首页 | 官方网站   微博 | 高级检索  
     

基于BP神经网络的AHP判断矩阵调整方法
引用本文:于建华,孙首群,杨凡,满微微.基于BP神经网络的AHP判断矩阵调整方法[J].现代制造工程,2011(3).
作者姓名:于建华  孙首群  杨凡  满微微
作者单位:上海理工大学机械学院,上海,200093
基金项目:国家自然科学基金资助项目,国家863计划资助项目,上海市教育委员会重点学科建设项目
摘    要:提出了一种基于BP神经网络的判断矩阵一致性调整方法.该方法在建立判断矩阵一致性调整BP神经网络模型的基础上,通过对原始判断矩阵施之一定的数学变换,构造出完全一致性判断矩阵并将其作为BP神经网络的教师信号.经BP神经网络算法的调整使原始判断矩阵逐渐逼近构造的完全一致性判断矩阵,从而达到一致性指标要求.计算实例表明,此种方法是可行及可靠的.

关 键 词:BP神经网络  层次分析法  判断矩阵  完全一致性矩阵

A method of comparison matrix adjustment based on BP neural network
YU Jian-hua,SUN Shou-qun,YANG Fan,MAN Wei-wei.A method of comparison matrix adjustment based on BP neural network[J].Modern Manufacturing Engineering,2011(3).
Authors:YU Jian-hua  SUN Shou-qun  YANG Fan  MAN Wei-wei
Affiliation:YU Jian-hua,SUN Shou-qun,YANG Fan,MAN Wei-wei(College of Mechanical Engineering,University of Shanghai for Science & Technology,Shanghai 200093,China)
Abstract:A new method based on BP neural network to adjust the consistency of comparison matrix was presented.A BP network model was established.A completely consistent comparison matrix is constructed by mathematical transform of the original comparison matrix which was regarded as the teacher signal of the BP neural network.The original comparison matrix is converted to an approximation of the completely consistent matrix by the BP network algorithm adjustment,so as to satisfy the consistency index requirements.Nu...
Keywords:BP neural network  Analytical Hierarchy Process(AHP)  comparison matrix  completely consistent matrix  
本文献已被 CNKI 万方数据 等数据库收录!
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-23

京公网安备 11010802026262号