FUNDAMENTAL STUDY ON TWIST DETECTION OF TRACK USING VEHICLE BODY RESPONSE CHARACTERISTICS BASED ON MBS

Abstract

 In recent years, a severe labor shortage has made the automation and streamlining of track maintenance an urgent priority. As a potential solution, the estimation of track irregularities using simplified onboard measurements with low-cost portable devices has attracted significant attention. As track irregularities are reflected in vehicle body responses through vehicle-track dynamic interaction, it is possible to estimate track conditions by measuring responses, such as carbody acceleration. The objective of this study is to develop a methodology for detecting twist irregularities based on vehicle body responses. Multi-body simulation (MBS) was employed to reproduce vehicle responses while traveling over tracks with twist and alignment irregularities. The findings indicate that twist irregularities elicit distinct carbody vibration modes compared to those induced by alignment irregularities, resulting in varied response characteristics. Based on these findings, a filtering approach capable of differentiating between the two types of irregularities was investigated. Furthermore, the effects of traveling speed and wavelength on vehicle responses to twist irregularities were analyzed. The findings suggest that the characteristics of carbody roll angular velocity are significantly influenced by both input frequency and the phase relationship associated with the time lag between front and rear bogies. These findings establish a fundamental basis for high-precision, low-cost track condition monitoring.