Validation of Voltage Sags Short Term Measurements Based on Predicted Stochastic Simulation

This work presents a procedure to validate sags monitoring results based on a stochastic assessment of voltage sag indices. The main practical use of this methodology is to analyze the accuracy of sag indices obtained from very short monitoring periods. Probabilistic models of several factors can be taken into account: lines and busbars fault rate, pre-fault voltage, fault type distribution, fault location uncertainty and fault resistance distribution. Confidence intervals based on the percentile method and hypothesis tests (ANOVA and t-est) are the statistical tools selected to perform the validation of voltage sags magnitude and frequency. A case study based on the evaluation of a six-month monitoring period shows the applicability of the proposed methodology.