Los puntos clave no están disponibles para este artículo en este momento.
Abstract This study presents a comprehensive prediction model for the shear strength of slender reinforced concrete (RC) beams, applicable to both beams with and without shear reinforcement. The model is grounded in a unified mechanical framework that combines the contributions of distinct shear transfer mechanisms at the ultimate limit state—namely, uncracked concrete compression, aggregate interlock, dowel action, residual tensile stress, and, where applicable, shear reinforcement through a variable‐angle truss mechanism. For beams without stirrups, a novel size effect factor—based on width‐to‐depth ( b / d ), effective span‐to‐depth ( l eff / d ), and aggregate size‐to‐depth ( a g / d ) ratios—was introduced and optimized via genetic algorithms. The model was trained using a large experimental database and validated on independent data, ensuring strong generalizability. Validation showed high predictive accuracy: for beams without stirrups, R 2 = 0.9205 and RMSE = 34.79 kN; for beams with stirrups, R 2 = 0.8343 and RMSE = 46.46 kN. Compared to design codes (ACI 318‐19, EN 1992‐1‐1:2023, CSA A23.3‐19) and paper‐based equations, the proposed model demonstrated superior performance, with V exp / V pred means of 1.05 (no stirrups) and 1.04 (with stirrups), and lower standard deviations. Mechanism‐wise, the model quantified average contributions: in beams without shear reinforcement, concrete compression (49.0%) and aggregate interlock (44.5%) were dominant; in beams with shear reinforcement, stirrups (48.7%) and dowel action (25.8%) provided the main resistance, with lesser contributions from concrete compression (23.0%) and negligible residual effects. These insights, visualized through stacked bar charts, reveal the shifting role of mechanisms depending on reinforcement presence. Overall, the model provides a reliable, mechanism‐based framework for RC shear prediction and offers potential extensions to dynamic loads, new materials, and more complex geometries.
Aure et al. (Sun,) studied this question.