基于改进BP神经网络的混沌时间序列预测方法对比

The Chaotic Time Series Prediction Comparison Based on Improved BP Neural Network Methods

针对BP神经网络预测混沌时间序列存在的易陷入局部极小值和收敛速度较慢的问题,选取了两种改进预测模型,即GA-BP预测模型和PSO-BP预测模型.并将这两种模型对Lorenz混沌时间序列进行了预测比较实验.实验表明,两种改进模型比BP神经网络预测模型具有更好的预测性能,并且PSO-BP预测模型较GA-BP预测模型的预测精度更高.

Based on the problem that BP neural network prediction of chaotic time series is easy to fall into local minimum and slow convergence speed, we chose two kinds of improved prediction model, namely the GA-BP prediction model and PSO-BP prediction model. Experimental results show that two kinds of improved model has better prediction performance than the BP neural network prediction model, and the accuracy of PSO-BP prediction model is better than GA-BP model.