Fast approximate simulation of seismic waves with deep learning

Abstract

We simulate the response of acoustic seismic waves in horizontally layered media using a deep neural network. In contrast to traditional finite-difference modelling techniques our network is able to directly approximate the recorded seismic response at multiple receiver locations in a single inference step, without needing to iteratively model the seismic wavefield through time. This results in an order of magnitude reduction in simulation time from the order of 1 s for FD modelling to the order of 0.1 s using our approach. Such a speed improvement could lead to real-time seismic simulation applications and benefit seismic inversion algorithms based on forward modelling, such as full waveform inversion. Our proof of concept deep neural network is trained using 50,000 synthetic examples of seismic waves propagating through different 2D horizontally layered velocity models. We discuss how our approach could be extended to arbitrary velocity models. Our deep neural network design is inspired by the WaveNet architecture used for speech synthesis. We also investigate using deep neural networks for simulating the full seismic wavefield and for carrying out seismic inversion directly.

10 pages, 8 figures

References in corpus (3)

• Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multi-Layer Calorimeters
• Fast approximate simulation of seismic waves with deep learning
• Probabilistic Neural-Network Based 2D Travel Time Tomography

Cited by in corpus (15)

• Physics-informed Neural Networks (PINNs) for Wave Propagation and Full Waveform Inversions
• Finite Basis Physics-Informed Neural Networks (FBPINNs): a scalable domain decomposition approach for solving differential equations
• Solving the acoustic VTI wave equation using physics-informed neural networks
• Wavefield reconstruction inversion via physics-informed neural networks
• Fast approximate simulation of seismic waves with deep learning
• Rapid Seismic Waveform Modeling and Inversion with Neural Operators
• Deep Neural Helmholtz Operators for 3D Elastic Wave Propagation and Inversion
• Instantaneous physics-based ground motion maps using reduced-order modeling
• Unsupervised Learning of Full-Waveform Inversion: Connecting CNN and Partial Differential Equation in a Loop
• Accelerating Bayesian microseismic event location with deep learning
• Emerging Directions in Geophysical Inversion
• Reduced-order modeling for complex 3D seismic wave propagation
• Seismic wave propagation and inversion with Neural Operators
• Modeling the Asymptotic Behavior of Higher-Order Aftershocks with Deep Learning
• Real-Time Reconstruction of Ground Motion During Small Magnitude Earthquakes: A Pilot Study