| The shield method is widely used in urban rail transit construction in China.During the shield construction process,there are many participants and the construction period is also long.As the diameter and depth of the tunnel continue to increase,the stratum and construction environmental conditions will become more complicated.More and more influencing factors are facing.Therefore,it is very meaningful and necessary to adopt better methods and technical means for construction management and control.Based on the advantages of intelligent algorithms and BIM technology in non-linear problem processing,information integration and collaboration respectively,the paper researches the construction parameters,personnel,materials and machinery management and control method for shield construction.The main research contents include: improvement on optimization method of shield machine tunneling parameters,matching method for shield machine tunneling parameters with limited engineering data sample size,shield construction personnel,materials and machinery management and control plan information model based on BIM technology,integration method between shield construction personnel,materials and machinery information and progress control plan network,management and control system about personnel,materials and machinery for shield construction incorporating intelligent algorithms and BIM technology.Based on the above research content,the research works and results are as follows:(1)The improvement and verification for PSO algorithm performance and application in shield machine tunneling parameters optimizationAccording to the curve characteristics of the exponential decreasing function and the range of the inertia weight in PSO algorithm,a new inertia weight dynamic adjustment equation is constructed.In order to further reduce the possibility of falling into the local optimal value at the later iteration stage,a specific adjustment strategy for the particle swarm during the iteration process is given,proposing exponential adjustment inertia weight immune particle swarm optimization(EAIW-IPSO).12 sets of standard test functions are applied to test the EAIW-IPSO performance.Meanwhile,the improved PSO algorithms from other scholars are compared.The minimum,average value,root mean square error and iterative curve are used to analyze the accuracy,stability and iteration characteristics.The experimental results show that the EAIW-IPSO algorithm has higher accuracy,stability and better iterative characteristics in both single-peak and multi-peak problems.At the same time,EAIW-IPSO algorithm has a stronger ability to avoid falling into local extremes.Based on the EAIW-IPSO algorithm,an optimization model of shield machine tunneling parameters is constructed.The geometric factors and soil factors of the tunnel are considered in the model.BP neural network optimized by GA algorithm is used to first establish the nonlinear relationship model between geometric parameters,soil parameters,shield tunneling parameters and ground settlement.Non-linear model training is carried out through these parameters and the corresponding ground settlement values obtained in the actual project.Based on the well-trained model,the EAIW-IPSO algorithm and PSO algorithms improved by other scholars are used to optimize the shield machine tunneling parameters,obtaining the optimal shield machine tunneling parameter value under specific geometric parameters and soil parameters.According to actual case application results,the ground settlement value obtained by the EAIW-IPSO algorithm is the smallest.Therefore,the performance of the EAIW-IPSO algorithm is verified again,providing a more accurate and reliable method for the optimization of the shield machine tunneling parameters.(2)Construction and verification of the shield machine tunneling parameter matching model with limited sample dataSVM is used to establish nonlinear relationship model between the parameter variables and the target value in a class of problems about shield machine tunneling parameter matching.Then,the EAIW-IPSO algorithm is used to optimize the parameter variables based on the nonlinear relationship model.Through two sets of examples in this type of problem,experiments are performed.By sequentially reducing the number of training samples,the performance changes of the model under different sample size conditions are analyzed.In the experiments,the matching models based on BP and PSO algorithms are also compared.The results show that the matching model based on SVM and EAIW-IPSO can still obtain more accurate and stable matching results under the condition that the number of samples is decreasing.The tunnel geometric parameter is the depth-span ratio.Soil parameters are water level,gravity,cohesion,internal friction angle and pressure measurement coefficient.The shield machine tunneling parameters are the synchronous grouting amount,thrust,cutter head torque,soil tank pressure and penetration.These parameters are all as the input parameter variables of SVM model.The ground settlement is as the objective value.Considering the limitation of obtaining samples,SVM model training and testing are carried out through actual shield construction data.A suggestion is given that the number of training samples required for shield tunneling parameter matching based on SVM and EAIW-IPSO is 28 to 49 groups under the condition that the average relative error of SVM test is within 3%～10%.Based on developed parameter matching system module,30 sets of data in actual shield construction are used to train SVM model.EAIW-IPSO is used to optimize the shield tunneling parameters under specific geometric parameters and soil parameters.By comparing the matching models of BP and PSO,the application results in the example show that the matching model based on SVM and EAIW-IPSO is superior in sample fitting,model generalization ability,and optimization results.(3)Information model construction for personnel,materials and machinery management and control plan in shield projectAiming at the defects of creating shield segment information model in Revit,a plug-in for rapid generation of segment information model was realized through secondary development technology.Based on the Revit family,the related materials and mechanical equipment information model in shield construction were created.And other information models in starting venue were created based on the Revit project.The information models were imported into Navisworks,and the corresponding element management sets were established.The personnel,materials and machinery construction plan information were further integrated with the basic geometric information model of the component to realize the information model establishment.Therefore,based on the information model of the construction management and control plan,the personnel,materials and machinery information of the shield construction process can be better visualized,which provides more intuitive and effective guidance for construction management.(4)The integration method between shield construction personnel,materials and machinery information and progress control plan networkRevit was used to create component family of work in the schedule control plan network,and performe parameter setting.Through Revit API and Window Forms technology,a plug-in that can operate the schedule network personnel,materials and machinery information has been developed.The plug-in can achieve the following functions: adding and querying personnel,materials and machinery information;quick calculation and assignment modification for working time parameter values;visualization of building component models in each work.According to pre and post work and duration,schedule network that integrates multi-dimensional information was formed in Revit.The developed plug-in then was used to quickly calculate the time parameter values in the network,add the management and control plan information and visualize the building components corresponding to the specific work.So as to realize the integration of shield construction building components,personnel,materials and machinery information and schedule control plan network.The above operations enable relevant personnel to more intuitively and clearly understand the specific construction objects under each work and the key information about personnel,materials and machinery required during the construction process.(5)Design and development of management and control system about personnel,materials and machinery for shield construction incorporating intelligent algorithms and BIM technologyBased on the management and control content of personnel,materials and machinery in shield construction process,the system functional module design is performed by combining with the value of intelligent algorithms and BIM technology.Microsoft SQL server 2012 was used to establish the system database,and the corresponding database forms were designed.In the Microsoft Visual Studio 2012,the front-end design of the system,the back-end control code writing,and socket-based communication between the client and server are implemented.The each module function was tested by using personnel,materials and machinery information in actual shied construction project.The test results verify the feasibility of the system.The intelligent algorithm and BIM technology are integrated into the system,which improves the ability of the system to solve the nonlinear problems in the management and control of shield construction personnel,materials and machines,and realizes the visualization of personnel,materials and machines information in the system. |
PDF Full Text Request