ABSTRACT Elevated temperatures ( T ) significantly affect the strength and fracture behavior of concrete. This study investigates the coupled effects of relative ligament depth (( W ‐ a 0 )/ d max ) and different elevated T . A calculation method for the fictitious crack growth length (Δ a fic ) in light‐weight concrete (LWC) and normal‐weight concrete (NWC) is proposed. A mesoscopic fracture model is established based on small‐scale experimental results, simultaneously determining the size‐independent tensile strength ( f t ) and fracture toughness ( K IC ) of LWC and NWC at different T . Using the proposed theoretical model, the material parameters ( K IC and f t ) for LWC and NWC are obtained through linear fitting and normal statistical analysis. The validity of the proposed mesoscopic fracture model is confirmed by comparing the computational results derived from both methods with experimental data. The results showed that for the f t values from the two methods, only one dataset exceeds the upper limit of test dispersion by 0.07 and 0.08 MPa, respectively, while all other data fall within the dispersion range. In addition, based on the material parameters obtained through the two methods, the complete fracture failure curves of LWC and NWC under different T are constructed. Finally, a simplified linear relationship is developed between the peak load ( P max ) and the material parameters. The P max of LWC and NWC at different elevated T are predicted, with all data falling within the 95% confidence interval ( μ ± 2 σ ), further validating the reliability of the proposed mesoscopic fracture model.
Li et al. (Tue,) studied this question.