From Zero-Initialization to Physics-Based Determination: A Paradigm for Modeling Rock Damage under Cyclic Hardening

This study addresses a fundamental flaw in how damage mechanics is applied to rocks: the incorrect assumption that rock starts with zero damage, ignoring its natural porosity. This leads to inaccurate models, especially when simulating repeated stress cycles. We used CT scanning to observe how microscopic pores and cracks in rock evolve under cyclic loading. Results show that under low stress, pores close, causing hardening, while under high stress, cracks grow, causing softening. We found that when porosity minimizes under repeated low stress, hardening stops, indicating a nearly damage-free state. Based on this, we propose a practical method: pre-conditioning rock samples with high-cycle loading at 50% of their strength to establish this damage-free baseline. This allows for accurate calculation of the initial damage, providing a more realistic and widely applicable framework for predicting rock behavior in engineering. This dataset includes the experimental data and CT slices related to the above research content.