Statistical analysis and interpretation of resistivity sounding data in earthing grid design

This paper presents a statistical analysis and interpretation of resistivity sounding data, with the objective of establishing an optimal earth model for rocky soil. The parameters obtained are used as input data for a CAD of the earthing grids of an electric power substation. In this analysis, the soil is represented as several vertical zones, each of which is modelled as a set of horizontal layers. Field observations are carried out at each sounding point using general configurations of four symmetrical electrodes; the Schlumberger and Wenner configurations are special cases of these. Apparent resistivity is computed from the field measurements using a new expression which takes into account the cylindrical form of the electrodes. Sample values of the resistivity transform are then computed from the field apparent resistivity data using an exponential approximation of Stefanescu kernel function. The observed soil is subdivided into zones using a statistical analysis of the field resistivity transform data. Parameters for the horizontally stratified earth model are obtained by an iterative interpretation of the average resistivity transform of a single zone. During the iterative interpretation, the sample values of the average resistivity transform computed from field observation data are compared with those computed from the layer model parameters. The comparisons are efficient, since they use a variant of the Marquardt method.