Lightning performance assessment of power distribution lines by means of stratified sampling Monte Carlo method

This paper proposes the application of a recursive stratified sampling technique for improving the assessment of the lightning performance of power distribution lines through the application of the Monte Carlo method. The accuracy of the results is expressed by means of the confidence interval of the estimated annual number of expected overvoltages. This paper shows the results for a single overhead line, with and without surge arresters, and the case of a feeder with laterals. The adoption of the stratified sampling techniques allows for a computational time reduction up to more than 75% with respect to a standard Monte Carlo procedure. The conceived calculation procedure is applicable to any type of network, regardless of the presence of nonlinear devices and avoiding the application of any heuristic rule.