基于遗传算法的BP神经网络隧道施工参数正反演分析与应用  

Application of Genetic Algorithm Based ANN to Forward and Back Analysis on Tunnel Construction Parameters

作　　者：王睿[1,2] 漆泰岳[1,2] 冯剑 雷波[1,2] 李延[1,2] 

机构地区：[1]西南交通大学交通隧道工程教育部重点实验室,四川成都610031 [2]西南交通大学土木工程学院,四川成都610031 [3]中国市政工程西南设计研究总院有限公司,四川成都610031

出　　处：《铁道学报》2016年第4期98-104,共7页Journal of the China Railway Society

基　　金：国家自然科学基金(51278423;51478395)

摘　　要：相对于监测数据采集的高效性,隧道施工中对现场突发状况缺乏高效的应对措施。本文结合工程实例,采用正交表及对应三维数值计算模型,得到隧道施工参数与对应隧道变形的样本集,应用基于遗传算法的BP神经网络Matlab程序,通过对施工参数进行正演分析,实现相对高效的施工反馈;在实测数据基础上,通过进一步的反演分析,可优化施工参数,实现施工工艺的经济优选。工程应用结果表明,该方法的分析结果能够满足工程施工精度要求,有效提高施工过程中突发状况的应对效率,同时也为设计中参数的优化选择提供参考,为建立隧道工程施工的高效反馈机制提供新方法和新思路。Compared with the high efficiency of monitoring data collection during tunnel construction,effective counter measures are not available to deal with emergencies on the construction site.In this paper,combined with practical work,a sample set of the tunnel construction parameters and the corresponding tunnel deformation was established through an orthogonal layout and corresponding numerical 3Dcalculation model.The Matlab program of a genetic algorithm based BP neural network was applied to forward analysis of the tunnel construction parameters in order to achieve relatively efficient construction feedback.Based on the measured data,the construction parameters can be optimized through further back analysis to realize economic optimization of construction technology.The results showed that this method satisfied the precision requirements of construction,effectively improving the efficiency to deal with emergencies in the construction process.Furthermore,it can be applied to the optimization of parameters during design,offering a new way in establishing high-efficiency tunnel construction feedback mechanism.

关 键 词：遗传算法 BP神经网络 正演分析 反演分析 

分 类 号：U456.3[建筑科学—桥梁与隧道工程]