High-Dimensional Wavefield Solutions Using Physics-Informed Neural Networks with Frequency-Extension

Abstract

Solving the wave equation is essential to seismic imaging and inversion. The numerical solution of the Helmholtz equation, fundamental to this process, often encounters significant computational and memory challenges. We propose an innovative frequency-domain scattered wavefield modeling method employing neural operators adaptable to diverse seismic velocities. The source location and frequency information are embedded within the input background wavefield, enhancing the neural operator's ability to process source configurations effectively. In addition, we utilize a single reference frequency, which enables scaling from larger-domain forward modeling to higherfrequency scenarios, thereby improving our method's accuracy and generalization capabilities for larger-domain applications. Several tests on the OpenFWI datasets and realistic velocity models validate the accuracy and efficacy of our method as a surrogate model, demonstrating its potential to address the computational and memory limitations of numerical methods.

Integrating U-net with full-waveform inversion for an efficient salt body construction

“…Recently, deep learning has been utilized in many geophysical applications including modelling, processing, interpretation, and inversion, overcoming many limitations of the conventional methods (Huang et al, 2022;Song et al, 2021;Liu et al, 2022;Harsuko and Alkhalifah, 2022;AlAli and Anifowose, 2022;Zhou et al, 2020;Xiong et al, 2018;Kazei et al, 2020;Yang and Ma, 2019;Araya-Polo et al, 2018). In salt model building, deep learning contributes through two general approaches.…”