Research On Nonlinear Intelligent Estimation Methods For Construction Cost Of Metro Interval Shield Engineering

The investment estimation in feasibility study stage determines the effect of subway construction and operation in the later stage,and has a very important impact on subway project.Investment estimation is an important technical and economic index for the economic evaluation of metro shield construction.The accuracy of investment estimation plays a decisive role in the investment control of the entire subway section.The investment estimation method proposed in this paper is based on the consideration of the construction period of the project and overcomes the simple lag of the traditional investment estimation method.The investment estimation of the subway interval shield is carried out by using the artificial intelligence method.In this paper,the subway section which is constructed by shield method is taken as the research object,considering from the angle of construction period,the shield engineering of subway section is decomposed by Project Breakdown Structure(PBS).This paper analyzes the engineering characteristics of the investment estimation in the construction period of shield tunneling,collects the relevant data and information of the subway section that has been built or is under construction,and analyzes and collates these information.Thus,a relatively perfect basic information database for investment estimation of shield tunneling in subway section is established.On the basis of this,the basic engineering characteristic system of shield section engineering of metro is established.At the same time,the characteristic index of these engineering is taken into account.There may be redundancy and correlation between the markers.The rough set theory is used to reduce the attribute of these engineering characteristic indexes.According to the magnitude of attribute importance,several indexes which have great influence on the cost of shield tunneling in subway section are obtained.For the later use of MATLAB software to forecast the investment in the construction period,the workload is reduced,and the goal of using relatively little information to predict the relatively accurate estimate is achieved.The reduced engineering characteristic attributes are sorted out and the quantitative data table is obtained.Then the construction period of the proposed subway interval shield is predicted by using BP neural network model with MATLAB software.Finally,the theoretical model and system built in this paper are applied to the Beijing Metro project,and the rationality and reliability of the model and system are verified.