首页»
在线发表
Petroleum Science > DOI: https://doi.org/10.1016/j.petsci.2022.01.013
A comparison of deep learning methods for seismic impedance inversion Open Access
文章信息
作者:Si-Bo Zhang, Hong-Jie Si, Xin-Ming Wu, Shang-Sheng Yan,
作者单位:
投稿时间:
引用方式:Si-Bo Zhang, Hong-Jie Si, Xin-Ming Wu, Shang-Sheng Yan, A comparison of deep learning methods for seismic impedance inversion, Petroleum Science, 2022, , https://doi.org/10.1016/j.petsci.2022.01.013.
文章摘要
Abstract
Deep learning is widely used for seismic impedance inversion, but few work provides in-depth research and analysis on designing the architectures of deep neural networks and choosing the network hyperparameters. This paper is dedicated to comprehensively studying on the significant aspects of deep neural networks that affect the inversion results. We experi- mentally reveal how network hyperparameters and architectures affect the inversion perfor- mance, and develop a series of methods which are proven to be effective in reconstructing high-frequency information in the estimated impedance model. Experiments demonstrate that the proposed multi-scale architecture is helpful to reconstruct more high-frequency details than a conventional network. Besides, the reconstruction of high-frequency informa- tion can be further promoted by introducing a perceptual loss and a generative adversarial network from the computer vision perspective. More importantly, the experimental re- sults provide valuable references for designing proper network architectures in the seismic inversion problem.
Deep learning is widely used for seismic impedance inversion, but few work provides in-depth research and analysis on designing the architectures of deep neural networks and choosing the network hyperparameters. This paper is dedicated to comprehensively studying on the significant aspects of deep neural networks that affect the inversion results. We experi- mentally reveal how network hyperparameters and architectures affect the inversion perfor- mance, and develop a series of methods which are proven to be effective in reconstructing high-frequency information in the estimated impedance model. Experiments demonstrate that the proposed multi-scale architecture is helpful to reconstruct more high-frequency details than a conventional network. Besides, the reconstruction of high-frequency informa- tion can be further promoted by introducing a perceptual loss and a generative adversarial network from the computer vision perspective. More importantly, the experimental re- sults provide valuable references for designing proper network architectures in the seismic inversion problem.
关键词
-
seismic inversion; impedance; deep learning; network architecture