Distributed optimal reactive power dispatch based on parallel particle swarm optimisation algorithm

This paper presents a parallel particle swarm optimisation (PPSO) approach for solving the distributed optimal reactive power dispatch (ORPD) problem. The proposed method is a combination of parallel computation based on message passing interface (MPI) and particle swarm optimisation algorithm. After dividing a large scale power system into a group of small subsystems, ORPD problem of each subsystem is solved paralleled. Due to the size of each ORPD problem is reduced, it become easier to solve the optimization problem. Results from each subsystem will be combined together to find solution of whole system. The algorithm is evaluated on the IEEE 118 bus test system on two platforms: a PC cluster and a supercomputer. Test results show that the quality of solutions to the ORPD problems is improved and the calculation time can be reduced enormously so as to fulfill the practical requirement for potential application in power systems operation.