Abstract: In order to improve the neural network structure and setting method of parameters, based on the Particle Swarm Optimization (PSO) and BP Neural Network (BPNN), an algorithm of BP neural network optimized Improved Particle Swarm Optimization (IPSOBPNN) is proposed. In the algorithm, PSO is used to obtain better network initial threshold and weight so as to compensate the defect of connection weight and thresholds choosing of BPNN, thus BPNN can have faster convergence and greater learning ability. The efficiency of the proposed prediction method is tested by the simulation of the chaotic time series of Duffing system. The simulations results show that the proposed method has higher forecasting accuracy compared with the BPNN and BP neural network optimized Particle Swarm Optimization (PSOBPNN), so it is proved that the algorithm is feasible and effective in the chaotic time series.