Comprehensive study on identification of water inrush sources from deep mining roadway

To accurately identify the location of the water inrush sources, correlation analysis and factor analysis were used to discuss the similarity of hydrochemical control mechanism of different aquifers and inrush water. On this basis, Fisher and BP neural network are used to judge the probability of inrush water. Finally, it is verified by hydrological observation holes. According to the water sources mixing model, the water inrush sources of 11601 working face were mainly No. 13 stratum Carboniferous limestone water with a small amount of No. 10 stratum Carboniferous limestone water, and the mixing ratio is 11:1. Combining the results of water inrush source identification, water yield mutation, geophysical exploration results, and underground drilling, it is speculated that there are two faults in this working face. The normal fault stops at the top of No. 13 stratum Carboniferous limestone aquifer, and the reverse fault develops to No. 13 stratum Carboniferous limestone aquifer. This research is of great significance for the identification of mine water sources and guidance for the prevention and control of water inrush.