Electric distribution system planning with fuzzy injections

Electric distribution planning problem deals with a great number of variables, integer and real, and presents consumptions and forecasting uncertainties which most of times cannot be modeled in a probabilistic way. The inclusion of independent producers, which generate without control from utilities, introduce additional uncertainties. Moreover, there is a multiobjective problem, once it is intended to minimize investment and exploring costs, maximizing reliability and other consumer’s satisfaction criteria. This contribution presents a multi-stage formulation of this problem. Uncertainties are fuzzy modeled and the original mixed-integer problem is simplified in a linear one. The degree of violation of technical restrictions leads to the definition of robustness and severity indices which allow an additional characterization of alternatives. As final result the model produces a limited number of non dominated solutions which are to be analyzed, a posteriori, by the decision maker. This contribution also includes an illustrative example of electric distribution planning when independent production is considered.