Bo Wu

Bo Wu

Auto-generated profile, based on publicly available information

Bo Wu

State Key Laboratory of Petroleum Resources and Engineering 1

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 April 2024 in Geophysics
Verified
Journal Article
Closed access
BWBo Wu
Bo Wu
Gang YaoGang Yao
Gang Yao
XMXiao Ma
Xiao Ma
HCHanming Chen
Hanming Chen
DWDi Wu
Di Wu
静曹静杰 曹
静杰 曹
Least-squares reverse time migration (LSRTM) is currently one of the most advanced migration imaging techniques in the field of geophysics. It uses least-squares inversion to fit the observed data, resulting in high-resolution imaging results with more accurate amplitudes and better illumination compensation than conventional reverse time migration (RTM). However, noise in the observed data and the Born approximation forward operator can result in high-wavenumber artifacts in the final imaging results. Moreover, iteratively solving LSRTM leads to one or two orders of computational cost higher than conventional RTM, making it challenging to apply extensively in industrial applications. Simultaneous source acquisition technology can reduce the computational cost of LSRTM by reducing the number of wavefield simulations. However, this technique also can cause high-wavenumber crosstalk artifacts in the migration results. To effectively remove the high-wavenumber artifacts caused by these issues, simultaneous source and deep learning are combined to speed up LSRTM as well as suppress high-wavenumber artifacts. A deep residual neural network (DR-Unet) is trained with synthetic samples, which are generated by adding field noise to synthesized noise-free migration images. Then, the trained DR-Unet is applied on the gradient of LSRTM to remove high-wavenumber artifacts in each iteration. Compared to directly applying DR-Unet denoising to LSRTM results, embedding DR-Unet denoising into the inversion process can better preserve weak reflectors and improve denoising effects. Finally, the proposed LSRTM method is tested on two synthetic data sets and a land data set. The tests demonstrate that the proposed method can effectively remove high-wavenumber artifacts, improve imaging results, and accelerate convergence speed.

18 October 2022 in Energies
Verified
Journal Article
Open access
PDF file
YHYang Han
Yang Han
BWBo Wu
Bo Wu
Gang YaoGang Yao
Gang Yao
XMXiao Ma
Xiao Ma
DWDi Wu
Di Wu
Finite-difference methods are the most widely used methods for seismic wavefield simulation. However, numerical dispersion is the main issue hindering accurate simulation. In the case where the finite-difference scheme is known, the time dispersion can be predicted mathematically and, thus, can be eliminated. However, when only pre-compiled software is available for wavefield simulation, which is common in practical applications, the software-used algorithm becomes a black box (unknown). Therefore, it is challenging to obtain the mathematical expression of the time dispersion, resulting in difficulty in eliminating the time dispersion. To solve this problem, we propose to use deep learning methods to eliminate time dispersion. We design a semi-supervised framework based on convolutional and recurrent neural networks for eliminating time dispersion caused by seismic wave modeling. The framework of our proposed neural network includes two main modules: Inverse Model and Forward Model, both of which have learnable parameters. The Inverse Model is used for eliminating time dispersion while the Forward Model is used for regularizing the training. Particularly, this framework includes two steps: Firstly, using the compiled modeling software to generate two data sets with large and small time steps. Secondly, we train these two modules for transformation between large time-step data (with time dispersion) and small time-step data (without time dispersion) by labeled and unlabeled data sets. In this work, the labeled data set is a paired data set with large time-step data and their corresponding small time-step data; the unlabeled data set is the large time-step data that need time-dispersion elimination. We use the unlabeled data set to guide the network. In this learning framework, re-training is required whenever the modeling algorithms, time interval, or frequency band is changed. Hence, we propose a transfer learning training method to extend from the trained model to another model, which reduces the computational cost caused by re-training. This minor drawback is offset overwhelmingly by the modeling efficiency gain with large time steps in large-scale production. Tests on two models confirm the effectiveness of the proposed method.

29 August 2026 in Geophysics
Verified
Journal Article
Closed access
BWBo Wu
Bo Wu
Gang YaoGang Yao
Gang Yao
PZPingmin Zhang
Pingmin Zhang
DWDi Wu
Di Wu
Full-waveform inversion (FWI) based on the full wave equation uses travel-time and amplitude information from seismic records to recover high-resolution subsurface models. As seismic records are nonlinearly related to subsurface parameters, FWI is inherently a nonlinear inverse problem. Although global inversion methods can effectively handle such nonlinearity, they are computationally expensive. Therefore, local gradient methods are widely adopted. To avoid explicitly computing the Jacobian matrix, these approaches use the adjoint-state method, which computes the gradient by cross-correlating the forward wavefield from the state equation with the adjoint wavefield from the adjoint equation. However, when FWI is based on the first-order VTI elastic wave equation, the adjoint-state equation is inconsistent with the state equation. This introduces two challenges: (1) the programming complexity increases because the forward and adjoint wavefields require separate simulation codes, and (2) the adjoint-state equation cannot use the same free-surface boundary conditions as the state equation. In FWI, the inability of the adjoint-state equation to correctly impose free-surface boundary conditions can adversely affect the final inversion results. To address these issues, a skew-adjoint form of the first-order VTI elastic wave equation was proposed. In the skew-adjoint VTI elastic wave equation, the operator and its adjoint operator had the same form, differing only by a negative sign. Because it was mathematically equivalent to the first-order VTI elastic wave equation, the proposed skew-adjoint formulation allowed the state and adjoint-state equations to use consistent free-surface boundary conditions. Finally, the effectiveness of our method was demonstrated using two synthetic data sets.