The accurate prediction of the geological conditions ahead of a tunnel plays an important role in tunnel construction. Among all forward geological prospecting methods, the seismic detection method is widely applied. However, due to the characteristics of the tunnel and the complexity of the geological conditions, the seismic wavefield is complicated. Carrying out a more realistic forward modeling method is vital for fully understanding the law of seismic wave propagation and the characteristics of seismic wavefield in the tunnel. In this paper, the 3D staggered-grid finite-difference scheme in the cylindrical coordinate system based on the decoupled nonconversion elastic wave equation is used to carry out the numerical simulation. This method can avoid the diffraction interferences produced at the edges of the tunnel face in the Cartesian coordinate system. Based on this forward modeling method, the characteristics of wavefield and propagation laws of seismic waves under three kinds of common typical unfavorable geological models were explored, which can provide theoretical guidance to seismic data interpretation of tunnel seismic forward prospecting in practice.

Elastic waveform inversion plays a vital role in estimating the Earth's subsurface property. The inversion of multiple elastic parameters from observation data has been regarded as a cutting-edge and challenging issue due to its severe non-linearity and ill-posed nature. Deep learning approaches have recently demonstrated incredible potential in simulating non-linear mapping and made remarkable achievements in geophysical inversion. In this work, we consider multi-parameter elastic inversion as a multi-task learning problem and propose to accomplish the three tasks sequentially by a deep neural network with sequential decoders, which we name ElasInvNet. Specifically, we reconstruct the three elastic parameters, P-velocity, S-velocity, and density, one after the other, and use the features from the former task as the prior information to assist the later tasks' reconstruction. We verified the effectiveness of the proposed ElasInvNet through comprehensive comparisons and ablation studies.