Optimal planning of secondary distribution systems considering distributed generation and network reliability

In this paper a new methodology for optimal planning of low voltage distribution systems is presented. The problem is formulated as a mixed integer non-linear model, which in addition to traditional aspects (costs of installing and upgrading of elements, and operative costs), considers penetration of distributed generators (DG) and the network reliability. The DG are modeled as PQ-nodes and the network reliability is involved by evaluating the non-supplied energy in the objective function. The set of constrains are associated with technical and operative characteristics of the electric networks. In order to take into account the unbalance load, a three-phase model of the elements is used. To solve the problem a tabu search algorithm is employed. To verify the proposed methodology a test distribution system is considered, obtaining results of lower investment costs compared with those acquired in traditional expansion plans.