Heurística construtiva e otimização bioinspirada aplicadas à expansão de sistemas de transmissão de energia elétrica

Abstract

The static transmission expansion planning of electrical systems problem consists in determining, among a pre-defined set of candidate expansion circuits, the ones that must be built to minimize the operational costs (deficit) and investment costs in the electrical networks thus meeting the forecast demand in a given planning horizon. This hard-solution optimization problem presents some particular characteristics, such as: (i) non-convex solution region, which means a large number of feasible solutions leading most of the algorithms, used in this situation, to converge to a local optimum; (ii) the combinatorial nature of the planning process which usually leads to the combinatorial explosion related to investment alternatives, resulting in a high computational effort; (iii) the existence of islanded electrical systems. These features illustrate the main difficulties in the development of fast, efficient and robust algorithms to solve the static planning of the transmission expansion of electrical systems. Considering this problem, this work proposes a two-step resolution strategy: (a)Initially, a constructive heuristic algorithm is used in order to obtain a feasible initial solution for the problem; (B) Since this initial solution is known, it is transferred to the multimodal optimization process, based on the echolocation phenomenon. The echolocation is a recent multimodal optimization method when compared with other bioinspired multimodal methods and its application on electric power systems is still incipient so, its utilization may be a motivation. The obtained results indicate that the proposed solution strategy provides increased efficiency for the multimodal optimization process by the search for optimality, since the optimal solution can be obtained in a small number of iterations of bioinspired search process.