##### A New Physics-based Approach for Estimating a Buried Object’s Location, Orientation and Magnetic Polarization from EMI Data

Authors:
Fridon Shubitidze, David Karkashadze, Ben Barrowes, Irma Shamatava and Kevin O’Neill

Journal name: Journal of Environmental and Engineering Geophysics

Issue: Vol 13, No 3, September 2008 pp. 115 - 130

Special topic: Unexploded Ordnance

Language: English

Info: Article, PDF ( 2.14Mb )

Summary:

A new physics-based expression is presented for determining a buried object’s location, orientation and magnetic polarizibility. The approach assumes the target exhibits a dipolar response and requires only three global values: a magnetic field vector H, a vector potential A and a scalar magnetic potential y, all at a single location in space. Among these values, only the scattered magnetic field, H, is measurable with current electromagnetic induction sensors.
Therefore, in order to estimate the scattered magnetic scalar and vector potentials from data, a numerical technique called the normalized surface magnetic source (NSMS) method is employed. Originally, in the NSMS model, the scattered magnetic field outside the object is reproduced mathematically by equivalent magnetic charges distributed on a three-dimensional (3-D) closed surface. Here, a two-dimensional (2-D) implementation of the NSMS that uses
elementary magnetic dipoles, instead of magnetic charges distributed on a planar surface placed under the measurement grid, is utilized. These sources are used to estimate the scattered magnetic field’s vector potential A and scalar magnetic potential y without a priori knowledge of the object’s location and orientation. The amplitudes of the NSMS are determined by matching the measured magnetic field with the NSMS modeled field. Once the NSMS amplitudes are determined, H, A, and y are simulated on or above the measurement grid. The
theoretical basis of the new approach, as well as the practical realization of the 2-D NSMS algorithm used to estimate H, A, and y above the measurement grid from actual data, is illustrated. Several numerical and experimental tests for actual EMI sensors are presented.

Back to the article list