开采沉陷预计参数模糊聚类分析研究

发布时间：2018-05-10 10:15

本文选题：预计参数 + 模糊聚类分析　；参考：《辽宁工程技术大学》2015年硕士论文

【摘要】：随着“三下”开采强度的增加,开采沉陷地质灾害日益严重。为了减少“三下”开采引起的开采沉陷灾害,能否准确地进行地表变形预计就显得至关重要,因此如何选取预计参数也成为了一个比较关键的问题。针对现有预计参数求取方法存在的不科学性和不确定性问题,采用模糊聚类分析和回归分析法来确定预计参数与地矿特征之间的复杂关系。不仅采用特征选取组合方法来确定开采沉陷主要地矿特征,还采用改进的模糊聚类算法将开采沉陷岩移观测站划分为四个相似现象群,最后从隶属关系模糊与否出发,建立了基于隶属关系判定算法的预计参数分段模型,克服了现有预计参数求取方法的不足。通过对徐州矿区和辽西北矿区的实例分析,验证了基于隶属关系判定算法的分段模型适用于单个和区域性矿区的准确性和可靠性,为先验信息比较缺乏的新矿区提供了预计参数选取方法。结果表明：此模型在开采沉陷预计参数预测方面具有重大的理论价值和指导意义,这也为全国矿区开采沉陷的准确预测、灾害预警评价和“三下”开采合理设计研究提供了新的技术手段和有力的信息支持。
[Abstract]:With the increase of mining intensity, the geological hazard of mining subsidence is becoming more and more serious. In order to reduce the subsidence disaster caused by "three down" mining, it is very important to predict the surface deformation accurately, so how to select the predicted parameters becomes a key problem. In view of the unscientific and uncertain problems existing in the existing prediction methods, fuzzy cluster analysis and regression analysis are used to determine the complex relationship between the prediction parameters and the geological features. Not only the feature selection combination method is used to determine the main geological and mineral characteristics of mining subsidence, but also the improved fuzzy clustering algorithm is used to divide the mining subsidence observation station into four similar phenomenon groups. A segmented model of prediction parameters based on membership relationship decision algorithm is established, which overcomes the shortcomings of the existing methods for obtaining predicted parameters. By analyzing the examples of Xuzhou mining area and northwest Liaoning mining area, the accuracy and reliability of the segmental model based on the subordination relation decision algorithm for single and regional mining areas are verified. It provides a method to select the predicted parameters for the new mining area which is lack of prior information. The results show that this model has great theoretical value and guiding significance in predicting the predicted parameters of mining subsidence, which is also an accurate prediction of mining subsidence in China. The research of disaster early warning evaluation and reasonable design of three-down mining provides new technical means and powerful information support.
【学位授予单位】：辽宁工程技术大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD327

【参考文献】