Full waveform inversion applied in defect investigation for ballastless undertrack structure of high-speed railway

Abstract

Until the end of 2015, China will build a national high-speed railway network with total mileage of exceed 20,000km. The Ballastless undertrack structure is wildly used in this network. Filed investigation shows that the damage and defects in ballastless undertrack structure will grow by long-term dynamic loading during the operation and bring potential risk to the operation safety. For better characterization of these internal problems, a time-domain full waveform inversion (FWI) method based on quasi-linear method and random search algorithm is developed as nondestructive testing (NDT) method. Tikhonov regularization is adopted to alleviate the ill conditioned matrix. Also, cross convolutions between observed data and forward response waveforms are employed to allow this FWI method to get rid of the influence of the source. According to filed investigation, two numerical case studies considered with horizontal discontinuity of material are carried out. It is verified that the presented technical is capable for characterizing anomalies of various zones in buried low-velocity layers. Meanwhile the FWI method is applied to experimental data in a real full scale model of ballastless high-speed railway undertrack structure. The inversion work successfully reveals the preset defects and damages inside the CA mortar layer and embankment. The results are generally consistent with the preset model.