Estimation of Soil Parameters From Driven Rod Measurements

This paper presents a methodology for the statistical estimation of soil parameters from usual measurements. In particular, driven rod measurements and the fall of potential method are employed to estimate the parameters of an equivalent two layer soil model. Statistical estimation techniques, coupled with an accurate computer model for grounding system analysis are employed. These techniques provide a) the best estimate of the earth model parameters in the least square sense and b) a measure of the quality of the estimates. Specifically, the methodology provides a quantitative measure of error in the estimated parameters expressed in terms of probability that error will not exceed a given value. The procedure has been applied using field test data and has been shown to provide a satisfactory soil model.