基于改进灰狼优化算法的网络流量预测模型

Network traffic predicting model based on improved grey wolf optimization algorithm

针对传统网络流量预测模型泛化能力弱和准确度低的缺点,提出一种组合小波包分解(WPD)和灰狼横纵多维混沌寻优算法(CCGWO)优化Elman神经网络的短期网络流量预测模型(WPD-CCGWO-ELMAN)。网络流量在小波包的作用下分解成多个频段序列,各子序列通过CCGWO-ELMAN神经网络优化模型进行单步或多步预测处理,然后重构并叠加各预测值,得到未来短时间段内的网络流量值。实验结果表明,该模型具有较好的预测精度和鲁棒性,并能掌握网络流量时间序列的变化规律。

In order to solve the problem caused by the traditional network traffic forecasting model＇s weak generalization ability and low convergence rate,this paper proposed a model of combining WDP and CCGWO to optimize elman nenural network for short-term network traffic forecasting（ WPD-CCGWO-ELMAN）. Network traffic was decomposed into multiple band series by wavelet packet. While the CCGWO-ELMAN optimized neural network was enrolled to predict the sub-band series by singlestep or multi-step. And by superimposing and reconstructing all the forecasting series,it got the traffic value of short period in the near future. The experimental results show that the model has high convergence rate and robustness,which can grasp the change of network traffic time series.