I. Introduction
Reactive power dispatch (RPD) problem is one of complex optimization problems. It determines the optimal operating state of a power system in order to minimize the transmission power losses without jeopardizing particular physical and operating constraints. Its control variables are the generator voltages, tap ratios of transformers, and reactive power injection of volt-ampere-reactive (VAR) sources. The generator voltage magnitudes are continuous nature whereas the transformer tap settings, and reactive power injected from capacitor banks are discrete variables. Thus, the RPD problem is considered as a highly nonlinear, nonconvex optimization problem, consisting of both continuous and discrete control variables [1]–[2] . It has been solved effectively by conventional optimization techniques [3] such as interior point methods [4], [5], linear programming (LP) [6], nonlinear programming [7], quadratic programming [8], mixed integer nonlinear programming [9], and sequential quadratic programming [10].