Compressive Response and Damage Distribution of Fiber-Reinforced Concrete with Various Saturation Degrees.

Tunnels frequently experience issues such as lining spalling and water leakage, making the stability of tunnel support critical for engineering safety. Given that tunnels are subjected to various ground stress disturbances and groundwater influences, it is essential to investigate the mechanical properties and damage mechanisms of tunnel support materials under different loading paths and saturation levels. Fiber-reinforced concrete (FRC) is widely used for tunnel support; in this study, uniaxial compression tests were conducted on FRC with different fiber contents (0%, 0.5%, 1.0%) under varying loading paths (monotonic, pre-peak cyclic loading, full cyclic loading). The stress-strain behavior, volumetric strain, and elastic modulus were analyzed. The results indicate that increasing fiber content enhances strength and stiffness, while higher water content leads to a significant water-weakening effect, reducing both parameters. To classify crack types, the logistic regression (LR) algorithm is employed based on the AF-RA features, identifying tensile damage (which accounts for 60-80%) as more dominant than shear damage. Using this classification, AE event distributions reveal the spatial characteristics of internal damage in FRC. Gaussian process regression (GPR) is further applied to predict the AE parameters, enabling the assessment of the tensile and shear damage responses in FRC. The location and magnitude of the predicted wave crest indicate extreme damage levels, which become more pronounced under a higher saturation condition. A damage constitutive model is proposed to characterize the post-peak softening behavior of FRC. The numerical verification demonstrates good agreement with the experimental results, confirming the model's capability to describe the softening behavior of FRC under various fiber and water contents.