Force- and Displacement-Controlled Non-linear FE Analyses of RC Beam with Partial Steel Bonded Length

Abstract

The current work deals with the comparison between two different computer formulations, namely, force- and displacement-controlled approaches for non-linear FE analyses of an RC beam in flexure with partially developed and asymmetric steel reinforcement. Instead of inputting load–slip data of direct tension pullout test to ANSYS code for defining the bond property, a different approach is adopted. In the study, load slip data are extracted from bond stress model of Yousaf (Performance of self-compacting concrete in bond at beams intersection. PhD dissertation, 2015) where the model was developed using data of digital strain gauge mounted on main steel bar in flexural beam testing. The load slip data are fed to computer as a set of real contestants choosing large-deformation non-linear plastic analysis scheme, discrete modeling approach for material modeling and program-chosen incremental scheme following the Newton–Raphson method. Complete harmony is found in computer and experimental results advocating the correctness and validation of the approach adopted for computer analysis. Among both formulations adopted for ANSYS analysis, the results reveal that the displacement-controlled approach is easy and efficient in terms of time-saving and less disk space requirement along with the ability to give falling branch of load–deflection response, if element displacement capacity still exists. Furthermore, it gives an early estimate of the load carrying capacity of the structural element along with suitable values of convergence and non-linear solution parameters. On the other hand, force-based analysis for a beam with unsymmetrical reinforcement detailing seems to yield more realistic and practical results in terms of mid-span deflection and beam cracking behavior compared with assumed symmetric displacement-controlled technique. From the study, the superiority of the force-controlled approach in comparable conditions is established through observed performance simulation with that of practical beam results along with the applicability of the approach adopted in the study. A comparison of ANSYS and experimental stress of partially developed rebar shows that the used bond model performs extremely well.