Sensitivity Analysis of Impedance Measurement Algorithms Implemented in Intelligent Electronic Devices

Abstract

In this paper a global sensitivity analysis technique for testing impedance measurement algorithms is described. The technique is based on the Analysis of Variance (ANOVA) statistical method. Accuracy of impedance measurement algorithms, when influenced by uncertainties, can be systematically analyzed. This technique divides variance of a measurement algorithm output into components related to uncertain parameters (factors) and interactions between factors. As an application example, we simulate transmission line faults with varying values of fault parameters (factors) according to the Sobol’s quasi-random sampling. The algorithm for automating this task was developed via DIgSILENT Programming Language (DPL). The SIMLAB software is used for generating samples in a factor space according to the Sobol’s quasi-Monte Carlo technique.