Failure mode classification and deformability evaluation for concrete beams reinforced with FRP bars

The failure mode of fiber-reinforced polymers reinforced concrete (FRP-RC) beams is a concern in the capacity evaluation. Therefore, developing a robust method to identify the failure mode of FRP-RC beams is warranted. This paper proposes a support vector machine (SVM) algorithm, together with comprehensive compiled experimental databases and a validated analytical model, to identify the failure mode of FRP-RC beams. The SVM analysis yields data-driven classification rules that not only achieved superior performance over ACI 440 guidelines but are also valid for a large range of design parameters. The proposed classification rules are a function of the balanced reinforcement ratio (pfb) and height to depth ratio (h/b) of FRP-RC beams. In addition, the verified classification rules recommend increasing the transition zone defined by ACI 440 from 1.4pfb to 1.55pfb. The study also utilized the verified analytical model to conduct a comprehensive deformability evaluation of FRP-RC members. The results indicate that FRP-RC beams have sufficient deformability before failure regardless of the failure mode. Moreover, to ensure minimum deformability requirements for FRP-RC beams, a limit should be specified on the modulus of elasticity -to-the ultimate tensile strength of the FRP ratio. Based on the analysis, a proposed limit of 200 is recommended.

Automated summary

This paper proposes a method using support vector machine algorithm to identify the failure mode of FRP-RC beams, and obtains classification rules that outperform ACI 440 guidelines. The study shows that FRP-RC beams have sufficient deformability before failure, and suggests limiting the modulus of elasticity-to-the ultimate tensile strength of the FRP to ensure minimum deformability requirements.