Font Size: a A A

Reliability Assessment Of High-Pier Bridge-Steep Slope Interaction System Based On GA-SVM

Posted on:2023-03-18Degree:MasterType:Thesis
Country:ChinaCandidate:H H LiuFull Text:PDF
GTID:2532306914953379Subject:Civil engineering
Abstract/Summary:PDF Full Text Request
China’s transportation infrastructure construction is developing to the southwest,and a large number of Bridges are distributed on high and steep slopes or even cliffs with extremely complicated engineering conditions.In the process of bridge operation,the non-uniform constraint effect of steep slope on bridge pile foundation is significant,and the bridge substructure and foundation are prone to failure under vehicle load.In addition,under the influence of vehicle load,external environment and steep slope,some components of the bridge will fail to varying degrees,such as excessive deflection,cracking of box girder,and excessive displacement of pile foundation.Therefore,the study of bridge-steep slope interaction mechanism is of great scientific value and engineering significance for reliability and safety assessment of bridge structure.Aiming at the above problems,this paper studies the interaction mechanism of bridge-steep slope through finite element simulation.Considering the influence of steep slope on the substructure,combined with support vector machine algorithm and genetic algorithm,the reliability of high-pier bridge-steep slope interactive system was studied.The main work is as follows:(1)Combined with the pile foundation case of a steep slope bridge engineering,the pile-soil interaction model is established by using ABAQUS finite element software,and the validity of the finite element simulation results of pile foundation deformation under steep slope is verified.The effects of load,pile and site parameters on horizontal displacement and bending moment of pile top are studied,and the interaction mechanism of pile and soil is revealed.According to the reliability theory,support vector machine and genetic algorithm,the process of calculating reliability by GA-SVM method is compiled.Based on the proposed method,reliability analysis of four examples was carried out,and the first three examples were explicit functional functions.The results of the proposed method were compared with those of the design checking point method,and the effectiveness of the proposed method in structural reliability analysis was verified.The fourth example is the implicit function of the pile foundation finite element model of the above steep slope bridge,and the failure probability and reliability index of the pile top displacement failure mode are calculated.(2)Based on the engineering background of a continuous rigid frame bridge,the finite element analysis models of bridge superstructure and steep slope pile foundation system were established respectively,and the reaction force at the consolidation of the pier bottom of the superstructure model was taken as the external force imposed on the bridge cap.The failure modes and limit state equations of bridge critical sections are determined by referring to relevant literatures.Latin hypercube sampling method was used to extract samples(input values),and the corresponding response values(output values)were obtained through finite element calculation.Training and test sets for SVM model were obtained.Based on the reliability algorithm of GA-SVM,the reliability index and failure probability of the bridge system are calculated.(3)Based on the reliability analysis of structural system,the time-varying reliability of steep slope bridge system was analyzed by time discrete method considering the degradation of bridge material properties and the attenuation of shear strength with time,and the reliability indexes of different service time in the bridge design reference period were obtained.Combined with the results of structural time-varying reliability analysis,the service safety of the bridge is evaluated.
Keywords/Search Tags:bridge engineering, steep bridge, support vector machine, genetic algorithm, reliability
PDF Full Text Request
Related items