• Geophysics and Geophysical Exploration
• /
• v.5 no.4
• /
• pp.316-322
• /
• 2002

A computationally efficient inversion scheme has been developed using the extended Born or localized nonlinear approximation to analyze electromagnetic fields obtained in a single-hole environment. The medium is assumed to be cylindrically symmetric about the borehole, and to maintain the symmetry vertical magnetic dipole source is used throughout. The efficiency and robustness of an inversion scheme is very much dependent on the proper use of Lagrange multiplier, which is often provided manually to achieve desired convergence. In this study, an automatic Lagrange multiplier selection scheme has been developed to enhance the utility of the inversion scheme in handling field data. The inversion scheme has been tested using synthetic data to show its stability and effectiveness.

• Geophysics and Geophysical Exploration
• /
• v.8 no.2
• /
• pp.129-136
• /
• 2005

The damped least-squares inversion has become a most popular method in finding the solution in geophysical problems. Generally, the least-squares inversion is to minimize the object function which consists of data misfits and model constraints. Although both the data misfit and the model constraint take an important part in the least-squares inversion, most of the studies are concentrated on what kind of model constraint is imposed and how to select an optimum regularization parameter. Despite that each datum is recommended to be weighted according to its uncertainty or error in the data acquisition, the uncertainty is usually not available. Thus, the data weighting matrix is inevitably regarded as the identity matrix in the inversion. We present a new inversion scheme, in which the data weighting matrix is automatically obtained from the analysis of the data resolution matrix and its spread function. This approach, named 'extended active constraint balancing (EACB)', assigns a great weighting on the datum having a high resolution and vice versa. We demonstrate that by applying EACB to a two-dimensional resistivity tomography problem, the EACB approach helps to enhance both the resolution and the stability of the inversion process.

• Geophysics and Geophysical Exploration
• /
• v.17 no.2
• /
• pp.95-103
• /
• 2014

This paper presents a practical inversion method for recovering a three-dimensional (3D) resistivity model and static shifts simultaneously. Although this method is based on a Gauss-Newton approach that requires a sensitivity matrix, the computer time can be greatly reduced by implementing a simple and effective procedure for updating the sensitivity matrix using the Broyden's algorithm. In this research, we examine the approximate inversion procedure and the weighting factor ${\beta}$ for static shifts through inversion experiments using synthetic MT data. In methods using the full sensitivity matrix constructed only once in the iteration process, a procedure using the full sensitivity in the earlier stage is useful to produce the smallest rms data misfit. The choice of ${\beta}$ is not critical below some threshold value. Synthetic examples demonstrate that the method proposed in this paper is effective in reconstructing a 3D resistivity structure from static-shifted MT data.

• Proceedings of the KSEEG Conference
• /
• 2001.04a
• /
• pp.332-335
• /
• 2001

• Geophysics and Geophysical Exploration
• /
• v.24 no.4
• /
• pp.149-157
• /
• 2021

Given that induced polarization (IP) and direct current (DC) resistivity surveys are similar in terms of data acquisition, most DC resistivity systems are equipped with a time-domain IP data acquisition function. In addition, the time-domain IP data include the DC resistivity values. As such, IP and DC resistivity data are intimately linked, and the inversion of IP data is a two-step process based on DC resistivity inversions. Nevertheless, IP surveys are rarely applied, in contrast to DC resistivity surveys, as proper inversion software is unavailable. In this study, through numerical modeling and inversion experiments, we analyze the problems with the conventional inverse mapping technique used to invert time-domain IP data. Furthermore, we propose a modified inverse mapping technique that can effectively suppress inversion artifacts. The performance of the technique is confirmed through inversions applied to synthetic IP data.

• Geophysics and Geophysical Exploration
• /
• v.24 no.4
• /
• pp.171-179
• /
• 2021

We outline a process for estimating Cole-Cole parameters from time-domain induced polarization (IP) data. The IP transients are all inverted to 2D Cole-Cole earth models that include resistivity, chargeability, relaxation time, and the frequency exponent. Our inversion algorithm consists of two stages. We first convert the measured voltage decay curves into time series of current-on time apparent resistivity to circumvent the negative chargeability problem. As a first step, a 4D inversion recovers the resistivity model at each time channel that increases monotonically with time. The desired intrinsic Cole-Cole parameters are then recovered by inverting the resistivity time series of each inversion block. In the second step, the Cole-Cole parameters can be estimated readily by setting the initial model close to the true value through a grid search method. Finally, through inversion procedures applied to synthetic data sets, we demonstrate that our algorithm can image the Cole-Cole earth models effectively.

• Geophysics and Geophysical Exploration
• /
• v.11 no.4
• /
• pp.326-334
• /
• 2008

The resistivity method has been used to image the electrical properties of the subsurface. Especially, this method has become suitable for monitoring since data could be rapidly and automatically acquired. In this study, we developed a time-lapse inversion algorithm for the interpretation of resistivity monitoring data. The developed inversion algorithm imposes a big penalty on the model parameter with small change, while a minimal penalty on the model parameter with large change compared to the reference model. Through the numerical experiments, we can ensure that the time-lapse inversion result shows more accurate and focused image where model parameters have changed. Also, applying the timelapse inversion method to the leakage detection of an embankment dam, we can confirm that there are three major leakage zones, but they have not changed over time.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.227-241
• /
• 2022

Full waveform inversion (FWI) in the field of seismic data processing is an inversion technique that is used to estimate the velocity model of the subsurface for oil and gas exploration. Recently, deep learning (DL) technology has been increasingly used for seismic data processing, and its combination with FWI has attracted remarkable research efforts. For example, DL-based data processing techniques have been utilized for preprocessing input data for FWI, enabling the direct implementation of FWI through DL technology. DL-based FWI can be divided into the following methods: pure data-based, physics-based neural network, encoder-decoder, reparameterized FWI, and physics-informed neural network. In this review, we describe the theory and characteristics of the methods by systematizing them in the order of advancements. In the early days of DL-based FWI, the DL model predicted the velocity model by preparing a large training data set to adopt faithfully the basic principles of data science and apply a pure data-based prediction model. The current research trend is to supplement the shortcomings of the pure data-based approach using the loss function consisting of seismic data or physical information from the wave equation itself in deep neural networks. Based on these developments, DL-based FWI has evolved to not require a large amount of learning data, alleviating the cycle-skipping problem, which is an intrinsic limitation of FWI, and reducing computation times dramatically. The value of DL-based FWI is expected to increase continually in the processing of seismic data.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2006.06a
• /
• pp.131-135
• /
• 2006

A seismic waveform inversion for prestack seismic data based on the Gauss-Newton method is presented. The Gauss-Newton method for seismic waveform inversion was proposed in the 80s but has rarely been studied. Extensive computational and memory requirements have been principal difficulties. To overcome this, we used different sizes of grids in the inversion stage from those of grids in the wave propagation simulation, temporal windowing of the simulation and approximation of virtual sources for calculating partial derivatives, and implemented this algorithm on parallel supercomputers. We show that the Gauss-Newton method has high resolving power and convergence rate, and demonstrate potential applications to real seismic data.

• Geophysics and Geophysical Exploration
• /
• v.9 no.1
• /
• pp.86-92
• /
• 2006

The full-waveform inversion algorithm using normalised seismic wavefields can avoid potential inversion errors due to source estimation required in conventional full-waveform inversion methods. In this paper, we have modified the inversion scheme to install a weighted smoothness constraint for better resolution, and to implement a staged approach using normalised wavefields in order of increasing frequency instead of inverting all frequency components simultaneously. The newly developed scheme is verified by using a simple two-dimensional fault model. One of the most significant improvements is based on introducing weights in model parameters, which can be derived from integrated sensitivities. The model-parameter weighting matrix is effective in selectively relaxing the smoothness constraint and in reducing artefacts in the reconstructed image. Simultaneous multiple-frequency inversion can almost be replicated by multiple single-frequency inversions. In particular, consecutively ordered single-frequency inversion, in which lower frequencies are used first, is useful for computation efficiency.