A Multi-task Deep Learning Model for the Denoising, Interpolation, and Wavefield Separation of DAS-VSP Data

Distributed acoustic sensing (DAS) is an innovative data acquisition method and is usually combined with vertical seismic profiling (VSP) technique for downhole seismic exploration, offering some advantages such as low cost, high resistance to temperature and pressure, and a high spatial sampling rate. A pre-processing workflow can significantly improve the quality of pre-stack seismic data for subsequent inversion and high-resolution imaging. Denoising, interpolation, and wavefield separation are three necessary pre-processing steps for the pre-stack DAS-VSP records. Traditionally, these three steps are carried out independently, resulting in certain shortcomings, such as the amplitude attenuation of recovered signals, higher computational cost, and extensive parameter adjustments. In this work, we propose a multi-task pre-processing model (MTPM) to combine the three steps together. It can simultaneously perform denoising, interpolation, and wavefield separation operations on DAS-VSP data using only a single well-trained model. The network architecture of MTPM is a combination of two classical frameworks: a convolutional neural network (CNN) and a Transformer. Specifically, this proposed MTPM comprises a Front-Net and a Post-Net. The Front-Net is a Transformer-based encoder-decoder structure enhanced by a U-Net++ block. This hybrid CNN-Transformer structure can simultaneously extract both local and global features using the convolution and multi-head self-attention operations, which are important for denoising and interpolation tasks. The Post-Net is a pure Transformer that focuses only on global features and thus enhances the performance of wavefield separation. Furthermore, we design a two-stage training strategy for MTPM to combine the three tasks together. In our experiments, we use synthetic and field DAS-VSP records to test the effectiveness of MTPM, which demonstrates better denoising, interpolation, and wavefield separation performance compared to other methods.