Reactive power dispatch and planning using a non-linear branch-and-bound algorithm

Abstract

This study proposes the use of a non-linear branch-and-bound (B&B) algorithm to solve the reactive power dispatch and planning problem of an electrical power system. The problem is formulated as a mixed integer non-linear programming (MINLP) problem. The MINLP is relaxed resulting in a set of non-linear programming (NLP) problems, which are solved at each node of the B&B tree through a primal dual-interior point algorithm. The non-linear B&B algorithm proposed has special fathoming criteria to deal with non-linear and multimodal optimisation models. The fathoming tests are redefined, adding a safety margin value to the objective function of each NLP problem before they are fathomed through the objective function criteria, avoiding convergence to local optimum solutions. The results are presented using three test systems from the specialised literature. The B&B algorithm found several optimum local solutions and the best solution was found after solving some NLP problems, with little computational effort.