关键词: BP 神经网络, 导水裂隙带高度, 影响因素, 样本

Abstract: In order to predict exactly the forming height of water flowing fractured zone, based on summing up methods and theories to forecast the height of water flowing fractured zone, combined with a lot of practical information analysis, the main eight factors influencing the height of water flowing fractured zone have been found out, which are mining depth, coal seam inclination angle, coal seam thickness, coal seam hardness, rock structure, the uniaxial compressive strength of rock, mining thickness and goaf plagioclase. According to the typical cases in China, the forecasting model of the height of water flowing fractured zone based on BP neural networks has been built, and the input samples and testing samples needed by BP neural networks have been determined. The optimization of network model has been got after Matlab software is used in the training of network. The height of water flowing fractured zone of 11011 workface in Zhaogu No.1 Mine, Jiaozuo cola field has been forecasted according to the established network model. Compared with measured results, it has shown that the calculation results obtained by the BP neural network model are closer to the reality than those obtained according to the empirical formula.

Key words: BP neural network, the height of water flowing fractured zone, influencing factors, sample