| With the increasing tension of coal resources, under-water mining has become a hot issue of energy development issues. Especially in Mid-Eastern China,shallow covered thick loose aquifer, in mining process in the Shang covered rock occurred water sand bursting Chung of disaster often occurred. In addition,the data of coal mine water inrush is complex, nonlinear and small sample, which increases the difficulty of risk evaluation and prediction. It brings great threat to the safety production of coal mine,and seriously restricts the development of industry. In order to ensure the safety of coal mine mining, this paper relies on the powerful database processing and programming ability of MATLAB,then takes the 3 coal seam of Ge Ting coal mine as an example to evaluate the risk of water inrush in overlying strata.Firstly,we collected the data of coal mine geology,hydrogeology,engineering geology. Based on the direct current method,3D seismic exploration and controllable source ,the water-filling performance of the roof rock is explored. In this paper, it is discussed in detail for the distribution of the regional aquifers and aquiclude,and the water abundance,distribution and development of each fault.Then we analyzed theoretically the mechanism of roof water inrush,,explored the factors that affect the water resistance of roof strata,and studied the relationship between disaster formation and various influencing factors. Study on failure mechanism of overlying rock and analysis of mining overburden failure characteristics. Using the FLAC3D process of simulative coal seam mining under thick alluvium and thin bedrock,during excavation of overburden failure law and mechanism analysis of determining overlying strata to the influencing factors of water and sand gushing,and many factors and correlation analysis between sand overlying water piping.Finally,on the basis of MATLAB2012a, this paper constructs a model to determine the main controlling factors of sand and water inrush from the analytic hierarchy process,and combines with the sensitivity analysis of YAAHP to select the evaluation index of the sand inrush risk of the thick unconsolidated layer thin bedrock.With the help of MATLAB2012a software,we construed risk assessment model of principal component entropy weight and determined the weight of evaluation index. The Ostu algorithm in image segmentation was introduced to determine the threshold of water inrush in thick unconsolidated layers. Then we carried out the risk zoning according to the results of the model calculation. Finally,the model is designed and applied to the coal mine to provide the risk assessment and prediction for the coal mining under the same geological conditions.After analysis,the risk zoning is as follows: research target of dangerous regional and more dangerous regional are mainly concentrated in the area of drilling No. 1,9~14, 32~33 and No.38,which are located in the northwest,southwest and northern regions of the coal seam. The safe area and the safer area are mainly concentrated in the area of drilling No.3~4, 6~9,15~30 and No.34~37, which are located in the eastern and western part of the coal seam. The transition area is mainly concentrated in No5,17~23 and No.39,which is located in the middle part of the coal seam. |