Skip to Main Content
Optimizing multi-objective reactive power dispatch in power systems is an effective way of improving voltage quality, decreasing active power losses and increasing voltage stability margin. This is a non-linear, constrained, non-convex, mixed discrete-continuous variable problem. Recently, computational intelligence-based methods such as genetic algorithms (GAs), differential evolution (DE) algorithms, particle swarm optimization (PSO) algorithms and immune algorithms (IAs) have been applied to solve this problem. This paper employs dynamic multi-group self-adaptive differential evolution (DMSDE) algorithm to solve multi-objective reactive power optimization problem. In DMSDE, the population is divided into multiple groups which exchange information dynamically. Further, in the mutation phase, the best vector among the three randomly vectors is used as the base vector while the difference vector is determined by the remaining two vectors. Moreover, two parameters, F and CR, are self-adapted. The presented method is tested on IEEE 30-bus, IEEE 57-bus and IEEE 118-bus power systems. The numerical results, when compared with other algorithms, show that DMSDE is an efficient tool to solve dispatch reactive power flow problem.