The working face of North China coalfield is threatened by water inrush from the bottom aquifer. Accurate prediction of the failure depth of the working face floor (FFD) is the key to control the water disaster of the coal field. The measured data of floor failure depth of 20 working faces in 10 mining areas including Jining mining area, Fengfeng mining area, Huainan mining area and Jiaozuo mining area were collected through investigation. The multiple regression model was used to fit the formula, and the information entropy in information theory was used to calculate the weight of each factor. The prediction method of floor failure depth in North China coalfield based on multiple regression model and information entropy was obtained. Compared with the three empirical formulas recommended in the specification and the formula proposed by Dong et al., the predicted value obtained by this method was closer to the measured value, and the average relative error was only 8.4 %. Finally, eight working faces of four coal mines in North China were taken as an example for application, and good application results were obtained, which verified the universality of this method in North China coalfield. Finally, the local sensitivity analysis of the weight of each factor in the model is carried out. The sensitivity of each factor weight from large to small is the working face slope length, mining depth, mining thickness and coal seam inclination. The results show that this method has high accuracy in predicting the floor failure depth of working face in North China coalfield, which can provide reference for the calculation of floor failure depth of working face in North China coalfield.
 

North China coalfield; Floor failure depth; Multiple regression; Forecasting model