Sensor Placement for Pantograph Fault Detection Based on PCA/KPCA Method

Jiaqi Qin, Moussa Hamadache, Roger Dixon

Research output: Chapter in Book/Report/Conference proceedingConference contribution


In the UK, rail transport is very important to its transportation system. Due to the advantages of electric trains, the electrification of railways has been an irreversible trend. At the end of 2019, about 38% of the UK railway network has been electrified. Some experts predict that the removal of diesel only passenger trains from the National Rail network is achievable by 2040. Pantograph is a device that allows electric trains to obtain power from overhead lines. When it works abnormally, property damage and casualties may be cause, such as the Littleport accident in January 2012. Where pantograph assembly fell from the roof of a passenger train, and broke two windows on its way to the ground. Overhead line and the train body were damaged, and apassenger was injured. Therefore, it is essential to detect faults happen in pantograph. In this paper, a new fault-detection method is developed based on principal component analysis (PCA) and kernel principal component analysis (KPCA) algorithm. In addition, each sensor’s contribution rate for a specific fault (e.g., air cylinder fault) is also obtained, which can help optimize the positions of the sensors in subsequent research.
Original languageEnglish
Title of host publicationHigh Speed Rail: Education Interchange 2020
Number of pages2
Publication statusAccepted/In press - 14 Dec 2020


  • fault-detection
  • pantograph
  • PCA
  • KPCA
  • contribution rate


Dive into the research topics of 'Sensor Placement for Pantograph Fault Detection Based on PCA/KPCA Method'. Together they form a unique fingerprint.

Cite this