To accurately identify faults in Solid Oxide Fuel Cell (SOFC) systems and overcome the challenges of traditional methods such as difficulties in parameter determination, high result randomness, and limitations in complex data processing, a fault identification method based on fusion clustering and Radial Basis Function Support Vector Machine (RBF-SVM) is proposed. This method employs a clustering algorithm based on a fusion mechanism to process the performance degradation fault data of the SOFC system reformer catalyst, obtaining clustering labels under different fault levels, and then uses the RBF-SVM algorithm to train the classified fault data to develop a fault identification model. This model can map fault feature information to higher-dimensional spaces to enhance fault identification capabilities and is further applied to identify air and fuel leakage faults. Experimental results demonstrate that the proposed fusion clustering algorithm effectively classifies faults, and the RBF-SVM fault identification model achieves 99.3% accuracy in identifying reformer catalyst degradation faults, 99.9% accuracy for air leakage faults, and 99.4% accuracy for fuel leakage faults.
本文提出一种基于融合聚类和径向基核函数支持向量机(Radial Basis Function Support Vector Machine,RBF-SVM)的故障识别方法,主要贡献如下:1)基于数据对SOFC系统进行故障识别,同时对SOFC系统进行全工况实验并采集大量跨工况数据,确保故障识别算法能够充分训练并挖掘故障信息;2)提出一种融合聚类算法用于SOFC系统故障分类,对未标记故障数据进行划分和标记,得到标签数据,克服传统聚类方法的不足;3)基于支持向量机(Support Vector Machine,SVM)算法,为SOFC系统设计了融合聚类和RBF-SVM算法对故障数据进行训练并用于故障识别,通过径向基核函数将故障特征信息映射到更高维空间,使故障信息能更容易区分。
本实验从一套以天然气为燃料的SOFC系统中采集运行数据。一个完整的SOFC系统由多个辅助发电设备(Balance of Plant,BOP)构成,能够完成独立发电。该系统由4个子系统构成:电堆子系统、气体供应子系统、尾气回收子系统和电控子系统,结构如图2所示。其构成部件包括电堆、重整器、空气热交换器、空气-燃料热交换器、尾气燃烧室、水蒸发器、除硫器等。
SINGLAM K, NIJHAWANP, OBEROIA S. Correction to: Hydrogen fuel and fuel cell technology for cleaner future: A review[J]. Environmental Science and Pollution Research, 2021, 28(15): 19536. DOI: 10.1007/s11356-021-13191-3 .
[2]
XUK, SHENC C, XUC, et al. Modeling and control-oriented thermal characteristics under variable load of the solid oxide fuel cell[J]. Journal of Solid State Electrochemistry, 2023, 27(8): 2083-2099. DOI: 10.1007/s10008-023-05477-y .
[3]
CHENJ W, HUZ C, ZHANGH S. A novel control scheme to mitigate temperature profile variation in a SOFC System[J]. Energy Reports, 2022, 8: 10604-10613. DOI: 10.1016/j.egyr.2022.08.185 .
[4]
YANJ, CHENH L. Power generation via solid oxide fuel cell using ex-reformed oxygen-containing low-concentration methane[J]. Materials Chemistry and Physics, 2024, 313: 128736. DOI: 10.1016/j.matchemphys.2023.128736 .
[5]
ZHANGL, LIX, JIANGJ H, et al. Dynamic modeling and analysis of a 5-kW solid oxide fuel cell system from the perspectives of cooperative control of thermal safety and high efficiency[J]. International Journal of Hydrogen Energy, 2015, 40(1): 456-476. DOI: 10.1016/j.ijhydene.2014.10.149 .
DUK, LINK S, SONGC, et al. Current status and prospects of solid oxide fuel cell stacks and related systems[J]. Proceedings of the CSEE, 2025, 45(10): 3877-3892. DOI: 10.13334/j.0258-8013.pcsee.232375(Ch ).
NIEL, YANGT, ZHANGJ J, et al. Multi-attribute decision-making and k-means clustering based relay selection method for vehicular safety messages dissemination[J]. Journal of Wuhan University (Natural Science Edition), 2023, 69(5): 609-616. DOI: 10.14188/j.1671-8836.2022.0198(Ch ).
[10]
XIONGX Y, YANGL, WUY F, et al. External temperature field test and leakage fault diagnosis for SOFC stacks[J]. International Journal of Hydrogen Energy, 2023, 48(7): 2788-2800. DOI: 10.1016/j.ijhydene.2022.10.175 .
[11]
DINGH H, HUQ R, WANGY Z, et al. A mechanism-based convex model of fuel cell systems considering the effect of auxiliary system[J]. IEEE Transactions on Sustainable Energy, 2025, 16(4): 3124-3127. DOI: 10.1109/TSTE.2025.3564106 .
[12]
DUY, LIY H, RENP, et al. Oxygen transfer at mesoscale catalyst layer in proton exchange membrane fuel cell: Mechanism, model and resistance characterization[J]. Chemical Engineering Journal, 2024, 494: 153021. DOI: 10.1016/j.cej.2024.153021 .
[13]
GUOS, ZHAOY Q, WANGX Y, et al. Leakage mechanism model of proton exchange membrane fuel cell sealing structures under vibration conditions[J]. International Journal of Green Energy, 2024, 21(12): 2912-2922. DOI: 10.1080/15435075.2024.2338386 .
[14]
WUX L, XUY W, PENGJ X, et al. Novel hybrid modeling and analysis method for steam reforming solid oxide fuel cell system multifault degradation fusion assessment[J]. ACS Omega, 2023, 8(40): 36876-36892. DOI: 10.1021/acsomega.3c03928 .
[15]
XUY W, JIANGC, PENGJ X, et al. Fault prognosis method for solid oxide fuel cells based on mechanism degradation process model and particle filtering[J]. IEEE Transactions on Power Electronics, 2023, 38(6): 6831-6840. DOI: 10.1109/TPEL.2023.3246068 .
WUX, HUC, WANGQ, et al. VMD based fault diagnosis method for SOFC stack seal failure and leakage[J]. Electric Power, 2024, 57(12): 139-147 (Ch).
[18]
ZHANGJ H, YUS, LEIL B, et al. Experimental detection of fuel leakage in solid oxide fuel cells based on voltage signals[J]. International Journal of Hydrogen Energy, 2024, 77: 575-581. DOI: 10.1016/j.ijhydene.2024.06.144 .
[19]
YANGJ, LIZ J, BIANR L, et al. Fault diagnosis of SOFC system based on single cell voltage analysis[J]. International Journal of Hydrogen Energy, 2021, 46(48): 24531-24545. DOI: 10.1016/j.ijhydene.2021.04.114 .
[20]
ZHENGY, WUX L, ZHAOD Q, et al. Data-driven fault diagnosis method for the safe and stable operation of solid oxide fuel cells system[J]. Journal of Power Sources, 2021, 490: 229561. DOI: 10.1016/j.jpowsour.2021.229561 .
[21]
WUX L, MEIJ, XUY W, et al. Stack performance classification and fault diagnosis optimization of solid oxide fuel cell system based on Bayesian artificial neural network and feature selection[J]. Journal of Power Sources, 2024, 620: 235198. DOI: 10.1016/j.jpowsour.2024.235198 .
[22]
YIF Y, SUNY, ZHANGJ M, et al. Early flooding fault diagnosis method of fuel cell based on feature amplification transformer[J]. International Journal of Hydrogen Energy, 2025, 119: 13-24. DOI: 10.1016/j.ijhydene.2025.03.241 .
CHENM, CHENX P, CHENW, et al. Research on division and recognition of grate cooler operating conditions based on fusion clustering[J]. Control Engineering of China, 2023, 30(2): 361-367. DOI: 10.14107/j.cnki.kzgc.20200861(Ch ).
HUT, WUM T, HUW, et al. Reactivity classification prediction of coal fly ash based on machine learning[J]. Journal of Central South University (Science and Technology), 2023, 54(10): 3829-3839 (Ch).