We propose a multiscale surrogate framework for steel-fiber–reinforced concrete (SFRC) that links single-fiber mechanics to structural flexure. At the mesoscale, we model single-fiber pull-out with explicit fiber, matrix, and interface, combining Mode-II bond degradation with rate-dependent friction and (for hooked ends) local bending/straightening. From the obtained load–slip curves we extract orientation-resolved surrogate bridging laws—compact relations that retain the effects of friction, loading rate, fiber geometry (smooth; hooked with one, two, or three folds), and inclination. These laws form a reusable database covering arbitrary orientations. At the meso–macroscale, the structural response is assembled by weighting the orientation-resolved surrogates with the measured fiber orientation histogram and expected fiber counts across the ligament. At the macroscale, we analyze notched three-point bending beam (3PB) with a central cohesive crack, where matrix traction–separation carries the fracture energy and the fiber contribution enters through the surrogate bridging traction. The framework is validated against single-fiber pull-out and 3PB tests for multiple hooked-end fibers (3D/4D/5D) and inclinations, reproducing peak load and post-peak softening trends. By calibrating physics at the fiber scale and propagating it through orientation-aware surrogates, the approach achieves predictive accuracy at substantially lower computational cost than explicit fiber simulations, while remaining extensible to other fiber types and rates. • A multiscale framework links single- ber pull-out behavior to structural response of SFRC. • Rate-aware, orientation-resolved ber bridging laws are calibrated from pull- out tests. • A cohesive-crack formulation captures exural behavior in notched 3PB specimens. • Mesh robustness is demonstrated without crack-band regularization. • Fiber geometry, content, and orientation govern peak load and post-peak ductility.
Poveda et al. (Thu,) studied this question.