The fields from a finite electrical dipole—A new computational approach

Abstract

Controlled‐source, frequency‐domain, and time‐domain electromagnetic methods require accurate, fast, and reliable methods of computing the electric and magnetic fields from the source configurations used. Except for small magnetic dipole sources, all electric and magnetic sources are composed of lengths of straight wire, which may be grounded. If the source‐receiver separation is large enough, the composite electrical dipoles may be considered to be infinitely small, and in a 1-D earth model the fields are expressed as Hankel transforms of an input function, which depends only on the model parameters. The Hankel transforms can be evaluated using the digital filter theory of fast Hankel transforms. However, the approximation of the infinitely small dipole is not always valid, and fields from a finite electrical dipole must be calculated. Traditionally, this is done by numerical integration of the fields from an infinitesimal dipole, thus increasing computation time considerably. The fields from the finite electrical dipole are expressed as Hankel transforms and as integrals of Hankel transforms. The theory of fast Hankel transforms is extended to include integrals of Hankel transforms, and a method is devised for calculating the filter coefficients. Unlike the fast Hankel transform, the computation involved in the integrated Hankel transforms is not a true convolution, and so a set of filter coefficients must be calculated for each source‐receiver configuration. Furthermore, the method is extended to include the calculation of potential differences where one more integration is involved, which is what is actually measured in the field. The computation of filter coefficients is very fast, and for standard configurations, the coefficients need be computed only once. The method is as fast, accurate, and reliable as the fast Hankel transforms method, and is up to an order of magnitude faster than the usual numerical integration.