Use of Rusck Formula and Multi-Nonlinear Regression to Discard Events Required to Calculate the Lightning Performance of Overhead Lines

Abstract

In this work, two heuristic techniques used to reduce the number of simulations of lightning-induced voltages, which are required to assess the lightning performance of distribution lines according to the Monte Carlo methodology, are presented. The first uses the Rusck Formula to discard the events that do not cause damage to the line. The second is based on a multi-nonlinear regression of the previously simulated events. Both techniques were applied to estimate the flashover rate of straight and complex lines above flat ground with finite conductivities. As a result, between 80 % to 90 % of the total number of simulations were eliminated with a reduced error.