Font Size: a A A

Research On The Key Technology In Construction Control Of Long Continuous Girder Bridge

Posted on:2014-06-24Degree:MasterType:Thesis
Country:ChinaCandidate:D J LiFull Text:PDF
GTID:2252330401985542Subject:Bridge and tunnel project
Abstract/Summary:PDF Full Text Request
Until now, the continuous girder bridge construction control theory development has become increasingly mature, but with the deepening of the engineering requirements, design theory, construction method and the increasing maturity of the development and application of high strength material, the continuous girder bridge has towards long span and long unit development trend. But there is construction process complex, large cumulative errors and closure of multiple control problems we need to solve and study. Yellow River bridge as the engineering background in Inner Mongolia (59.7m+11×100m+59.7m), this paper based on the characteristics of long continuous girder bridge, based on the construction control and the finite element model analysis, use the least squares support vector machines (LS-SVM) elevation error prediction analysis of cantilever stage, and research the closure of long continuous girder bridge, concrete research content and results are as follows:First of all, in combination with the practical situation of Inner Mongolia Yellow River Bridge, construction control content is determined; the influence factors of construction control error are analyzed, and determine the factors influencing the cantilever construction elevation error.Secondly, selected the field measured data as the research sample, elevation error prediction model is established according to steps such as processing samples, choosing a kernel function, selecting parameters, training and predicting. Using the established model to predict the elevation error of the cantilever construction stage, find the elevation error prediction model based on the LS-SVM prediction accuracy is high, and have strong learning ability, stability and efficiency, through the LS-SVM prediction compared to the measured values, the SVM model prediction and the BP neural network model prediction and the LS SVM dimensionality reduction model prediction. Through the elevation control validation of cantilever stages verification, the error prediction based on the LS-SVM of this bridge elevation control is feasible, and successfully solved due to the segment less predictive control problem in cantilever construction stage.The last, this chapter on the engineering example put forward four feasible closure scheme, including three for optimization solution, with the deformation and internal force of main girder as the elements, the joining of different order scheme under each working condition of the final closure, the cumulative edge of vertical displacement, the section on stress and displacement of bearing for comparative analysis. Research found that different closure order scheme, structural system conversion time, displacement and stress difference are very big, it is necessary to optimize research, comprehensive comparison choose a reasonable solution. And the closure phase of construction control was studied, through into the bridge construction control results of verification, validation showed that the closure order method of optimization technique and the construction control technology, well solved the key problem in joining long continuous girder bridge construction control, and can provide reference and basis for the same type bridge construction control.
Keywords/Search Tags:continuous girder bridge, construction control, LS-SVM, prediction error, closure research
PDF Full Text Request
Related items