Switch Machine Fault Diagnosis Method Based on Kalman Filter and Support Vector Machines

You are here: Homepage > Search

Switch Machine Fault Diagnosis Method Based on Kalman Filter and Support Vector Machines

Li, Xiang / Qin, Yong / Wang, Zhipeng / Kan, Jiayu / Zhang, Xiaofeng

Switch machines are used for operating railway turnout; its error can cause delays, increase operating costs and may even lead to train accidents. Therefore, the fault diagnosis technology for the switch machine has received more and more attention. This paper proposes a fault diagnosis method based on the action current of switch machine. Firstly, the Kalman filter is used to preprocess the collected action current to reduce the influence of the unavoidable error of the measurement. In addition, we can further improve the accuracy of fault diagnosis by extracting the characteristics of the action current curve, like the maximum, minimum and average value, etc. Finally, we use DAG-SVMs to intelligently diagnose switch failures. Experiments show that the accuracy of classification after Kalman filter preprocessing is better than that of direct classification.

Access

Check access

Check availability in my library

Order at Subito €

Export, share and cite

Document information

Title :

Switch Machine Fault Diagnosis Method Based on Kalman Filter and Support Vector Machines

Additional title:

Lect. Notes Electrical Eng.

Contributors:

Qin, Yong (editor) / Jia, Limin (editor) / Liu, Baoming (editor) / Liu, Zhigang (editor) / Diao, Lijun (editor) / An, Min (editor) / Li, Xiang (author) / Qin, Yong (author) / Wang, Zhipeng (author) / Kan, Jiayu (author)

Conference:

International Conference on Electrical and Information Technologies for Rail Transportation ; 2019 ; Qingdao, China October 25, 2019 - October 27, 2019

Published in:

Proceedings of the 4th International Conference on Electrical and Information Technologies for Rail Transportation (EITRT) 2019 ; Chapter : 69 ; 727-735

Lecture Notes in Electrical Engineering ; 639

Publication date :

2020-04-04

Size :

9 pages

ISBN :

978-981-15-2866-8 , 978-981-15-2865-1

ISSN :

1876-1119 , 1876-1100

DOI :