Off-line condition monitoring of powered train doors.

Powered passenger train door unreliability is a major source of in-service delay and cancellation to passenger train services. This unreliability costs UK Train Operating Companies significant amounts of money through Railtrack charges, customer compensation payments and loss of revenue as well as door maintenance and repair procedures. The successful and reliable identification of incipient powered door faults could therefore help to reduce these charges and subsequent maintenance costs, increase train reliability and potentially reduce safety-related incidents.This thesis presents the research and development of an off-line condition monitoring tool for electric powered passenger train doors, the Door Analysis Tool. The Door Analysis Tool was developed through a two-year, collaborative TCS project between Sheffield Hallam University and Interfleet Technology, an international railway engineering consultancy. The project was partly funded by the Department of Trade and Industry.This thesis focuses on a number of primary areas of work, namely: Introduction to powered train doors and condition monitoring in the railway industry; Identification of suitable data collection and condition monitoring equipment forpowered train doors; Development of suitable data analysis and interpretation algorithms; Development of a commercially viable prototype condition monitoring product forpowered passenger train doors, the Door Analysis Tool; Validation of the Door Analysis Tool through theoretical and practical testing; Identification of further development opportunities and further future work.The Door Analysis Tool uses clamp-on current probes to measure the door motor current characteristic and traction interlock status during opening and closing cycles to monitor the performance of train doors. It offers a range of opportunities to different parties within the railway industry to help reduce the impact of powered train door unreliability and thereby improve train performance through the identification of faulty train doors prior to costly in-service failures.The practical testing of the prototype Door Analysis Tool showed that it functions as intended in a train maintenance depot environment and that it is capable of collecting, analysing and classifying data relating to train door performance.The theoretical tests presented the prototype Door Analysis Tool with nine different sets of data, each representing a range of faulty and fault-free door operations. The Door Analysis Tool was able to correctly classify six of the nine tests. Three of the tests incorporating minor fault conditions were not highlighted by the door performance classification, however all of the theoretical faults were identifiable through inspection of the analysis parameters, which can be stored after each test. The sensitivity of the Door Analysis Tool analysis routines can be adjusted to tailor the tool to different systems.