Modeling track geometry degradation using Support Vector Machine technique

Can Hu, Xiang Liu

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

5 Scopus citations

Abstract

Analyzing track geometry defects is of crucial importance for railway safety. Understanding when a defect will need to be repaired can help in both planning a preventive maintenance schedule and reducing the probability of track failures. This paper discusses the data cleaning and analysis processes for modeling track geometry degradation. An analytical data model named the Support Vector Machine (SVM) was developed to model the deterioration of track geometry defects. This paper mainly focuses on the following three defect types - surface, cross level and dip. The model accounts for traffic volume, defect amplitude, track class, speed and other potential factors. Results demonstrate that the proposed analytical data model can have a prediction accuracy above 70%.

Original languageEnglish (US)
Title of host publication2016 Joint Rail Conference, JRC 2016
PublisherAmerican Society of Mechanical Engineers
ISBN (Electronic)9780791849675
DOIs
StatePublished - 2016
Event2016 Joint Rail Conference, JRC 2016 - Columbia, United States
Duration: Apr 12 2016Apr 15 2016

Publication series

Name2016 Joint Rail Conference, JRC 2016

Other

Other2016 Joint Rail Conference, JRC 2016
CountryUnited States
CityColumbia
Period4/12/164/15/16

All Science Journal Classification (ASJC) codes

  • Transportation

Fingerprint Dive into the research topics of 'Modeling track geometry degradation using Support Vector Machine technique'. Together they form a unique fingerprint.

Cite this