Computational Modeling Of Realistic Experiment Boundary Conditions And Restraints

Liping Kang,Roberto T. Leon,Xilin Lu

2012-01-01

Abstract:Experimental simulation of an idealized structure in the laboratory typically requires a careful consideration to achieve the idealized boundary conditions and restraints; however, the real test conditions involve friction, rotational restraints, fabrication and construction errors that may have a significant effect on the structural response. Computational analyses with idealized boundary conditions and restraints cannot capture these realistic effects thus resulting in large errors in displacement, rotation, and capacity. This paper presents a computational modeling approach of realistic boundary conditions and restraints to simulate these effects in OpenSEES. Translational and/or rotational springs are used in the computational modeling to represent the friction and rotational restraints, and a methodology is presented to estimate the force vs. displacement and moment vs. rotation spring relations based on the global structural response. Additionally an approach to consider initial deformations due to assembly errors is also presented to account for locked-in stresses. The result shows that: 1). Friction and rotational springs increase the capacity of the analyzed connections; 2). Global unsymmetrical force vs. displacement behavior of the analyzed connection is obtained with unsymmetrical spring behavior or locked-in stresses from initial deformation; and 3). The proposed computational modeling approach captures the possible significance of each influencing factor on the global structural response. The results highlight the need to provide additional experimental data such as estimates of the frictional response and measured strain from rotational restraints and initial deformations during structural assembly so that the effect of these terms can be accounted for. With the additional experimental data and the significance of influencing factors, we can accurately predict the realistic structural response computationally.