Providing accurate mining-induced water fractured zone over the longwall goaf prediction methods has become increasingly significant because of the grand economic and technological challenge for deeper extraction. Taking the No. 839 coalface in the Qingdong mining area, as a case study, a comprehensive study of the water-flowing fractured zone (WFFZ) formation process and peak height were carried out by the radial basis function (RBF) neural network model, parallel electrical monitoring (PEM) technology, numerical simulation model and the empirical formula method. The results show that the destruction does not occur simply layer-by-layer from the bottom up. Rather, the fractured areas appear first in soft layers above the goaf. With the advance of footage, the destruction gradually extends from soft layers to hard layers, and the WFFZ is formed when the fractured areas fit closely together. Also, the change process of the WFFZ shape is closely related to rock mass heterogeneity. In addition, the results of the comparison suggest that the RBF neural network model had more accuracy in predicting the height of the WFFZ than the numerical simulation model and empirical formula method.
The change law of the shape of the WFFZ is revealed by means of field investigation and numerical simulation.
The prediction method of the height of the WFFZ is established based on the neural network model.
The RBF model has more accuracy than the numerical simulation model and empirical formula method.