Domain reduction using GRASP construction phase for transmission expansion planning problem

Abstract

This paper proposes a new strategy to reduce the combinatorial search space of a mixed integer linear programming (MILP) problem. The construction phase of greedy randomized adaptive search procedure (GRASP-CP) is employed to reduce the domain of the integer variables of the transportation model of the transmission expansion planning (TM-TEP) problem. This problem is a MILP and very difficult to solve specially for large scale systems. The branch and bound (BB) algorithm is used to solve the problem in both full and the reduced search space. The proposed method might be useful to reduce the search space of those kinds of MILP problems that a fast heuristic algorithm is available for finding local optimal solutions. The obtained results using some real test systems show the efficiency of the proposed method. © 2012 Springer-Verlag.