HCTNet: Robust prestack seismic inversion using a hybrid convolutional transformer
- 7 January 2025
- journal article
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 90 (4) , N17-N32
Abstract
Accurately estimating subsurface elastic parameters is essential to characterize oil and gas reservoirs. Prestack amplitude-variation-with-angle (AVA) inversion methods are used to estimate subsurface elastic parameters. The traditional physical model-based linear AVA inversion methods are challenging for accurately calculating the velocity and density of the prestack data with large incidence angles. Data-driven deep-learning networks have received much attention because of their ability to construct nonlinear relationships between a source and target domains. For prestack seismic inversion, the limitation of wells affects the inversion performance of elastic parameters. Here, we construct a hybrid neural network with a 1D convolutional neural network (CNN) and transformer for prestack seismic inversion. We use a hybrid dilated convolutional block with a residual connection mechanism (RHDC) to initially extract features from prestack angle-domain gathers. RHDC has a larger receptive field than traditional convolutional layers, which allows the extraction of larger-scale features. Our method is a prestack elastic parameter simultaneous inversion network based on a multitask learning strategy. We include a seismic data reconstruction branch to improve the inversion performance, which is equivalent to a regularization method that focuses on the overall structure and consistency of the seismic data when inverting elastic parameters. We simulate synthetic data based on the feature distribution of the field data. For field data inversion, the network is pretrained using synthetic data with limited labels, and then the well-trained model is applied to train on field data, reducing reliance on wells. We use a hybrid loss function in the training stage and tune the loss weights across the three inversion branches to balance the inversion performance. Compared with two state-of-the-art deep-learning networks, i.e., a residual CNN and a hybrid-seismic-prior-guided neural network, our method has better horizontal continuity and noise robustness on synthetic and field data examples.Keywords
This publication has 31 references indexed in Scilit:
- EQCCT: A Production-Ready Earthquake Detection and Phase-Picking Method Using the Compact Convolutional TransformerIEEE Transactions on Geoscience and Remote Sensing, 2023
- Seismic Data Interpolation Based on Multi-Scale TransformerIEEE Geoscience and Remote Sensing Letters, 2023
- Intelligent AVA Inversion Using a Convolution Neural Network Trained with Pseudo-Well DatasetsSurveys in Geophysics, 2023
- Nonstationary phase-corrected full-waveform inversion with attenuation compensation in viscoacoustic mediumJournal of Geophysics and Engineering, 2022
- AVO Inversion Based on Transfer Learning and Low-Frequency ModelIEEE Geoscience and Remote Sensing Letters, 2021
- Robust deep learning seismic inversion witha prioriinitial model constraintGeophysical Journal International, 2021
- Two-hidden-layer feed-forward networks are universal approximators: A constructive approachNeural Networks, 2020
- Automatic Waveform Classification and Arrival Picking Based on Convolutional Neural NetworkEarth and Space Science, 2019
- Data-driven simultaneous seismic inversion of multiparameters via collaborative sparse representationGeophysical Journal International, 2019
- Color TV: total variation methods for restoration of vector-valued imagesIEEE Transactions on Image Processing, 1998