Abstract: To meet the requirements of building a core backbone network and raise the ability to support important loads, which was presented by the differentiate planning and design, a global searching algorithm called quantum binary particle swarm optimization（QBPSO） was proposed in this paper on the basis of the features of searching model. The niche technique and the crossover operation of genetic algorithm were introduced to increase the global searching ability of the QBPSO. At the same time, the graph repair strategy was used to improve the efficiency of the QBPSO and the improved QBPSO was applied to core backbone network searching in IEEE-118 model. Simulation results show that this method can effectively search out the core backbone network, which can meet the requirements of flow restraint and connectivity.