黄土缓坡丘陵采煤塌陷预测中概率积分法适用性研究

发布时间：2018-08-01 17:23

【摘要】：我国煤炭资源的85%来自井工开采。井工开采必然导致地表下沉、裂缝的产生,从而影响矿区生产、人民生活。加之,我国的煤矿区大多分布在生态脆弱的晋、陕、蒙区,因此,对井工开采导致的地面塌陷进行深入探讨迫在眉睫,特别是在黄土丘陵区塌陷地的塌陷形态特征研究的基础上对塌陷地塌陷程度进行准确预测可为生态脆弱煤矿区的塌陷地的预测、治理及恢复提供依据。目前塌陷预测使用较多的方法为概率积分法,该方法在平原区较为适用,在黄土缓坡丘陵区的适用性有待研究。因此,本文以平朔井工三矿903工作面为研究对象,运用概率积分法作为平朔井工三矿903工作面塌陷预测的方法并对工作面上100×100m塌陷区样地进行了实地调查,通过对塌陷地各属性值(塌陷面积、塌陷深度、裂缝数等)进行统计分析和空间自相关分析,研究了平朔矿区塌陷地形态特征;最后将概率积分法的预测结果与实测结果对比分析,讨论该方法在黄土缓坡丘陵区塌陷预测的适用性。本文的主要研究结论有:(1)运用概率积分法,并借助中国矿业大学研制的开采沉陷软件(MSAS),对井工三矿903工作面进行塌陷预测。通过下沉等值线图和下沉云图,得到下沉最大区位于903工作面中心,为9269mm。对样地范围内塌陷预测结果表明,塌陷下沉范围为1-9m,而对样地塌陷地DEM进行分析结果表明,样地自东向西100米范围内塌陷程度逐渐加重,塌陷下沉范围为1-3.5m。(2)平朔矿区井工开采对地表的破坏主要表现为地表的下沉和裂缝。经预测计算和遥感解译,井工三矿903工作面占地塌陷面积有220.02 hm2,其中耕地占200.18hm2,林地占19.82hm2。对塌陷区样地(100×100 m)实地测量结果表明,样地范围内塌陷面积和裂缝数空间分布均匀,但塌陷深度空间分布并不一致,对塌陷地空间研究表明:样地的中部和南部塌陷较严重,而西北部塌陷不严重,这可能和井工开采的工艺及地质条件的差异性有关。(3)将塌陷区预测下沉值与实测值进行对比分析,计算得实际值的均方根误差是标准差的2.18倍,根据标准化的均方根误差定义,预测塌陷下沉值与实测塌陷下沉值之间的误差较大,这可能与丘陵区地质特征、样地范围等因素有关系。
[Abstract]:85% of China's coal resources come from well mining. Well mining will inevitably lead to surface subsidence and cracks, thus affecting mining production and people's lives. In addition, most of the coal mining areas in China are located in the ecologically fragile areas of Shanxi, Shaanxi and Mongolia. Especially on the basis of the study on the collapse morphological characteristics of the subsided land in the loess hilly area, the accurate prediction of the collapse degree of the subsided land can provide the basis for the prediction, treatment and restoration of the subsided land in the ecologically fragile coal mine area. At present, the probability integration method is widely used in the prediction of collapse, which is more suitable in the plain area, and the applicability of the method in the loess gentle slope hilly area needs to be studied. Therefore, this paper takes 903 face of Pingshuo coal mine as the research object, uses probability integration method as the method to predict the collapse of 903 face in Pingshuo coal mine, and makes a field investigation on the sample land of 100 脳 100m collapse area on the face. Based on the statistical analysis and spatial autocorrelation analysis of the collapsing land values (subsidence area, collapse depth, crack number, etc.), the morphological characteristics of subsidence land in Pingshuo mining area are studied. Finally, the prediction results of the probabilistic integration method are compared with the measured results, and the applicability of the method to the prediction of collapse in the loess hilly region is discussed. The main conclusions of this paper are as follows: (1) the collapse prediction of 903 face in Jinggong No.3 Mine is carried out by using probabilistic integration method and with the help of mining subsidence software (MSAS), developed by China University of Mining and Technology. The maximum subsidence area is located at the center of 903 face, which is 9269mm. The prediction results of subsidence in the sample plots show that the subsidence range is 1-9 m, and the DEM analysis results show that the subsidence degree of the sample plots is gradually increasing within 100m from east to west. The subsidence range is from 1 to 3.5 m. (2) the surface damage caused by mining in Pingshuo mining area is mainly manifested by surface subsidence and fracture. Through prediction calculation and remote sensing interpretation, the subsidence area of 903 face in Jinggong No. 3 Coal Mine is 220.02 mm ~ 2, of which the cultivated land is 200.18hm2and the woodland is 19.82hm2. The field survey results of the sample land (100 脳 100m) in the subsidence area show that the collapse area and the number of cracks are uniformly distributed in the sample area, but the spatial distribution of the collapse depth is not consistent. The research on the collapsing site space shows that the collapse in the central and southern part of the sample plot is more serious. However, the northwestern collapse is not serious, which may be related to the difference of mining technology and geological conditions. (3) by comparing the predicted subsidence value with the measured value, the root mean square error of the actual value calculated is 2.18 times of the standard deviation. According to the standard definition of root mean square error, the error between the predicted subsidence value and the measured subsidence value is large, which may be related to the geological characteristics of the hilly area and the range of sample plots.
【学位授予单位】：中国地质大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD327

【相似文献】