Groundwater inrush is a typical hydrologic natural hazard in mining engineering. Since 2000 to 2012, there have been 1110 types of mine groundwater inrush hazards with 4444 miners died or missing. As a general geological structure in the northern China coalfields, the karst collapse pillar (KCP) contains a significant amount of granular rocks, which can be easily migrated under high hydraulic pressure. Therefore, the KCP zone acts as an important groundwater inrush pathway in underground mining. Grouting the KCP zone can mitigate the risk of groundwater inrush hazard. However, the fracture or instability of the coal pillar near KCP can cause the instability of surrounding rock and even groundwater inrush hazard. To evaluate the risk of groundwater inrush from the aquifer that is caused by coal pillars instability within grouted KCP in a gob, an in-situ investigation on the deformation of the surrounding strata was conducted. Besides, a mechanical model for the continuous effect on the coal pillar with the floor-pillar-roof system was established; then, a numerical model was built to evaluate the continuous instability and groundwater inrush risk. The collective energy and stiffness in the floor-pillar-roof system are the two criterions for judging the stability of the system. As a basic factor to keep the stability of floor-pillar-roof system, the collective energy in coal pillar is larger than that in floor-roof system. Moreover, if the stiffness of floor-roof or coal pillar meets a negative value, the system will lose stability; thus, the groundwater inrush pathway will be produced. However, if there is a negative value occurring in floor-pillar-roof system meets, it indicates that the system structure is situated in a damage state; a narrower coal pillar will enlarge the risk of continuous instability in the system, leading to a groundwater inrush pathway easily. Continuous coal pillars show a lower probability of instability. Conversely, the fractured coal pillars have a greater probability of failure. The plastic zone and deformation of the roadway roof in the fractured coal pillar are larger than that of continuous coal pillar, indicating that the continuous coal pillars mitigate the risk of groundwater inrush hazard effectively.

Water inrush hazards can be effectively reduced by a reasonable and accurate soft-measuring method on the water inrush quantity from the mine floor. This is quite important for safe mining. However, there is a highly nonlinear relationship between the water outburst from coal seam floors and geological structure, hydrogeology, aquifer, water pressure, water-resisting strata, mining damage, fault and other factors. Therefore, it is difficult to establish a suitable model by traditional methods to forecast the water inrush quantity from the mine floor. Modeling methods developed in other fields can provide adequate models for rock behavior on water inrush. In this study, a new forecast system, which is based on a hybrid genetic algorithm (GA) with the support vector machine (SVM) algorithm, a model structure and the related parameters are proposed simultaneously on water inrush prediction. With the advantages of powerful global optimization functions, implicit parallelism and high stability of the GA, the penalty coefficient, insensitivity coefficient and kernel function parameter of the SVM model are determined as approximately optimal automatically in the spatial dimension. All of these characteristics greatly improve the accuracy and usable range of the SVM model. Testing results show that GA has a useful ability in finding optimal parameters of a SVM model. The performance of the GA optimized SVM (GA-SVM) is superior to the SVM model. The GA-SVM enables the prediction of water inrush and provides a promising solution to the predictive problem for relevant industries.