Abstract: The mining environment of deep coal seam has the characteristics of high confined water pressure, high ground stress and strong disturbance. Mining is more likely to induce high-pressure karst water to flow into the mine along the weak zone of geological bodies such as faults. In order to comprehensively and accurately predict the fault water inrush from coal seam floor in Huangyuchuan Coal Mine, geophysical methods such as three-dimensional seismic and transient electromagnetic were used to explore the fault distribution in the study area and determine the fault location and water conductivity. Aiming at the problem that it is difficult to quantify and predict the potential water inrush risk of working face crossing fault in pressured area, the factors affecting fault stability are screened and extracted, and the fuzzy analytic hierarchy process (FAHP) and grey correlation analysis (GRA) risk evaluation model are integrated to construct a FAHP-GRA coupling fault zone water inrush risk evaluation model. FAHP is used to determine the weights of 10 factors affecting fault water inrush, and GRA is used to determine the correlation degree of each index in the mathematical model. The risk level of fault water inrush is obtained by the correlation degree of each index obtained by evaluation. The results show that the weight of confined water pressure in the evaluation model is the largest (0.146 7), while the weight of mining depth and fault water conductivity is the smallest (0.029 34), and the grey weighted correlation degree of hydrogeological conditions is the largest (0.253); the NNE-NE trending faults in the pressured zone are mainly high-risk and relatively dangerous, and the NNW-NW trending faults are mainly relatively safe and safe. The evaluation model is suitable for the investigation of water inrush risk in coal mines with complex geological structures, and realizes the purpose of “effective evaluation, rapid diagnosis and precise prevention” of fault water inrush.