Goto

Layer | Fill | Outline |
---|

Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|

Fill | Stroke |
---|---|

Wan, Le (Department of Geology and Geophysics, University of Utah) | Puahengsup, Pechet (Department of Geology and Geophysics, University of Utah) | Zhdanov, Michael S. (Department of Geology and Geophysics, University of Utah)

**Summary**In this paper we develop a fast 3-D electromagnetic (EM) migration method for marine geophysical exploration. The developed migration algorithm is based on downward extrapolation of the observed EM field using a special form of finite-difference equation for the migration field. It allows us to migrate within the sea-bottom formations the EM signals observed by the sea-bottom receivers. The migration field is subsequently transformed in the resistivity image of the sea-bottom geoelectrical structures. This technique is in an order faster than the conventional inversion. It can be used for fast imaging of the marine magnetotelluric (MT) and controlled-source electromagnetic (CSEM) data in off-shore hydrocarbon (HC) exploration.

algorithm, electromagnetic migration, EM migration, equation, exploration, frequency domain, frequency domain electromagnetic migration, geoelectrical model, ilj, Imaging, inversion, metals & mining, migration, province, receiver, Reservoir Characterization, resistivity image, sea bottom, sea-bottom geoelectrical structure, Upstream Oil & Gas, Zhdanov

Industry:

- Materials > Metals & Mining (1.00)
- Energy > Oil & Gas > Upstream (1.00)

Oilfield Places:

- Europe > Norway > North Sea > Northern North Sea > North Viking Graben > Block 31/6 > Troll Field > Sognefjord Formation (0.99)
- Europe > Norway > North Sea > Northern North Sea > North Viking Graben > Block 31/6 > Troll Field > Fensfjord Formation (0.99)
- Europe > Norway > North Sea > Northern North Sea > North Viking Graben > Block 31/5 > Troll Field > Sognefjord Formation (0.99)
- (5 more...)

SPE Disciplines: Reservoir Description and Dynamics > Reservoir Characterization > Seismic processing and interpretation (0.90)

A thin sheet model consists of one or several horizontal,interpretation of 3-D electromagnetic data.

algorithm, anomaly, conductance, distortion, electric field, electromagnetic data, electromagnetic field, electromagnetic induction, equation, Fainberg, field component, inhomogeneous layer, interpretation, Inversion Algorithm, model parameter, resistive layer, thin sheet model, transverse resistance

The growing use of the controlled-source electromagnetic method (CSEM) for exploration applications has been driving the technical development of data acquisition, as well as three-dimensional (3D) modeling and imaging techniques. However, targeting increasingly complex geological environments also further enhances the problems inherent in large-scale inversion, such as nonuniqueness and resolution issues. In this paper, we report on two techniques to mitigate these problems. We use 3D joint CSEM and MT inversion to improve the model resolution. Further, a hybrid model parameterization approach is presented, where traditional cell-based model parameters are used simultaneously within a parametric inversion.

Large-scale inverse problems are usually under-determined, meaning that there are more unknowns, typically in the form of digitized model meshes, than data. This adds to the problem that errors are associated with every geophysical data. The resulting issue is referred to as the problem of non-uniqueness of inverse solutions. To mitigate this problem and to improve the resolution in an inversion, it is common to take advantage of complementary natures of different geophysical datasets. In electromagnetic imaging, magnetotelluric (MT) data, providing conductivity structure information on a gross scale, can be combined with CSEM data. With the latter method responding stronger to thin resistive targets, the joint CSEM and MT inversion has the potential of limiting ambiguities in the EM data interpretation relevant to many exploration scenarios.

However, even with improved resolution capabilities, the solutions of 3D large-scale cell-based (or pixel-based) inversions with finely sampled models usually are still nonunique. Several strategies have been reported to limit the ambiguities for reconstructed targets and its conductivities. For cell-based problems, model-smoothing constraints are usually applied, limiting the solutions to a class of geologically more meaningful ones, i.e. avoiding too high conductivity contrasts. A different approach is to actually address the under-determinacy by casting the problem into a parametric problem. Usually, particular geometric shapes are assumed in parametric solutions, requiring a priori information. A model parameterization can for example be based on interfaces known from seismic reflection data. The 2D sharp-boundary inversion algorithm by Smith et al. (1999) features a parameterization with variable nodebased boundaries and greatly limits the number of unknowns. Parametric inversion algorithms have also been used for the simultaneous reconstruction of both geometry and conductivity of unknown regions (Commer, 2003; Zhang et al., 2007). The obvious drawback of such methods is the necessity of sufficient background information in order to find a suitable model parameterization.

Here, we propose to use a hybrid approach, overlaying a cell-based inversion for a particular area of interest with a parametric inversion. This combines the advantages of cellbased and structure-based model parameters. We present two joint inversion examples using synthetic CSEM andMT data. The first example employs only cell-based model parameters, and simulates a survey in a marine environment. Second, we present an inversion study for a surface survey, using the hybrid parameterization approach.

Our inversion algorithm’s underlying finite-difference (FD) forward modeling algorithm for EM field simulation solves a modified form of the vector Helmholtz equation for scattered or total electric fields.

Alumbaugh, background, conductivity reconstruction, electromagnetic inversion, frequency, inverse problem, inversion, Inversion Algorithm, iteration, joint inversion example, magnetotelluric data, model parameter, model-based inversion, MT data, optimal conductivity reconstruction, reference list, Reservoir Characterization, resolution, Upstream Oil & Gas

SPE Disciplines:

We carry out the inversion of marine controlled-source electromagnetic data using real coded genetic algorithm to estimate the isotropic resistivity. Unlike linearized inversion methods, genetic algorithms belonging to class of stochastic methods are not limited by the requirement of the good starting models. The objective function to be optimized contains data misfit and model roughness. The regularization weight is used as a temperature like annealing parameter. This inversion is cast into a Bayesian framework where the prior distribution of the model parameters is combined with the physics of the forward problem to estimate the aposteriori probability density function in the model space. The probability distribution derived with this approach can be used to quantify the uncertainty in the estimation of vertical resistivity profile. We apply our inversion scheme on three synthetic data sets generated from horizontally stratified earth models. For all cases, our inversion estimated the resistivity to a reasonable accuracy. The results obtained from this inversion can serve as starting models for linearized/higher dimensional inversion.

Presentation Date: Monday, October 15, 2018

Start Time: 1:50:00 PM

Location: Poster Station 13

Presentation Type: Poster

annual meeting, Artificial Intelligence, conductivity, CSEM data, electromagnetic data, evolutionary algorithm, frequency-domain marine controlled-source electromagnetic data, genetic algorithm, geophysics, inversion, iteration, machine learning, Mallick, marine controlled-source electromagnetic data, model parameter, model space, optimization problem, Reservoir Characterization, resistive layer, resistivity, seg international exposition, synthetic data, Upstream Oil & Gas, well log data

SPE Disciplines:

Technology:

For efficiency, forward and inverse models are matrix of partial derivatives.

algorithm, anomalous conductivity, Artificial Intelligence, background conductivity structure, cell conductivity, conductivity, Conductivity Structure, derivative, electromagnetic inversion, equation, geophysics, Hohmann, inhomogeneity, inverse solution, inversion, iteration, layered earth, matrix, model response, Reservoir Characterization, resistivity, three-dimensional electromagnetic inversion

Thank you!