A Neural Network-Based Hybrid Framework for Least-Squares Inversion of Transient Electromagnetic Data

TY - JOUR

T1 - A Neural Network-Based Hybrid Framework for Least-Squares Inversion of Transient Electromagnetic Data

AU - Asif, Muhammad Rizwan

AU - Bording, Thue S.

AU - Maurya, Pradip K.

AU - Zhang, Bo

AU - Fiandaca, Gianluca

AU - Grombacher, Denys J.

AU - Christiansen, Anders V.

AU - Auken, Esben

AU - Larsen, Jakob Juul

N1 - Publisher Copyright: IEEE Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2022

Y1 - 2022

N2 - Inversion of large-scale time-domain transient electromagnetic (TEM) surveys is computationally expensive and time-consuming. The calculation of partial derivatives for the Jacobian matrix is by far the most computationally intensive task, as this requires calculation of a significant number of forward responses. We propose to accelerate the inversion process by predicting partial derivatives using an artificial neural network. Network training data for resistivity models for a broad range of geological settings are generated by computing partial derivatives as symmetric differences between two forward responses. Given that certain applications have larger tolerances for modeling inaccuracy and varying degrees of flexibility throughout the different phases of interpretation, we present four inversion schemes that provide a tunable balance between computational time and inversion accuracy when modeling TEM datasets. We improve speed and maintain accuracy with a hybrid framework, where the neural network derivatives are used initially and switched to full numerical derivatives in the final iterations. We also present a full neural network solution where neural network forward and derivatives are used throughout the inversion. In a least-squares inversion framework, a speedup factor exceeding 70 is obtained on the calculation of derivatives, and the inversion process is expedited ~36 times when the full neural network solution is used. Field examples show that the full nonlinear inversion and the hybrid approach gives identical results, whereas the full neural network inversion results in higher deviation but provides a reasonable indication about the overall subsurface geology.

AB - Inversion of large-scale time-domain transient electromagnetic (TEM) surveys is computationally expensive and time-consuming. The calculation of partial derivatives for the Jacobian matrix is by far the most computationally intensive task, as this requires calculation of a significant number of forward responses. We propose to accelerate the inversion process by predicting partial derivatives using an artificial neural network. Network training data for resistivity models for a broad range of geological settings are generated by computing partial derivatives as symmetric differences between two forward responses. Given that certain applications have larger tolerances for modeling inaccuracy and varying degrees of flexibility throughout the different phases of interpretation, we present four inversion schemes that provide a tunable balance between computational time and inversion accuracy when modeling TEM datasets. We improve speed and maintain accuracy with a hybrid framework, where the neural network derivatives are used initially and switched to full numerical derivatives in the final iterations. We also present a full neural network solution where neural network forward and derivatives are used throughout the inversion. In a least-squares inversion framework, a speedup factor exceeding 70 is obtained on the calculation of derivatives, and the inversion process is expedited ~36 times when the full neural network solution is used. Field examples show that the full nonlinear inversion and the hybrid approach gives identical results, whereas the full neural network inversion results in higher deviation but provides a reasonable indication about the overall subsurface geology.

KW - Computational modeling

KW - Conductivity

KW - Data models

KW - Forward modeling

KW - Jacobian matrices

KW - Jacobian matrix

KW - Logic gates

KW - Mathematical model

KW - Neurons

KW - inverse modeling

KW - neural networks

KW - transient electromagnetics (TEM).

UR - http://www.scopus.com/inward/record.url?scp=85105852890&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2021.3076121

DO - 10.1109/TGRS.2021.3076121

M3 - Journal article

AN - SCOPUS:85105852890

SN - 0196-2892

VL - 60

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

M1 - 4503610

ER -