Quantum-behaved particle swarm optimization (QPSO) is a good optimization technique which has been successfully applied in many research and application areas.But traditional QPSO algorithm is easy to fall into local optimum and the rate of convergence is slow.To solve these problems, an improved algorithm based on dynamic adjustment of the acceleration factor is proposed.The experiments on high dimensional function optimization showed that the improved algorithm has more powerful global exploration ability and faster convergence speed.