Deep learning for joint geophysical inversion of seismic and MT data sets
Глубокое обучение для совместной геофизической инверсии сейсмических и МТ-данных
2021-09-01
SCID: 54.1/ysj5pcap
Discuss with AI
apparent resistivityartificial neural networksjoint geophysical inversionmagnetotelluric dataseismic data
Figures from the paper
Abstract (AI)
Using more than one geophysical technique provides a more reliable way to delineate the subsurface structure than a single geophysical method. In this paper, we present a novel way for the joint inversion of seismic and MT datasets using artificial neural networks. Different rock models are taken to compute the joint relation between the porosity, shale content, velocity and resistivity of the strata. The velocity and resistivity models thus generated were used to compute seismic traces and apparent resistivity curves, the dataset used for training and testing the neural network. Such a method of performing joint inversion is advantageous when we want to check the reliability of our Machine Learning model in regions of low velocity or when a particular physical property does not vary across the layers while another does since, unlike other optimisation schemes used for the geophysical inversion, we do not need to rerun the model for every initial model. The efficacy of artificial neural networks (ANN) to perform the join inversion is tested, and in the process, a particular type of ANN architecture is developed.
Key Findings
1
A novel artificial-neural-network approach performs joint inversion of seismic and magnetotelluric datasets using their coupled subsurface-property relationships.
2
Joint ANN inversion can assess model reliability in low-velocity regions and cases where one physical property remains constant while another varies.
3
Synthetic rock models link porosity and shale content with seismic velocity and electrical resistivity to generate training and testing data.
4
The generated velocity and resistivity models produce seismic traces and apparent-resistivity curves for training and evaluating the neural network.
5
Unlike conventional optimization-based inversion, the approach does not require rerunning the model for every initial model; a dedicated ANN architecture is developed and tested.
Research Object
Subsurface rock strata characterized by porosity, shale content, seismic velocity, and electrical resistivity, observed through seismic and magnetotelluric (MT) data
Research Subject
Joint inversion of seismic and MT responses to infer the coupled relationships among porosity, shale content, seismic velocity, and resistivity using artificial neural networks
Publication Details
Publication Date
2021-09-01
Journal
Publisher
ISSN
Cited by
6
Access Type
Author Information
Download PDF
Subscribe to digest