Location

London

Event Website

http://www.csce2016.ca/

Description

Glass Fiber Reinforced Polymer (GFRP) plates present a viable option for strengthening corroded flanges and/or webs subjected to normal or shear stresses, to help restore their original local buckling strength. Given the large difference in elastic properties for the steel, adhesive, and GFRP, the use of the concept of transformed section is found to grossly overestimate the buckling strength of such system. An accurate prediction for the buckling strength of such systems necessitates the use of Three-dimensional Finite Element Analysis (3D FEA) modeling involving significant modelling and computational effort. The present paper is part of a larger study aimed at developing more computationally efficient solutions. Variational principles were formulated for the buckling analysis a single plate, two plates, and three-plates bonded through thin adhesive layer(s). Given (a) the complexity of the resulting expressions, and (b) the simplifying kinematics postulated in the formulation, it is desirable to devise a technique to assess the validity of the variational expressions obtained. Within this context, the present study develops a methodology to assess the validity of the variational expressions for the case of two-plate systems. A 3D FEA model was developed under Abaqus and the buckling stresses and associated mode shapes were determined. Through regression analysis, the observed mode shapes were successfully replicated by approximate functions. The approximate functions were then used in conjunction with the variational expression to predict the critical pressure combinations. The proximity of the critical pressure combinations predicted by the 3D FEA model and that based on the present solution suggest the validity of the assumptions made and the correctness of the variational principle.


Share

COinS
 
Jun 1st, 12:00 AM Jun 4th, 12:00 AM

STR-910: ASSESSMENT OF PLATE BUCKLING THEORY FOR TWIN-LAYERED PLATES USING THREE DIMENSIONAL FINITE ELEMENT ANALYSIS

London

Glass Fiber Reinforced Polymer (GFRP) plates present a viable option for strengthening corroded flanges and/or webs subjected to normal or shear stresses, to help restore their original local buckling strength. Given the large difference in elastic properties for the steel, adhesive, and GFRP, the use of the concept of transformed section is found to grossly overestimate the buckling strength of such system. An accurate prediction for the buckling strength of such systems necessitates the use of Three-dimensional Finite Element Analysis (3D FEA) modeling involving significant modelling and computational effort. The present paper is part of a larger study aimed at developing more computationally efficient solutions. Variational principles were formulated for the buckling analysis a single plate, two plates, and three-plates bonded through thin adhesive layer(s). Given (a) the complexity of the resulting expressions, and (b) the simplifying kinematics postulated in the formulation, it is desirable to devise a technique to assess the validity of the variational expressions obtained. Within this context, the present study develops a methodology to assess the validity of the variational expressions for the case of two-plate systems. A 3D FEA model was developed under Abaqus and the buckling stresses and associated mode shapes were determined. Through regression analysis, the observed mode shapes were successfully replicated by approximate functions. The approximate functions were then used in conjunction with the variational expression to predict the critical pressure combinations. The proximity of the critical pressure combinations predicted by the 3D FEA model and that based on the present solution suggest the validity of the assumptions made and the correctness of the variational principle.

http://ir.lib.uwo.ca/csce2016/London/Structural/63