Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach

Electromagnetic responses arising from the interaction of electromagnetic fields with the electrical properties of the subsurface structures provide useful information about such structures. Furthermore, if polarizable materials are present within the investigated structure, the observed data from t...

Full description

Saved in:
Bibliographic Details
Main Author: Mohamed Elkattan
Format: Article
Language:English
Published: The Korean Institute of Electromagnetic Engineering and Science 2025-07-01
Series:Journal of Electromagnetic Engineering and Science
Subjects:
Online Access:https://www.jees.kr/upload/pdf/jees-2025-4-r-301.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849393189761318912
author Mohamed Elkattan
author_facet Mohamed Elkattan
author_sort Mohamed Elkattan
collection DOAJ
description Electromagnetic responses arising from the interaction of electromagnetic fields with the electrical properties of the subsurface structures provide useful information about such structures. Furthermore, if polarizable materials are present within the investigated structure, the observed data from the electromagnetic measurements will contain inherently induced polarization responses, thereby introducing additional parameters for distinguishing subsurface materials. In this regard, the concept of introducing an inversion framework to extract information about induced polarization characteristics from electromagnetic measurements has recently attracted considerable attention. In this paper, an inversion setting is presented to estimate the induced polarization parameters of a stratified, polarizable, and layered medium from scattered electromagnetic fields. The proposed setting handles the inverse problem through a learning procedure that employs a neural network design. Several neural network design factors were tuned to achieve optimal performance. The proposed neural network with tuned design factors was also evaluated under noisy conditions. Error analysis verified the effectiveness of the proposed neural network design in inverting electromagnetic data to derive induced polarization parameters.
format Article
id doaj-art-973fc49f413e4868a2cf9d48cdf2383a
institution Kabale University
issn 2671-7255
2671-7263
language English
publishDate 2025-07-01
publisher The Korean Institute of Electromagnetic Engineering and Science
record_format Article
series Journal of Electromagnetic Engineering and Science
spelling doaj-art-973fc49f413e4868a2cf9d48cdf2383a2025-08-20T03:40:30ZengThe Korean Institute of Electromagnetic Engineering and ScienceJournal of Electromagnetic Engineering and Science2671-72552671-72632025-07-0125430731710.26866/jees.2025.4.r.3013749Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network ApproachMohamed ElkattanElectromagnetic responses arising from the interaction of electromagnetic fields with the electrical properties of the subsurface structures provide useful information about such structures. Furthermore, if polarizable materials are present within the investigated structure, the observed data from the electromagnetic measurements will contain inherently induced polarization responses, thereby introducing additional parameters for distinguishing subsurface materials. In this regard, the concept of introducing an inversion framework to extract information about induced polarization characteristics from electromagnetic measurements has recently attracted considerable attention. In this paper, an inversion setting is presented to estimate the induced polarization parameters of a stratified, polarizable, and layered medium from scattered electromagnetic fields. The proposed setting handles the inverse problem through a learning procedure that employs a neural network design. Several neural network design factors were tuned to achieve optimal performance. The proposed neural network with tuned design factors was also evaluated under noisy conditions. Error analysis verified the effectiveness of the proposed neural network design in inverting electromagnetic data to derive induced polarization parameters.https://www.jees.kr/upload/pdf/jees-2025-4-r-301.pdfelectromagneticinduced polarizationinversionneural network
spellingShingle Mohamed Elkattan
Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach
Journal of Electromagnetic Engineering and Science
electromagnetic
induced polarization
inversion
neural network
title Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach
title_full Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach
title_fullStr Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach
title_full_unstemmed Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach
title_short Reconstruction of Induced Polarization Information from Electromagnetic Data Using Neural Network Approach
title_sort reconstruction of induced polarization information from electromagnetic data using neural network approach
topic electromagnetic
induced polarization
inversion
neural network
url https://www.jees.kr/upload/pdf/jees-2025-4-r-301.pdf
work_keys_str_mv AT mohamedelkattan reconstructionofinducedpolarizationinformationfromelectromagneticdatausingneuralnetworkapproach