概率积分法沉陷预计与参数反演优化算法及实现

任意多边形工作面沉陷预计及精确的求取概率积分法反演参数是开采沉陷研究的重点。提出了使用Delaunay三角剖分将任意多边形开采区域划分为若干小三角形,然后进行沉陷预计的方法,解决了任意多边形沉陷预计的难题;综合模矢法和遗传算法优点,提出了组合算法,提高了概率积分法参数反演精度和效率;开发了集实测数据处理、预计模型参数反演和移动变形预计模块于一体的地表移动变形数据处理软件。优化算法的提出及系统开发,可以更好的服务于"三下采煤"、矿区建筑物稳定性评估、矿区土地复垦、采煤塌陷地治理规划设计等工作。

Method and Implementation of Subsidence Prediction Based on the Probability Integral Method and Inversion of Optimal Parameters

Subsidence prediction and obtained inversion parameters of probability integral method accurately are the key in studying mining subsidence. In this paper, a new prediction approach of dividing the arbitrary polygonal working face into some small triangular areas by using Delaunay triangulation has been put forward. Then, the subsidence caused by mining of each triangular area can be calculated easily. This method has solved the difficulty of subsidence prediction of arbitrary polygonal working face. A new algorithm based on the combination of genetic algorithm and pattern search has been proposed. The algorithm can improve the accuracy and efficiency of inversion of parameters. The software of data processing of surface movement and deformation has been developed. The main function includes process of measured data, subsidence prediction and inversion of optimal parameters. The algorithm and the software are expected to serve for "coal mining under buildings, under railroad and under water", stability evaluation of mining area buildings and land reclamation.