改进PSO优化BP神经网络的混沌时间序列预测

为提高BP神经网络预测模型的预测准确性，提出了一种基于改进粒子群算法优化BP神经网络的混沌时间序列预测方法。引入自适应变异算子对陷入局部最优的粒子进行变异，改进了粒子群算法的寻优性能; 利用改进粒子群算法优化BP神经网络的权值和阈值，训练BP神经网络预测模型求得最优解。将该预测方法应用到几个典型的非线性系统的混沌时间序列进行有效性验证，结果表明了该方法对典型混沌时间序列具有更好的非线性拟合能力和更高的预测准确性。

Prediction for chaotic time series of optimized BP neural network based on modified PSO

In order to improve forecasting model accuracy of BP neural network, an improved prediction method of optimized BP neural network based on modified Particle Swarm Optimization algorithm（PSO） is proposed. In this modified PSO algorithm, an adaptive mutation operator is proposed in PSO to change positions of the particles which plunge in the local optimization. The modified PSO is used to optimize the weights and thresholds of BP neural network, and then BP neural network is trained to search for the optimal solution. The availability of the proposed prediction method is proven by predicting several typical nonlinear systems. The simulation results have shown that the better fitting and higher accuracy are expressed in this improved method.