Progress on Cross-Scale Design and Machine Learning Prediction of Penetration Resistance of Hybrid Fiber Reinforced Concrete

Hybrid Fiber-Reinforced Concrete (HFRC) significantly enhances penetration resistance through multi-scale fiber hybridization and multi-stage energy dissipation mechanisms. Compared to single-type fiber-reinforced concrete, HFRC exhibits superior dynamic strength, energy absorption capacity, and crack resistance, establishing it as a key structural material in military protective engineering. This paper systematically reviews recent advances in cross-scale design and machine learning (ML) prediction of the penetration resistance of HFRC. The analysis begins by examining how different fiber combinations generate cross-scale synergistic effects that collectively improve the dynamic strength and anti-penetration capacity of HFRC. Subsequently, the mechanisms and multi-scale transmission pathways through which nanomaterials enhance the crater resistance and anti-spalling capacity of HFRC by strengthening the matrix and interfaces are examined. Furthermore, this review elucidates how multi-scale structural characteristics such as fiber distribution, orientation, and interfacial bonding synergistically govern the evolution of penetration-induced damage and the resulting failure patterns. Finally, the predictive efficacy of ML models for penetration resistance of HFRC is evaluated, along with potential integration pathways between ML and traditional numerical simulation.