Font Size: a A A

Reliability Analysis Of Wide-span Cable-stayed Bridge Based On Support Vector Machine

Posted on:2020-04-27Degree:MasterType:Thesis
Country:ChinaCandidate:H C GuoFull Text:PDF
GTID:2392330590964210Subject:Bridge and tunnel project
Abstract/Summary:PDF Full Text Request
Broad width and long-span cable-stayed bridges have been applied more and more in engineering because of their beautiful shape,good driving vision and good mechanical behavior.With the rapid development of long-span cable-stayed bridges,many problems have arisen.In recent years,the structural failure of long-span bridges has caused great harm to the development of society,economy and transportation.With the increase of bridge span,the bridge structure will become more complex,and the failure of bridge structure should be paid attention to.Therefore,the reliability analysis of broad width and long-span cable-stayed bridges becomes particularly important.Owing to the complexity and inefficiency of traditional structural reliability analysis methods,it is complicated and inefficient to calculate and analyze the reliability of broad width and long-span cable-stayed bridges.In this paper,support vector machine and intelligent algorithm are combined to realize the reliability analysis of cable-stayed bridges.The main contents are as follows:1.According to the existing reliability analysis methods and theories,the advantages and disadvantages of the first-order second-moment method,the second-order second-moment method,the MC method and the response surface method are analyzed.The support vector machine can accurately fit the function of cable-stayed bridge system,and this method can be used to analyze the reliability of Beijie waterway bridge.2.Based on MATLAB platform,support vector machine algorithm is used to fit the structural function of cable-stayed bridge.In order to ensure its accuracy,cross-validation method and genetic algorithm are used to optimize the parameters of support vector machine model.Then penalty function method is used to transform the non-linear constrained optimization problem into a simple function equation extremum problem.Finally,particle swarm optimization algorithm is used to solve the extremum problem.The solution is the solution of reliability index.The feasibility of the method is verified by classical examples,and the results are of high accuracy,which can be used for reliability analysis of Engineering structures.3.The finite element model of Beijie Waterway Bridge is established by using ANSYS software,and it is analyzed by combining support vector machine and intelligent algorithm.Aiming at the problems of stiffness failure and strength failure which often occur in engineering,the reliability of the main girder midspan deflection,cable stress and tower stress of Beijie Waterway Bridge is analyzed by using different vehicle load distribution methods,and the reliability is higher.Sensitivity analysis of the mean and standard deviation of each random variable is carried out to obtain the random variable which has the greatest impact on various failure modes.4.The structural system reliability of Beijie Waterway Bridge is estimated based on the general limit range method.The results show that the reliability of Beijie Waterway Bridge meets the requirements of safety range.Comparing the structural system reliability of Beijie waterway bridge,the mid-span deflection reliability of main girder,the stress reliability of cable-stayed cable and the stress reliability of main tower,the stress reliability of cable-stayed cable has the greatest impact on the structural system reliability of Beijie waterway bridge.It is necessary to pay attention to the selection,installation and later maintenance of cable.
Keywords/Search Tags:Support Vector Machine, Reliability Analysis, Wide Width, Cable-stayed Bridge, MATLAB
PDF Full Text Request
Related items