Optimization of tension stiffening model of concrete based on results of layered nonlinear analysis of FRP – reinforced concrete beams

Document Type : Research Article

Authors

1 M.Sc Student, Department of Civil Engineering, University of Sistan and Baluchestan, Zahedan, Iran.

2 Associate Professor, Department of Civil Engineering, University of Sistan and Baluchestan, Zahedan, Iran.

3 Professor, Department of Civil Engineering, University of Sistan and Baluchestan, Zahedan, Iran.

Abstract

Tension stiffening effect, which is due to participation of concrete in resisting tension between cracks, results in the reduction of strains of reinforced concrete member relative to pure bars. In some researches, this effect was involved at strain – stress behavior of concrete after cracking that few researches investigated FRP reinforced concrete. In Present investigation, comparison of experimental data with results of layered nonlinear analysis of wide collection FRP-reinforced concrete beams relative to prior researches was used for extracting of concrete strain – stress after cracking. The resulting model unlike previous models was developed for all type of FRP bars. The constants of model are obtained from optimization method based on genetic. 78 curves from 87 curves of FRP – reinforced concrete beams under four-point bending are used for deriving of model coefficients and the others are used for investigating of optimal model. Furthermore, 20 percent of load – deflection data of 78 beams are also randomly selected for controlling of optimal model instead of developing of model. The average natural logarithm of experimental/calculated deflection ratios for randomly selected data by applying initial model and optimal model is 1.0457 and 0.2668, respectively. According to ideal value of zero for natural logarithm of experimental/calculated deflection ratio, applying optimal model has improved this statistical parameter by 74 percent. Responses of nonlinear analysis by using initial model and optimized model are compared with those of code relations.

Keywords

Main Subjects