In the process of deep coal mining, the cyclic mining disturbance behavior produces significant cyclic loading and unloading effects, to ensure safe production in the mining area, it is necessary to study the damage characteristics and acoustic emission characteristics of the rock body under different cyclic loading and unloading effects. This study developed a TAWZ-5000/3000 rock true triaxial (disturbance) test system and used it to carry out true triaxial cyclic loading and unloading tests in different modes, with acoustic emission (AE) monitoring of the deformation and damage processes. The test results proved that a large number of cycles weakens the rock’s ability to control deformation, and this weakening ability is also related to the types of cyclic loading and unloading modes. The hysteresis loop characteristics and mechanical parameters show obvious path dependence. Based on the damage equivalence method, it is found that the absolute damage parameter increases with the increase of the number of cycles in the same mode, and the accumulation rate of the cumulative damage parameter is faster; the cumulative damage parameter in the equal amplitude cyclic loading and unloading (EAC) mode is always higher than that in the other two modes. The normalized AE ring count rate in the rock samples under cyclic loading and unloading has a significant relationship with the damage percentage, and the crack extension scale function (b-value) shows the “steady increase – strong fluctuation − significant decline” trend. By analyzing the AE characteristic parameters, namely rise time/amplitude (RA) and ringing count/duration (AF), it is found that the sandstones in EAC, graded cyclic loading and unloading (GC), and stepped cyclic loading and unloading (SC) modes are mainly damaged by tension-shear and shear, respectively. By analyzing the values of the acoustic emission parameters RA-AF, it is found that the sandstone is mainly damaged by tensile shear, tensile tension, and shear under EAC, GC, and SC modes, respectively, corresponding to the rock samples’ damage patterns. The study results have important guiding significance for preventing and controlling disasters in deep mines.