To explore the hydration characteristics and early strength evolution of classified fine tailings cemented backfill (CFTCB), a nuclear magnetic resonance (NMR) analysis and a volume resistivity test were performed on classified fine tailings filling slurry (CFTFS). The early hydration products of CFTCB were studied by scanning electron microscopy (SEM) and X-ray diffraction (XRD) phase analysis. Uniaxial compressive strength (UCS) test was carried out, and the microscopic characteristics and strength rules of the hydration reaction of CFTCB were analyzed. Based on the experiment, we found the law of water content change and porosity evolution. The early hydration reaction can be divided into the dissolution, setting, and hardening stages. The volume resistivity test results show that the volume resistance of filling slurry increases slowly at first then decreases, and finally increases rapidly. The variation trend of volume resistivity is related to the degree of hydration reaction. When combined with the hydration characteristics of backfill materials, the hydration reaction rate determines the growth rate of early strength of backfill, and the formation of hydration products is the reason for the early strength increase in backfill. The research conclusion has an important theoretical guiding value and engineering significance in mine filling production organization and filling ratio parameter optimization.

A reasonable arrangement of filling pipelines can solve the problems of low line magnification, a high flow rate, large pipe pressure, etc., in deep well filling slurry transportation. The transportation pressure loss value of filling slurry is the main parameter for the layout design of filling pipelines. At present, pressure loss data are mainly obtained through the loop pipe experiment, which has problems such as a large amount of labor, high cost, low efficiency, and a limited amount of experimental data. In this paper, combined with a new generation of artificial intelligence technology, the random forest machine learning algorithm is used to analyze and model the experimental data of a loop pipe to predict the pressure loss of slurry transportation. The degree of precision reaches 0.9747, which meets the design accuracy requirements, and it can replace the loop pipe experiment to assist with the filling design.

Cemented tailings backfill (CTB) plays an important role in mine filling operations. In order to study the long-term stability of CTB under the dynamic disturbance of deep wells, ultrafine cemented tailings backfill was taken as the research object, and the true triaxial hydraulic fracturing antireflection-wetting dynamic experimental system of coal and rock was used to carry out a static true triaxial compression test, a true triaxial compression test under unidirectional disturbance, and a true triaxial compression test under bidirectional disturbance. At the same time, the acoustic emission monitoring and positioning tests of the CTB were carried out during the compression test. The evolution law of the mechanical parameters and deformation and failure characteristics of CTB under different confining pressures is analyzed, and the damage constitutive model of the filling body is established using stochastic statistical theory. The results show that the compressive strength of CTB increases with an increase in intermediate principal stress. According to the change process of the acoustic emission ringing count over time, the triaxial compression test can be divided into four stages: the initial active stage, initial calm stage, pre-peak active stage, and post-peak calm stage. When the intermediate principal stress is small, the specimen is dominated by shear failure. With an increase in the intermediate principal stress, the specimen changes from brittle failure to plastic failure. The deformation and failure strength of CTB are closely related to its loading and unloading methods. Under a certain stress intensity, compared with unidirectional unloading, bidirectional unloading produces a greater deformation of the rock mass, and the failure strength of the rock mass is higher. This study only considers the confining pressure within the compressive limit of the specimen. Future research can be directed at a wider range of stresses to improve the applicability and reliability of the research results.