基于群组AHP的悬索桥状态评估指标权重确定  

Determination of Index Weights in Suspension Bridge Condition Assessment Based on Group-AHP

作　　者：许翔[1] 黄侨[1] 任远[1] 刘小玲[1] XU Xiang;HUANG Qiao;REN Yuan;LIU Xiaoling

机构地区：[1]东南大学交通学院,江苏南京210096

出　　处：《湖南大学学报（自然科学版）》2018年第3期122-128,共7页Journal of Hunan University:Natural Sciences

基　　金：国家自然科学基金资助项目(51208096);江苏省交通运输科技项目重大科技专项(2014Y02);江苏省交通运输科技项目(2012Y25)~~

摘　　要：为确定悬索桥综合技术状况评估模型中指标的权重,应用群组AHP方法确定相应的评估指标权重.首先,根据韦伯-费希纳定律确定了100/8.384~108/8.384形式的五级指数标度方法,对其合理性进行论述;其次,采用遗传算法对专家的判断矩阵一致性进行优化,优化后的判断矩阵一致性明显提高;随后采用粒子优化K均值聚类算法对专家意见进行聚类分析,并以主流专家意见为基础,采用对等共识模型计算专家权重,得到各评估指标最终的权重值;最后,以悬索桥评估模型为例,通过上述方法确定悬索桥模型中各指标的权重值,论证了方法的可行性和实用性.将本文的权重计算结果与规范权重进行对比,发现计算得到的大跨径悬索桥评估指标权重较规范权重更趋合理,尤其是索塔和附属设施的权重值.悬索桥评估指标权重的研究结论是对现行规范权重的补充,可以为规范的进一步修编提供参考.To determine the index weights of the condition assessment model for suspension bridges,group Analytic Hierarchy Process(AHP)approach was explored to determine the corresponding index weights.Firstly,the 10 0 8.384~10 8 8.384 scale method was adopted based on the Weber-Fechner law.Then,the genetic algorithm was used to optimize the consensus of judgment matrixes.The consensus of judgment matrixes was significantly improved after optimization.Moreover,the PSO-K-MEANS clustering algorithm was used to classify the experts'opinions,and the expert weights were calculated by the peer-to-peer consensus reaching model.In the end,the presented group-AHP method was applied to a suspension bridge assessment model to verify its feasibility and practicability.Compared with the index weights of the code,it was found that the index weights calculated in this paper were more reasonable than those of the code.The presented index weights of suspension bridges are the effective supplement to the current standard,which can provide reference to revise the current standard.

关 键 词：悬索桥 状态评估 指标权重 群组层次分析法 对等共识模型 

分 类 号：U446.3[建筑科学—桥梁与隧道工程]