Development of analytical flexural models for steel fibre-reinforced concrete beams with and without steel bars

ENGLISH ABSTRACT: There is an increasing demand for the development and use of innovative materials with reduced cost of construction while offering improved structural properties. Steel fibre reinforced concrete (SFRC) can be used as a structural material to substitute the conventional reinforcing bars partially or fully. However, there is little or no codified approach on the design procedures for SFRC members in the latest guidelines outlined in the draft 2010 Model code. It is against this background that analytical methods are derived in this study for the determination of the flexural capacity of strain-softening, deflection-hardening SFRC with and without steel reinforcing bars. Models used for the determination of the flexural capacity of SFRC rectangular sections are based on equivalent stress blocks for both compression and tensile stresses. These are derived from an elastic-perfect plastic model for compression and either an elastic-constant post-peak response or Rilem’s multi-linear model for tension, in which strain compatibility and force equilibrium theories are used. By employing the equivalent stress blocks for both tensile and compressive stress states, parameters are defined by converting the actual stress-strain distribution to an equivalent stress block, depending on the ratio of yield (or cracking) strain and post-yield (post-cracking) strains. Due to the simplicity of a drop-down tensile model and a bilinear compression model, these material models are used for the subsequent derivation of the flexural models for both SFRC with and without steel reinforcing bars. An experimental program is designed and executed for model verification. This includes material characterisation experiments for the determination of material model input parameters, and main beam flexural experiments for the determination of the beam bending capacity. An indirect tensile test is used for the characterisation of the tensile behaviour while a four-point bending test is used for beam bending behaviour. Both flexural models for SFRC with and without reinforcing bars have been verified to fairly predict the flexural capacity of the beams. However, the flexural model for SFRC with steel bars offers some challenges as to whether the synergetic effect of using both steel bars and steel fibres should be incorporated at the low fibre volumes as used in the verification exercise. Furthermore, the use of indirect methods to characterise tensile behaviour added some uncertainties in the material model parameters and hence may have affected the predictability of the model. More research on the verification of the models is required to enable the use of a wider concrete strength spectrum for the verification and possible modification of the models. Studies on the model uncertainty may also help determine the reliable safety factor for the use of the model in predicting design strength of beam sections at a prescribed reliability index.