Research on Temporal Patterns of Water–Rock Interaction in the Coal Mine Underground Reservoir Based on the Dynamic Simulation Test

[Image: see text] The temporal pattern of water–rock interaction is significant in predicting the ion concentration of the effluent in coal mine underground reservoirs. This study used the roof-caved rock samples and main incoming water (mine water and fissure water) of the Daliuta coal underground reservoir as the research object and designed four groups of dynamic simulation experiments of the water–rock interaction. Based on the main ion concentrations in the water sample at different reaction times, Q-type hierarchical cluster analysis (HCA) was used to classify the stages of water–rock interaction. The types and intensities of water–rock interaction in each stage were identified by combining the ion ratio and principal component analysis (PCA). Q-type HCA shows that the dynamic simulation experimental water samples can be divided into three categories according to the reaction time, representing the early, middle, and late stages of the water–rock interaction process. The influence of water quality on the division of the water–rock interaction stage is greater than that of rock characteristics. The ion ratio and PCA show that the dissolution of pyrite minerals, cation exchange reaction, and mineral adsorption mainly occur in the early stage of water–rock interaction, in which the cation exchange reaction plays a leading role in the change of ions in water. In the middle stage, the cation exchange reaction and the dissolution of carbonate minerals, such as calcite and dolomite, mainly occur, in which mineral dissolution is the main. In the late stage, the water–rock interaction is relatively weak, and the change of ion concentration in water is not obvious. This study proves the temporal patterns of water–rock interaction between caved rock and mine water (or fissure water) and differences in the types and intensities of water–rock interaction in each stage. The results can provide a theoretical basis for the optimization of the operation cycle of coal mine underground reservoirs and the prediction of effluent ion concentration.