Inversion of vehicle-induced signals based on seismic interferometry and recurrent neural networks

Abstract

Vehicle-induced vibrations provide useful signals for passive seismic exploration. Such signals are repeatable and environmentally friendly; hence, they can provide an economical way to analyze subsurface structures. We have developed a new workflow to monitor roads or railways by producing 1D subsurface S-wave velocities in real time. This workflow consists of two steps: seismic interferometry and recurrent neural networks (RNN). Seismic interferometry can efficiently retrieve surface waves by crosscorrelating vehicle-induced vibrations. RNN was designed to first encode the picked dispersion curve into a fixed-length vector and then decode the vector into 1D S-wave velocities. To simulate the railway vibrations, we first analyze the time-dependent characteristic of the high-speed-train source and verify its mathematical expression by comparing the frequency spectrums of the real and synthetic data. We then evaluate the RNN-based surface-wave dispersion inversion method and validate the designed network structure using the 3D SEG/EAGE overthrust model. Finally, seismic interferometry and RNN-based surface-wave inversion are applied to a synthetic train-induced data set, a 33 min field record of railway vibrations and a 76 min field data of road vibrations, respectively. The synthetic and field data tests indicated that our workflow can be a feasible and cost-effective tool for real-time monitoring of subsurface media along roads and railways.