Shazad Qadir & Zhir Latif
Civil Engineering
TIU – Sulaymaniyah
In this project, we are conducting optimization of the flexural strength and stiffness designs for reinforced concrete beams. We are applying surrogate modeling technique to predict the responses of the structural system in three-point flexure tests. We have used Box-Behnken design method by arranging three design input parameters which are: area of steel bars in the compression zone; area of steel bars in the tension zone; and area of steel bars in the shear zones.
Thirteen specimens of reinforced concrete beams with dimensions of (1.02*0.2*0.15)m were prepared by using 10 mm diameter size of longitudinal steel bars for both the compression and tension zones, and using 6 mm diameter size steel bars for the shear reinforcement. The specimens were tested under three-points flexure tests in the age of 28 days and both the failure load and the maximum deflection in the middle of the specimens were recorded. Compression and tension tests have been conducted to obtain the concrete data for the analysis and numerical modeling. Finite element modeling has been performed for all specimens using ATENA program in order to verify the experimental tests through numerical results and simulation of the failure pattern. The surrogate models for the flexural strength and the stiffness of all specimens were constructed utilizing least square method, regression analysis, and MATLAB codes.
Furthermore, optimization process has been conducted to support the factorial method for the predicted responses of the structural system under three-point bending test. The results indicated that the flexural strength of reinforced concrete beams is associated with two main factors: the reinforcement in the flexure zone and the shear reinforcement together at the same time. It is worthy to mention that it was detected that the stiffness property of reinforced concrete beams is highly dependent on both longitudinal reinforcement in the compression zone first and in the tension zone as well. The availability of both reinforcements together has a great role in interpreting the stiffness of the structural member under external loading. Also, the role of shear reinforcement is appreciable in supporting the design of the reinforced concrete beams in flexure and in ductility as well.
Moreover, Box-Behnken design method manifested excellent strength in building the surrogate models to predict the responses of the structural system under loading and, as a result, efficient design, safer design, and lower cost design. The numerical models generated in ATENA program have manifested a great capability in verifying the experimental tests where the results were very close and very satisfactory, and the numerical models verified the experimental tests through displaying the failure types and the crack patterns to a very good agreement.
Also, the numerical models can be adopted to predict and optimize the design of the reinforced concrete beams for flexure and ductility. It was recognized that the sensitivity analysis has a great capability in the optimization process by producing the minimum and maximum responses based on the factorial method of the predicted responses of the flexural strength and the stiffness from the surrogate models.
Finally, the results manifested an excellent tool to optimize the design of reinforced concrete beams for flexure and stiffness.
