Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A deep learning-enhanced framework for multiphysics joint inversion

Journal Article · · Geophysics
 [1];  [2];  [3];  [2];  [4];  [4];  [5]
  1. University of Houston, Department of Electrical and Computer Engineering, Houston, Texas, USA.; OSTI
  2. University of Houston, Department of Earth and Atmospheric Sciences, Houston, Texas, USA.
  3. University of Houston, Department of Information and Logistics Technology, Houston, Texas, USA.
  4. Cyentech Consulting LLC, Houston, Texas, USA.
  5. University of Houston, Department of Electrical and Computer Engineering, Houston, Texas, USA. (corresponding author)

Joint inversion has drawn considerable attention due to the availability of multiple geophysical data sets, ever-increasing computational resources, the development of advanced algorithms, and its ability to reduce inversion uncertainty. A key issue of joint inversion is to develop effective strategies to link different geophysical data in a unified mathematical framework, in which the information obtained from different models can complement each other. We have developed a deep learning-enhanced joint inversion framework to simultaneously reconstruct different physical models by fusing different types of geophysical data. Traditionally, structure similarity constraints are pursued by joint inversion algorithms using manually crafted formulations (e.g., cross gradient). The constraint is constructed by a deep neural network (DNN) during the learning process. The framework is designed to combine the DNN and a traditional independent inversion workflow and improve the joint inversion result iteratively. The network can be easily extended to incorporate multiphysics without structural changes. Numerical experiments on the joint inversion of 2D DC resistivity data and seismic traveltime are used to validate our method. In addition, this learning-based framework demonstrates excellent generalization abilities when tested on data sets using different geologic structures. It also can handle different sensing configurations and nonconforming discretization.

Research Organization:
Cyentech Consulting LLC, Cypress, TX (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
DOE Contract Number:
SC0020910
OSTI ID:
2421660
Journal Information:
Geophysics, Journal Name: Geophysics Journal Issue: 1 Vol. 88; ISSN 0016-8033
Publisher:
Society of Exploration Geophysicists
Country of Publication:
United States
Language:
English

References (60)

Seismic Impedance Inversion Using Fully Convolutional Residual Network and Transfer Learning journal December 2020
Seismic borehole tomography—Measurement system and field studies journal January 1986
A comparison of geophysical inversion and machine learning in inverse problems conference August 2019
A comparative study of structural similarity and regularization for joint inverse problems governed by PDEs journal December 2018
Joint inversion of refraction and gravity data for the three-dimensional topography of a sediment-basement interface journal October 2002
Joint inversion of multiple geophysical and petrophysical data using generalized fuzzy clustering algorithms journal November 2016
Model Fusion and Joint Inversion journal August 2013
A framework for petrophysically and geologically guided geophysical inversion using a dynamic Gaussian mixture model prior journal August 2019
Characterization of heterogeneous near-surface materials by joint 2D inversion of dc resistivity and seismic data: CHARACTERIZATION OF HETEROGENEOUS NEAR-SURFACE MATERIALS journal July 2003
Joint Inversion of Seismic and Magnetotelluric Data in the Parkfield Region of California Using the Normalized Cross-Gradient Constraint journal December 2014
A theory-guided deep-learning formulation and optimization of seismic waveform inversion journal March 2020
Uncertainty analysis of joint inversion using mixed Lp-norm regularization conference September 2020
Deep Learning-Based Inversion Methods for Solving Inverse Scattering Problems With Phaseless Data journal November 2020
Seismic‐controlled nonlinear extrapolation of well parameters using neural networks journal November 1998
Multiple-domain, simultaneous joint inversion of geophysical data with application to subsalt imaging journal May 2011
Seismic impedance inversion based on cycle-consistent generative adversarial network conference August 2019
A flexible and versatile joint inversion framework using deep learning conference August 2022
First arrival picking using U-net with Lovasz loss and nearest point picking method conference August 2019
Joint Inversion of Audio-Magnetotelluric and Seismic Travel Time Data With Deep Learning Constraint journal September 2021
Salt Reconstruction in Full-Waveform Inversion With a Parametric Level-Set Method journal June 2017
Petrophysically and geologically guided multi-physics inversion using a dynamic Gaussian mixture model journal August 2020
Physics-guided self-supervised learning for monochromatic noise removal conference September 2021
Joint Gramian inversion of geophysical data with different resolution capabilities conference September 2020
RNN-based gradient prediction for solving magnetotelluric inverse problem conference September 2020
Generalized joint inversion of multimodal geophysical data using Gramian constraints journal May 2012
Improved hydrogeophysical characterization using joint inversion of cross-hole electrical resistance and ground-penetrating radar traveltime data: JOINT INVERSION OF CROSS-HOLE ERT AND GPR journal December 2006
Violin Plots: A Box Plot-Density Trace Synergism journal May 1998
Joint inversion of seismic traveltimes and gravity data on unstructured grids with application to mineral exploration journal January 2012
Unveiling the 3D undercover structure of a Precambrian intrusive complex by integrating airborne magnetic and gravity gradient data into 3D quasi-geology model building journal November 2020
Image Quality Assessment: From Error Visibility to Structural Similarity journal April 2004
U-Net: Convolutional Networks for Biomedical Image Segmentation
  • Ronneberger, Olaf; Fischer, Philipp; Brox, Thomas
  • Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III https://doi.org/10.1007/978-3-319-24574-4_28
book November 2015
A Physics-Driven Deep-Learning Inverse Solver for Subsurface Sensing conference July 2020
Deep-learning tomography journal January 2018
3-D inversion of magnetic data journal March 1996
A Supervised Descent Learning Technique for Solving Directional Electromagnetic Logging-While-Drilling Inverse Problems journal November 2020
Physics-driven deep learning joint inversion conference September 2020
Deep learning-enhanced multiphysics joint inversion conference September 2021
Direct reservoir parameter estimation using joint inversion of marine seismic AVA and CSEM data journal May 2006
A Physics-Driven Deep-Learning Network for Solving Nonlinear Inverse Problems journal February 2020
A Bayesian model for gas saturation estimation using marine seismic AVA and CSEM data journal March 2007
Fast Electromagnetic Inversion of Inhomogeneous Scatterers Embedded in Layered Media by Born Approximation and 3-D U-Net journal October 2020
SimPEG: An open source framework for simulation and gradient based parameter estimation in geophysical applications journal December 2015
Analysis of inter-domain coupling constraints for multi-physics joint inversion journal November 2018
Wavefield reconstruction inversion via machine learned functions conference September 2020
A framework for 3-D joint inversion of MT, gravity and seismic refraction data journal November 2010
Deep Perceptual Loss for Multi-Physics Joint Inversion conference January 2022
Machine learning and geophysical inversion — A numerical study journal July 2019
Application of supervised descent method for 2D magnetotelluric data inversion journal June 2020
Joint inversion of marine magnetotelluric and gravity data incorporating seismic constraintsPreliminary results of sub-basalt imaging off the Faroe Shelf journal May 2009
A physics-augmented deep learning method for seismic data deblending conference September 2020
Uncertainty analysis of 3D potential-field deterministic inversion using mixed Lp norms journal October 2021
Deep learning joint inversion of seismic and electromagnetic data for salt reconstruction conference September 2020
Lithological classification assisted by the joint inversion of electrical and seismic data at a control site in northeast Mexico journal February 2010
Deep-learning inversion: A next-generation seismic velocity model building method journal July 2019
Joint inversion of gravity and traveltime data using a level-set-based structural parameterization journal November 2016
Deep Learning Enhanced Joint Inversion of Multiphysics Data with Nonconforming Discretization conference December 2021
Stochastic Seismic Waveform Inversion Using Generative Adversarial Networks as a Geological Prior journal November 2019
Deep learning electromagnetic inversion with convolutional neural networks journal May 2019
Solutions of Ill-Posed Problems (A. N. Tikhonov and V. Y. Arsenin) journal April 1979
Geophysical modeling via simultaneous joint inversion of seismic, gravity, and electromagnetic data: Application to prestack depth imaging journal March 2007

Similar Records

Three-dimensional cooperative inversion of airborne magnetic and gravity gradient data using deep-learning techniques
Journal Article · Sun Dec 31 23:00:00 EST 2023 · Geophysics · OSTI ID:2580172

A New Capability of E4D For 3D Parallel Joint Inversion of DC Resistivity And Traveltime Data on Unstructured Mesh
Technical Report · Wed May 27 00:00:00 EDT 2020 · OSTI ID:1633415

Deep learning multiphysics network for imaging CO2 saturation and estimating uncertainty in geological carbon storage
Journal Article · Tue Aug 30 00:00:00 EDT 2022 · Geophysical Prospecting · OSTI ID:1894062