Probability of derailment under earthquake conditions
Author(s)
Guillaud, Lucile M. (Lucile Marie)
DownloadFull printable version (19.26Mb)
Other Contributors
Massachusetts Institute of Technology. Dept. of Civil and Environmental Engineering.
Advisor
Daniele Veneziano.
Terms of use
Metadata
Show full item recordAbstract
A quantitative assessment of the probability of derailment under earthquake conditions is presented. Two derailment modes are considered: by vibratory motion - during the ground motion - and by permanent track deformation - after the motion ended. Criteria for derailment that apply to both modes are derived in terms of peak transversal acceleration and peak transversal displacement. This allows a direct comparison between the two causes of derailment. We find that the first mode of derailment (by vibratory motion) dominates over the second mode (by track damage). The model considers the effect of spatial non-homogeneities in soil and structural characteristic and the incoherence of the ground motion into the assessment of derailment risk. The lateral motion experienced by the train under non-synchronous vibration of the track is obtained as the superposition of two contributions: one is the track motion at a fixed location and the other is the motion as the train travels on deformed tracks. Under linear elastic conditions, a method to obtain the power spectral density function for ground acceleration is presented and used to obtain acceleration and displacement response spectra. (cont.) The second component of motion depends on speed. It is found that the train motion due to track deformation has small effects at ordinary speeds but that it becomes noticeable as the speed increases and the support spacing decreases. In general, it is shown that changes in soil and structural properties present a higher risk for derailment by vibratory motion. In some cases, the second component of train motion may increase the acceleration due to track motion at a single location by a factor of two. The analysis is first done assuming linear behavior of the soil and structure and then nonlinearities and permanent deformations are included. The elastic analysis is found to be adequate, except for structures with natural periods exceeding 1 second where the elastic analysis yields conservative estimates in comparison with the inelastic case.
Description
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 2006. Includes bibliographical references (leaves 141-144).
Date issued
2006Department
Massachusetts Institute of Technology. Department of Civil and Environmental EngineeringPublisher
Massachusetts Institute of Technology
Keywords
Civil and Environmental Engineering.