Fuzzy decision-making in distribution network restoration

Abstract

This paper describes a model for the restoration of distribution networks in an uncertain environment. The proposed methodology is based on fuzzy mixed integer programming. Customer loads and power flows in the network are considered as triangular fuzzy numbers. The network capacity constraints (line and supply transformer constraints) are treated as "soft". These constraints can be violated up to a certain degree defined by the decision-maker in the utility. The proposed methodology allows decision-maker to analyze different load restoration scenarios by adjusting the values of the degree of acceptance in each constraint related to the power flow in lines and supply transformers. In this way decision-maker can select the bad restoration scenario which provides a maximum cost reduction and satisfies a set of relaxed constraints.