首页 | 本学科首页   官方微博 | 高级检索  
     

基于小波神经网络的电梯交通流预测
引用本文:黄敏,崔宝同,顾树生.基于小波神经网络的电梯交通流预测[J].控制与决策,2006,21(5):589-592.
作者姓名:黄敏  崔宝同  顾树生
作者单位:1. 江南大学,控制科学与工程研究中心,江苏,无锡,214122
2. 东北大学,信息科学与工程学院,沈阳,110004
基金项目:国家自然科学基金项目(60274024).
摘    要:小波神经网络随着输入维数的增加,网络参数将呈指数倍增加,导致收敛速度下降.在研究统计学理论的基础上,提出了以结构风险最小化为目标的训练方法.该方法最大限度地保证了网络的泛化能力.将该网络应用于电梯交通流的预测,得到了比传统BP神经网络更优的效果.

关 键 词:小波神经网络  统计学习理论  结构风险最小化  电梯交通流
文章编号:1001-0920(2006)05-0589-04
收稿时间:2005-02-23
修稿时间:2005-06-20

Elevator Traffic Flow Prediction Based on Wavelet Neural Networks
HUANG Min,CUI Bao-tong,GU Shu-sheng.Elevator Traffic Flow Prediction Based on Wavelet Neural Networks[J].Control and Decision,2006,21(5):589-592.
Authors:HUANG Min  CUI Bao-tong  GU Shu-sheng
Affiliation:1. Control Science and Engineering Research Center, Southern Yangtze University, Wuxi 214122, China; 2. College of Information Science and Engineering, Northeastern University, Shenyang 110004, China.
Abstract:The number of parameters of wavelet neural networks(WNN) increases by exponential form with input dimension and the convergence speed decreases.An algorithm is presented through using structural risk minimization(SRM) based on statistical learning theory.The novel algorithm can ensure great probability for global optimization.WNN based on SRM is also used to solve the problem of traffic flow prediction of elevator system,more optimal results than typical BP network are obtained.
Keywords:Wavelet neural networks  Statistical learning theory  Structural risk minimization  Elevator traffic flow
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《控制与决策》浏览原始摘要信息
点击此处可从《控制与决策》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号