Font Size: a A A

The Research Of Monitoring Data Noise Reduction And Deformation Prediction Of Passenger Dedicated Line Engineering Settlement

Posted on:2015-03-14Degree:MasterType:Thesis
Country:ChinaCandidate:W J PengFull Text:PDF
GTID:2392330488998973Subject:Road and Railway Engineering
Abstract/Summary:PDF Full Text Request
Passenger Dedicated Lines(PDL for short)is in a rapidly development period in our country,the difficulty in the PDL construction is changed from strength problem to deformation problem,it has never been solved completely,the characteristics of short construction period and wide territorial scope lead to the deformation problem more complex.It can only be solved by control the deformation of the underline engineering in a limited scope now.So It is so important to deal with the monitoring data effectively and determine the PDL engineering's deformation trend and find its final settlement.The paper is based on the settlement deformation monitoring of underline engineering of 4th part of ?-section of chengdu-chongqing PDL,the underground engineering monitoring programme is established firstly,then reduce the noise by wavelet analysis,then use the data to conduct deformation prediction by neural network method,at last,predict the settlement deformation by ABAQUS.By comparing the predict result with the data after noise reduction,and Discuss what kind of prediction method in PDL engineering settlement prediction is more reasonable and accurate,the main research results in the paper are listed as follows:1.Make the passenger dedicated line construction monitoring scheme based on The relevant specification and technical requirements,and arrange Settlement deformation monitoring network.2.Introduce the principle of noise reduction by wavelet analysis,then find the best wavelet analysis to the settlement deformation monitoring data of PDL underground engineering by MATLAB.3.Introducing the principle of 3 kinds of neural network,which are time series improved BP neural network,RBF neural network,Eleman neural network.And then introduce how to build the time series neural network which is suitable for the settlement monitoring data.4.ABAQUS is used to build 3d model for settlement prediction analysis of 3 structures:roadbed,transition section,bridge,by comparing the result with the monitoring data,then judge the finite element method's feasibility in the settlement predict program of PDL underline engineering.
Keywords/Search Tags:Passenger Dedicated Lines, settlement deformation, wavelet analysis time series-neural network, finite element method
PDF Full Text Request
Related items