperature and pressures are managed. Some notable works in this area include: [6][7][8][13][14][15][16][17]24,25].…”

Section: Introductionmentioning

confidence: 99%

PEMFCs Model-Based Fault Diagnosis: A Proposal Based on Virtual and Real Sensors Data Fusion

Ariza,

Correcher,

Vargas-Salgado

2023

Sensors

View full text Add to dashboard Cite

Proton Exchange Membrane Fuel Cells (PEMFCs) are critical components in renewable hybrid systems, demanding reliable fault diagnosis to ensure optimal performance and prevent costly damages. This study presents a novel model-based fault diagnosis algorithm for commercial hydrogen fuel cells using LabView. Our research focused on power generation and storage using hydrogen fuel cells. The proposed algorithm accurately detects and isolates the most common faults in PEMFCs by combining virtual and real sensor data fusion. The fault diagnosis process began with simulating faults using a validated mathematical model and manipulating selected input signals. A statistical analysis of 12 residues from each fault resulted in a comprehensive fault matrix, capturing the unique fault signatures. The algorithm successfully identified and isolated 14 distinct faults, demonstrating its effectiveness in enhancing reliability and preventing performance deterioration or system shutdown in hydrogen fuel cell-based power generation systems.

show abstract

A novel method for polymer electrolyte membrane fuel cell fault diagnosis using 2D data

Cited by 34 publications

References 46 publications

In Situ Detection of Lithium‐Ion Battery Pack Capacity Inconsistency Using Magnetic Field Scanning Imaging

In Situ Detection of Lithium‐Ion Battery Pack Capacity Inconsistency Using Magnetic Field Scanning Imaging

Enabling intelligent HVSR loose fault diagnosis based on vibration perception

PEMFCs Model-Based Fault Diagnosis: A Proposal Based on Virtual and Real Sensors Data Fusion

Contact Info

Product

Resources

About