Finite Element Model Updating Using GA and Static Data for Studying Joints Behavior in Double Layer Grid

Double layer grids are one of the most popular types of precast space structures that are assembled and erected in construction site using bolted joint systems. If joints behavior was not considered in analytical model of these structures, there would be considerable discrepancy between analytical and experimental results. In order to predict more accurate analytical responses, joint behavior must be considered in the analysis. In the present work, with the aim of studying the behavior of the ball joint system that was utilized in construction of a double layer grid, the method of inverse problem and in particular indirect finite element model updating technique has been used. An appropriate finite element model capable of modeling joints behavior was build for a double layer grid that the results of its static displacement measurements were available. An optimization problem is defined in which the objective function contains difference between experimental and analytical responses of the double layer grid, and using the genetic algorithm this objective function was minimized. Consequently an approximation of the joint behavior that has been considered as optimization variable was obtained. The obtained results show a very good correlation between responses of the updated model and experimental responses. Also, obtained relationship for the behavior of ball joint system shows a nonlinear behavior with two specific zones, including initial lack of fit zone and then ideal behavior zone

کلیدواژه ها:

Model updating ، Genetic algorithm ، Double layer grid ، Ball joint system