Interruption modelling in medium voltage electrical networks

Failures frequently occur in developing countries' electrical energy distribution networks. This paper proposes an approach to network reliability through modelling the interruptions on medium voltage lines. This modelling is based on a representative feeder sampling of a source station in an urban area. In order to determine the probability law governing these interruptions, statistical techniques were used: density estimation using the kernel method and approximation by the least squares. The results we obtained show that, from the quality of the equipment and their maintenance, interruptions for a given network follow a truncated and shifted gamma distribution or a truncated normal law. For the managers of such systems, these results would allow, amongst other things, reduction in the probability of failure, thus improving operational safety on electricity distribution with medium voltage lines. (C) 2002 Elsevier Science Ltd. All rights reserved.