Application of Affinity Propagation Clustering Algorithm in Fault Diagnosis of Metro Vehicle Auxiliary Inverter

  • Junwei Gao
  • Zengtao Ma
  • Yong Qin
  • Limin Jia
  • Dechen Yao
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 288)


Auxiliary inverter is one of the most important electrical equipments of metro vehicle; its complex structure causes various faults frequently. In this paper, the fundamental of the Affinity Propagation algorithm is introduced, and its application on fault diagnosis of metro vehicle auxiliary inverter is studied. Fault signals including voltage frequency variation, pulse transient, and power interruption are simulated by using the MATLAB software; clustering center matrix is calculated on the basis of AP algorithm, and the fault samples are classified by calculating the similarity degree between samples and clustering center. The simulation results show that the AP algorithm without initial clustering center can be used in the field of fault diagnosis, and even has better results than FCM algorithm.


Affinity propagation Fault diagnosis Auxiliary inverter Clustering analysis 



This work is partially supported by the National Science and Technology Pillar Program (2011BAG01B05), the Foundation of Shandong Province (BS2011DX008,ZR2011FM008), and the State Key Laboratory of Rail Traffic Control and Safety Foundation (RCS2011K005), Beijing Jiaotong University.


  1. 1.
    Yao DC, Yang JW, Yin YF et al (2010) Fault diagnosis of railway bearing based on muti-method fusion techniques. Mach Des Res 26(3):70–73 (in Chinese)Google Scholar
  2. 2.
    Huang YC (2003) Condition assessment of power transformers using genetic-based neural networks. IEE Proc Sci Meas Technol 150(1):19–24CrossRefGoogle Scholar
  3. 3.
    Bezdek JC (1974) Cluster validity with fuzzy sets. J Cybern 3(3):58–72MathSciNetCrossRefGoogle Scholar
  4. 4.
    Ma ZT, Gao JW, Leng ZW et al (2013) A method of metro vehicle auxiliary inverter fault diagnosis based on fuzzy clustering arithmetic. J Qingdao Univ Eng Technol Ed 28(3):8–14 (in Chinese)Google Scholar
  5. 5.
    Ma ZT, Gao JW, Zhang B et al (2013) A method of metro vehicle auxiliary inverter fault diagnosis based on weighted fuzzy clustering algorithm. Lect Notes Electr Eng 254:633–639CrossRefGoogle Scholar
  6. 6.
    Gao JW, Yu JP, Leng ZW et al (2013) The application of PSO-LSSVM in fault diagnosis of subway auxiliary inverter. ICIC Express Lett 4(3):777–784Google Scholar
  7. 7.
    Meng XY, Han XJ (2007) The fuzzy C-means clustering algorithm and its application in the fault diagnosis of ships. Shipbuild China 48(4):98–103 (in Chinese)MathSciNetGoogle Scholar
  8. 8.
    Xiong H, Sun CX, Liao RJ et al (2005) Study on kernel-based possibilistic clustering and dissolved gas analysis for fault diagnosis of power transformer. Proc CSEE 25(20):163–167 (in Chinese)Google Scholar
  9. 9.
    Frey BJ, Dueck D (2007) Clustering by passing messages between data points. Science 315(5814):972–976MathSciNetCrossRefMATHGoogle Scholar
  10. 10.
    Xiao Y, Yu J (2008) Semi-Supervised clustering based on affinity propagation algorithm. J Softw 19(11):2803–2813 (in Chinese)CrossRefMATHGoogle Scholar
  11. 11.
    Wang KJ, Zhang JY, Li D et al (2007) Adaptive affinity propagation clustering. Acta Automatica Sin 33(12):1242–1246 (in Chinese)MathSciNetMATHGoogle Scholar
  12. 12.
    Yang JW, Cai GQ, Yao DC et al (2010) Fault diagnosis method for the rolling bearing of railway vehicle based on wavelet packet transform and BP neural network. China Railw Sci 31(6):68–73 (in Chinese)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Junwei Gao
    • 1
    • 2
  • Zengtao Ma
    • 1
  • Yong Qin
    • 2
  • Limin Jia
    • 2
  • Dechen Yao
    • 2
  1. 1.College of Automation EngineeringQingdao UniversityQingdaoChina
  2. 2.State Key Laboratory of Rail Traffic Control and SafetyBeijing Jiaotong UniversityBeijingChina

Personalised recommendations