Medium voltage power lines: Outages data analysis and modelling

Obtaining a reliable model that takes into account the counting nature of outages occurrence in Medium Voltage (MV) lines is a very prominent issue for Distribution System Operators (DSO). As a matter of fact in distribution system planning and management, accurate assessment of reliability performance is essential for making informed decisions. In addition, performance based regulation directs towards a more deep understanding of differences in reliability performance between networks. In this paper two Negative Binomial regression models are proposed for modelling yearly long and short duration outages occurrence in MV feeders; consequently a Monte Carlo simulation is performed in order to obtain a confidence interval for the yearly number of interruptions in a collection of feeders. The results are compared to real failure data from a distribution cluster of feeders in Italy.