摘要: 传统人工神经网络时间序列预测方法难以表达时间序列中的时间累积效应。为此,提出一种基于过程神经元网络的时间序列预测方法。采用双链结构的量子粒子群对过程神经元网络进行训练,以Mackey-Glass混沌时间序列预测为例进行实验。仿真结果表明,该方法的均方误差比普通神经网络低一个数量级。
关键词:
过程神经元网络,
量子粒子群,
双链结构,
时间序列预测,
算法设计,
网络训练
Abstract: Artificial Neural Network(ANN) has difficulty in expression of the temporal accumulation in the time series prediction. A prediction method which uses the Process Neural Networks(PNN) is presented. The algorithm of quantum particle swarm which has double chains structure is used to train the PNN. The effectiveness of the method and training algorithm are proved by the Mackey-Glass time series prediction. Simulation result shows that the mean-square error about this method is reduced one order of magnitude compared with ANN.
Key words:
Process Neural Networks(PNN),
quantum particle swarm,
double chain structure,
time series prediction,
algorithm design,
network training
中图分类号:
刘志刚, 杜娟, 许少华, 李盼池. 基于过程神经元网络的时间序列预测方法[J]. 计算机工程, 2012, 38(5): 199-201.
LIU Zhi-Gang, DU Juan, HU Shao-Hua, LI Fen-Che. Time Series Prediction Method Based on Process Neural Networks[J]. Computer Engineering, 2012, 38(5): 199-201.