Parallel optimal reactive power flow based on cooperative co-evolutionary differential evolution and power system decomposition

Differential evolution (DE) is a promising evolutionary algorithm for solving optimal reactive power flow problems, but it requires relatively large population to avoid premature convergence. In order to overcome this disadvantage, a novel decomposition and coordination method based on the cooperative co-evolutionary architecture and the voltage-var sensitivity-based power system decomposition technique is proposed and incorporated with DE in this paper. It is implemented with a three-level parallel computing topology on a PC-cluster. Based on the IEEE 118-bus system test case, the effectiveness of the proposed method has been verified by comparison with the parallel basic DE not using the decomposition and coordination technique.