Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GPU Accelerated 3D P-wave Source Free Adaptive Wavefield Reconstruction Inversion with an application to experimental VSP physical modeling data

Zhilong Fang, Jingjing Zong

Abstract

Wavefield reconstruction inversion (WRI) has been considered a potential solution to the issue of local minima inherent in conventional full waveform inversion (FWI) methods. However, most current WRI research has been confined to 2D problems due to the computational challenges posed by solving augmented systems for optimal data-fitting wavefields. This constraint limits WRI applicability to realistic 3D scenarios. This study introduces a GPU-accelerated 3D source-free adaptive WRI (GPU-SF-AWRI) method that overcomes these computational barriers by adaptively controlling the computational accuracy of wavefield simulation and optimizing GPU utilization, thus enhancing its suitability for 3D applications. The inclusion of an on-the-fly source estimation technique further boosts its performance on realistic problems. Numerical experiments reveal that the proposed GPU-accelerated method achieves a 195-fold speedup compared to CPU-based approaches. By incorporating adaptive accuracy and total variation regularization, we attain a 2-fold speedup while maintaining inversion accuracy. We applied the GPU-SF-AWRI method to numerical and actual Vertical Seismic Profiling (VSP) physical modeling P-wave data, confirming its efficacy in addressing real data challenges and mitigating local minima associated with conventional FWI.

GPU Accelerated 3D P-wave Source Free Adaptive Wavefield Reconstruction Inversion with an application to experimental VSP physical modeling data

Abstract

Wavefield reconstruction inversion (WRI) has been considered a potential solution to the issue of local minima inherent in conventional full waveform inversion (FWI) methods. However, most current WRI research has been confined to 2D problems due to the computational challenges posed by solving augmented systems for optimal data-fitting wavefields. This constraint limits WRI applicability to realistic 3D scenarios. This study introduces a GPU-accelerated 3D source-free adaptive WRI (GPU-SF-AWRI) method that overcomes these computational barriers by adaptively controlling the computational accuracy of wavefield simulation and optimizing GPU utilization, thus enhancing its suitability for 3D applications. The inclusion of an on-the-fly source estimation technique further boosts its performance on realistic problems. Numerical experiments reveal that the proposed GPU-accelerated method achieves a 195-fold speedup compared to CPU-based approaches. By incorporating adaptive accuracy and total variation regularization, we attain a 2-fold speedup while maintaining inversion accuracy. We applied the GPU-SF-AWRI method to numerical and actual Vertical Seismic Profiling (VSP) physical modeling P-wave data, confirming its efficacy in addressing real data challenges and mitigating local minima associated with conventional FWI.

Paper Structure

This paper contains 16 sections, 12 equations, 15 figures, 7 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Methodology
  3. 3D Source free wavefield reconstruction inversion
  4. GPU-Accelerated 3D Source-Free Adaptive WRI
  5. SF-AWRI with total variation regularization
  6. GPU acceleration
  7. Numerical experiments
  8. Camembert model
  9. Evaluation of GPU Acceleration Efficiency
  10. Evaluation of GPU-AWRI
  11. VSP numerical model
  12. Physical modeling experiment
  13. Experimental setup
  14. Wavefield preparation
  15. Inversion Result
  16. ...and 1 more sections

Figures (15)

  • Figure 1: A 2D example to show the process of source estimation, gradient calculation, and model updating at each iteration.
  • Figure 2: Accelerating matrix-vector multiplication using the GPU involves independent computations for each row of $\mathbf{S}$ with $\mathbf{p}$ and each row of $\mathbf{S}^\top$ with $\mathbf{s}$.
  • Figure 3: (a) True model and (b) initial model of the Camembert experiment.
  • Figure 4: Acceleration ratios corresponding to the choices of $n_b$ = 1, 2, and 4.
  • Figure 5: Results of (a) WRI with tolerance $\epsilon = 1 \times 10^{-6}$, (b) WRI with tolerance of $\epsilon = 4 \times 10^{-4}$, (c) AWRI, and (d) AWRI with TV regularization.
  • ...and 10 more figures