改進(jìn)飛蛾捕焰算法在網(wǎng)絡(luò)流量預(yù)測(cè)中的應(yīng)用
發(fā)布時(shí)間:2019-06-11 10:25
【摘要】:傳統(tǒng)BP神經(jīng)網(wǎng)絡(luò)對(duì)網(wǎng)絡(luò)流量時(shí)間序列預(yù)測(cè)精度低和泛化能力弱。為此,提出一種新的優(yōu)化BP神經(jīng)網(wǎng)絡(luò)的方法。通過小波包分解對(duì)網(wǎng)絡(luò)流量進(jìn)行多頻段序列分解,并采用飛蛾縱橫交叉混沌捕焰算法優(yōu)化的神經(jīng)網(wǎng)絡(luò),對(duì)各分解后的子序列進(jìn)行預(yù)測(cè),疊加各子序列的預(yù)測(cè)值,重構(gòu)獲取實(shí)際預(yù)測(cè)結(jié)果。仿真結(jié)果表明,與傳統(tǒng)BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)方法相比,該方法能捕獲網(wǎng)絡(luò)流量的變化規(guī)律,具有較好的預(yù)測(cè)精度、穩(wěn)定性和泛化能力。
[Abstract]:The traditional BP neural network has low prediction accuracy and weak generalization ability for network traffic time series. Therefore, a new method to optimize BP neural network is proposed. The multi-band sequence decomposition of the network traffic is carried out by wavelet packet decomposition, and the neural network optimized by the vertical and horizontal cross chaotic flame trapping algorithm of moths is used to predict the decomposed subsequences and superimpose the predicted values of each subsequences. The actual prediction results are obtained by reconstruction. The simulation results show that compared with the traditional BP neural network prediction method, this method can capture the change law of network traffic, and has better prediction accuracy, stability and generalization ability.
【作者單位】: 廣東工業(yè)大學(xué)計(jì)算機(jī)學(xué)院;
【基金】:國家自然科學(xué)基金(61502108) 廣東省自然科學(xué)基金(2014A030313512,2014A030313629) 廣東省重大科技專項(xiàng)(2014B010111007) 廣東省科技計(jì)劃項(xiàng)目(2013B011304007) 廣東省公益研究與能力建設(shè)專項(xiàng)(2016A010101027)
【分類號(hào)】:TP183;TP393.06
[Abstract]:The traditional BP neural network has low prediction accuracy and weak generalization ability for network traffic time series. Therefore, a new method to optimize BP neural network is proposed. The multi-band sequence decomposition of the network traffic is carried out by wavelet packet decomposition, and the neural network optimized by the vertical and horizontal cross chaotic flame trapping algorithm of moths is used to predict the decomposed subsequences and superimpose the predicted values of each subsequences. The actual prediction results are obtained by reconstruction. The simulation results show that compared with the traditional BP neural network prediction method, this method can capture the change law of network traffic, and has better prediction accuracy, stability and generalization ability.
【作者單位】: 廣東工業(yè)大學(xué)計(jì)算機(jī)學(xué)院;
【基金】:國家自然科學(xué)基金(61502108) 廣東省自然科學(xué)基金(2014A030313512,2014A030313629) 廣東省重大科技專項(xiàng)(2014B010111007) 廣東省科技計(jì)劃項(xiàng)目(2013B011304007) 廣東省公益研究與能力建設(shè)專項(xiàng)(2016A010101027)
【分類號(hào)】:TP183;TP393.06
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 高玉潼;潘成勝;;一種網(wǎng)絡(luò)流量預(yù)測(cè)算法的研究與實(shí)現(xiàn)[J];沈陽理工大學(xué)學(xué)報(bào);2007年04期
2 陳振偉;王茜;黃繼紅;;基于多神經(jīng)網(wǎng)絡(luò)的網(wǎng)絡(luò)流量預(yù)測(cè)模型[J];計(jì)算機(jī)安全;2010年12期
3 于靜;王輝;;基于組合模型的網(wǎng)絡(luò)流量預(yù)測(cè)[J];計(jì)算機(jī)工程與應(yīng)用;2013年08期
4 葉苗;王勇;;基于支持向量回歸學(xué)習(xí)機(jī)的網(wǎng)絡(luò)流量預(yù)測(cè)[J];桂林工學(xué)院學(xué)報(bào);2007年02期
5 張正本;翟海慶;;基于組合模型的網(wǎng)絡(luò)流量預(yù)測(cè)[J];河南機(jī)電高等專科學(xué)校學(xué)報(bào);2008年06期
6 周丹琪;喬國平;張泉方;;基于K-Factor GARMA模型的網(wǎng)絡(luò)流量預(yù)測(cè)[J];江南大學(xué)學(xué)報(bào)(自然科學(xué)版);2008年02期
7 姜明;吳春明;張e,
本文編號(hào):2497116
本文鏈接:http://sikaile.net/guanlilunwen/ydhl/2497116.html
最近更新
教材專著