Fault diagnosis of proton exchange membrane fuel cells based on MVMD and ISCSO-HKELM

doi:10.3969/j.issn.2097-0706.2024.12.003

Abstract

Abstract:

To address the issue of low fault diagnostic accuracy in proton exchange membrane fuel cell（PEMFC） systems caused by high temperatures and strong background noise in raw signals，a fault diagnosis model based on multivariate variational mode decomposition（MVMD） and improved sand cat swarm optimization（ISCSO） algorithm for optimizing the hybrid kernel extreme learning machine（HKELM） was proposed. Wavelet hard thresholding denoising（WHTD） was applied to filter the raw PEMFC signals，and MVMD was employed to decompose the denoised signals into a series of intrinsic mode functions（IMFs）.The optimal IMFs were selected for signal reconstruction based on variance contribution rate， correlation coefficient，and information entropy.The sand cat swarm optimization（SCSO） algorithm was enhanced using logistic mapping，refracted opposition-based learning（ROBL），nonlinear dynamic factors，and golden sine strategy，resulting in an improved ISCSO algorithm. The ISCSO algorithm was then applied to optimize HKELM，and the improved ISCSO-HKELM algorithm was utilized to extract features from the reconstructed signals for fault diagnosis.The proposed WHTD-MVMD-ISCSO-HKELM fault diagnosis model was compared with other algorithms，and the experimental results demonstrated that the proposed approach significantly improved fault diagnostic accuracy，indicating feasibility and superiority.

Key words: hybrid kernel extreme learning machine, improved sand cat swarm optimization algorithm, multivariate variational mode decomposition, proton exchange membrane fuel cell, fault diagnosis

CLC Number:

TK01：TP181

DU Dongsheng, LIAN He, DENG Xiangshuai, REN Yiming, ZHAO Zhemin. Fault diagnosis of proton exchange membrane fuel cells based on MVMD and ISCSO-HKELM[J]. Integrated Intelligent Energy, 2024, 46(12): 17-28.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Table 1

Fig.9

Fig.10

Table 2

Fig.11

References 32

[1]	ZHANG W J, WANG Y L. Modification and durability of carbon paper gas diffusion layer in proton exchange membrane fuel cell[J]. Ceramics International, 2023, 49(6):9371-9381.
[2]	殷卓成, 王贺, 段文益, 等. 氢燃料电池汽车关键技术研究现状与前景分析[J]. 现代化工, 2022, 42(10):18-23. doi: 10.16606/j.cnki.issn0253-4320.2022.10.004
	YIN Zhuocheng, WANG He, DUAN Wenyi, et al. Research status and prospect of key technologies for hydrogen fuel cell vehicle[J]. Modern Chemical Industry, 2022, 42(10):18-23. doi: 10.16606/j.cnki.issn0253-4320.2022.10.004
[3]	LU J J, GAO Y, ZHANG L Y, et al. A novel dynamic radius support vector data description based fault diagnosis method for proton exchange membrane fuel cell systems[J]. International Journal of Hydrogen Energy, 2022, 47(84):35825-35837.
[4]	LI M F, CHEN Z P, DONG J B, et al. A data-driven fault diagnosis method for solid oxide fuel cell systems[J]. Energies, 2022, 15(7):2556-2556.
[5]	刘佳昕, 周风波. EMD算法的改进及在信号去噪中的应用[J]. 电子制作, 2024, 32(4):73-75.
[6]	LIU Z Z, DING K, LIN H B, et al. A novel impact feature extraction method based on EMD and sparse decomposition for gear local fault diagnosis[J]. Machines, 2022, 10(4):242.
[7]	PATRICK F, GABRIEL R, PAULO G. Empirical mode decomposition as a filter bank[J]. IEEE Signal Processing Letters, 2004, 11(2):112-114.
[8]	杨茂, 陈郁林. 基于EMD分解和集对分析的风电功率实时预测[J]. 电工技术学报, 2016, 31(21):86-93.
	YANG Mao, CHEN Yulin. Real-time prediction for wind power based on EMD and set pair analysis[J]. Transactions of China Electrotechnical Society, 2016, 31(21):86-93.
[9]	LU S, WANG J H, XUE Y G. Study on multi-fractal fault diagnosis based on EMD fusion in hydraulic engineering[J]. Applied Thermal Engineering, 2016,103:798-806.
[10]	DRAGOMIRETSKIY K, ZOSSO D. Variational moded ecomposetion[J]. IEEE Transactions on Signal Processing, 2014, 62(3):531-544.
[11]	崔素晓, 崔彦平, 武哲, 等. 基于MVMD-MOMEDA的齿轮箱故障诊断方法[J]. 河北科技大学学报, 2023, 44(6):551-561.
	CUI Suxiao, CUI Yanping, WU Zhe, et al. Gearbox fault diagnosis method based on MVMD-MOMEDA[J]. Journal of Hebei University of Science and Technology, 2023, 44(6):551-561.
[12]	GHANBARI E, AVAR A. Short-term wind power forecasting using the hybrid model of multivariate variational mode decomposition (MVMD) and long short-term memory (LSTM) neural networks[J]. Electrical Engineering, 2024(prepublish):1-31.
[13]	XU R L, ZHENG J Y, MEI F, et al. Short-Term Photovoltaic Power Generation Based on MVMD Feature Extraction and Informer Model[J]. Applied Sciences, 2024, 14(14):6279-6279.
[14]	曾海, 许德章. 基于模仿学习的气管插管机器人非结构环境作业策略[J]. 淮阴工学院学报, 2022, 31(3):31-40.
	ZENG Hai, XU Dezhang. Unstructured environment operation strategy of tracheal intubation robot based on imitation learning[J]. Journal of Huaiyin Institute of Technology, 2022, 31(3):31-40.
[15]	李冬玉, 娄柯, 尹杰, 等. 基于自适应ASO算法的微电网优化调度[J]. 淮阴工学院学报, 2023, 32(5):50-55.
	LI Dongyu, LOU Ke, YIN Jie, et al. Optimal dispatching of microgrid based on adaptive ASO algorithm[J]. Journal of Huaiyin Institute of Technology, 2023, 32(5):50-55.
[16]	何坤敏, 王霄, 杨靖, 等. 基于RF特征优选和WOA-ELM的风电齿轮箱故障诊断[J]. 电子测量技术, 2023, 46(5):57-64.
	HE Kunmin, WANG Xiao, YANG Jing, et al. Fault diagnosis of wind turbine gearbox based on RF feature optimization and WOA-ELM[J]. Electronic Measurement Technology, 2023, 46(5):57-64.
[17]	CHENG M H, JIAO L, YAN P, et al. Prediction and evaluation of surface roughness with hybrid kernel extreme learning machine and monitored tool wear[J]. Journal of Manufacturing Processes, 2022,84:1541-1556.
[18]	WANG J S, CHUANG F C. An accelerometer-based digital pen with a trajectory recognition algorithm for handwritten digit and gesture recognition[J]. IEEE Transactions on Industrial Electronics, 2012, 59(7):2998-3007.
[19]	HUANG G, SONG S J, GUPTA N, et al. Semi-supervised and unsupervised extreme learning machines.[J]. IEEE transactions on cybernetics, 2014, 44(12): 2405-2417. doi: 10.1109/TCYB.2014.2307349 pmid: 25415946
[20]	WANG S D, LIU Z B, JIA Z, LI Z H, et al. Incipient fault diagnosis of analog circuit with ensemble HKELM based on fused multi-channel and multi-scale features[J]. Engineering Applications of Artificial Intelligence, 2023,117:1-22
[21]	VAHID A T, AMIR S, JAWAD R, et al. Sand cat swarm optimization-based feedback controller design for nonlinear systems[J]. Heliyon, 2023, 9(3):13885-13885.
[22]	高佰宏, 刘朝晖, 刘华. 基于SCSO-GRU模型的网络流量预测[J]. 计算机与现代化, 2020(4):72-77.
	GAO Baihong, LIU Zhaohui, LIU Hua. Network traffic prediction based on SCSO-GRU model[J]. Computer and Modernization, 2020(4):72-77
[23]	TÜBER K, PÓCZA D, HEBLING C. Visualization of water buildup in the cathode of a transparent PEM fuel cell[J]. Journal of Power Sources, 2003, 124(2):403-414.
[24]	马杭, 陆文总, 耿世宇, 等. 基于改进阈值函数的小波降噪方法研究[J]. 激光杂志, 2023, 44(10):19-24.
	MA Hang, LU Wenzong, GENG Shiyu, et al. Research on Wavelet denoising method based on improved threshold function[J]. Laser Journal, 2023, 44(10):19-24.
[25]	周孟然, 张易平, 汪胜和, 等. 基于MVMD-CapSA-DBN的工业多元负荷分类研究[J]. 河南师范大学学报(自然科学版), 2023, 51(3):123-130.
	ZHOU Mengran, ZHANG Yiping, WANG Shenghe, et al. Research on industrial multiple load classification based on MVMD-CapSA-DBN[J]. Journal of Henan Normal University(Natural Science Edition), 2023, 51(3):123-130.
[26]	LV L, WANG W H, ZHANG Z Y, et al. A novel intrusion detection system based on an optimal hybrid kernel extreme learning machine[J]. Knowledge-Based Systems, 2020, 195(Prepublish):105648-105648.
[27]	TIAN Z, LI S, WANG Y, et al. Wind power prediction method based on hybrid kernel function support vector machine[J]. Wind Engineering, 2018, 42(3):252-264.
[28]	郁纪, 肖文波, 李欣蕊, 等. 基于SCSO-SVM算法的光伏组件故障识别[J]. 科学技术与工程, 2024, 24(3):1066-1074.
	YU Ji, XIAO Wenbo, LI Xinrui, et al. Fault identification of PV modules based on SCSO-SVM algorithm[J]. Science Technology and Engineering, 2024, 24(3):1066-1074.
[29]	杨晓燕, 谢满承, 郭小璇, 等. 基于改进原子轨道搜索算法优化随机森林分类器的光伏系统故障诊断[J]. 综合智慧能源, 2023, 45(10):53-60. doi: 10.3969/j.issn.2097-0706.2023.10.007
	YANG Xiaoyan, XIE Mancheng, GUO Xiaoxuan, et al. PV system fault diagnosis based on random forest classifier optimized by improved atomic orbital search algorithm[J]. Integrated Intelligent Energy, 2023, 45(10):53-60. doi: 10.3969/j.issn.2097-0706.2023.10.007
[30]	余修武, 黄露平, 刘永, 等. 融合柯西折射反向学习和变螺旋策略的WSN象群定位算法[J]. 控制与决策, 2022, 37(12):3183-3189.
	YU Xiuwu, HUANG Luping, LIU Yong, et al. Cauchy refraction opposition-based learning and variable helix mechanism of elephant herding localization algorithm in WSN[J]. Control and Decision, 2022, 37(12):3183-3189.
[31]	李旭炯, 孙林花, 杨郭明. 基于改进粒子群算法的光伏系统附加向心属性最大功率跟踪研究[J]. 综合智慧能源, 2022, 44(3):70-76. doi: 10.3969/j.issn.2097-0706.2022.03.011
	LI Xujiong, SUN Linhua, YANG Guoming. MPPT for PV systems appended with centripetal attribute based on improved PSO algorithm[J]. Integrated Intelligent Energy, 2022, 44(3):70-76. doi: 10.3969/j.issn.2097-0706.2022.03.011
[32]	ZHAO F Q, ZHANG L X, ZHANG Y, et al. An improved water wave optimisation algorithm enhanced by cma-es and opposition-based learning[J]. Connection Science, 2020, 32(2):1-30.

故障	IMF分量编号	方差贡献率	相关系数	信息熵
膜干	IMF1	0.236 0	0.277 6	9.961 2
	IMF3	0.180 4	0.483 0	9.565 2
	IMF4	0.172 3	0.499 7	9.652 9
	IMF5	0.075 0	0.415 7	9.698 0
	IMF7	0.044 7	0.266 7	9.577 5
水淹	IMF2	0.204 8	0.608 9	9.684 6
	IMF3	0.102 3	0.501 9	9.512 0
	IMF5	0.044 7	0.334 9	9.524 6
	IMF6	0.043 2	0.269 3	9.528 2
	IMF8	0.018 8	0.200 0	9.701 8

模型	准确率
ELM	76.00
WOA-ELM	84.00
HKELM	85.33
WHTD-MVMD-SCSO-HKELM	90.67
WHTD-MVMD-ISCSO-HKELM	98.85