In this study a hybrid particle swarm optimization (PSO) algorithm is proposed, called CLONEPAC-PSO. Specifically, the local-neighborhood variant of PSO is enhanced with the constriction factor approach and a new operator, reflecting the physical force of passive congregation, appeared in swarms. Passive congregation is an attraction of a particle to other group members where there is no display of social behavior. The results obtained by CLONEPAC-PSO in the reactive power and voltage control of an IEEE 30-bus test system are compared with those given by the global variant PSO algorithm based on passive congregation and another evolutionary algorithm, demonstrating the superiority of CLONEPAC-PSO.