【作者】 王斌；

【导师】 刘历波； 李旭光；

【作者基本信息】 河北工程大学 ， 建筑与土木工程（专业学位）， 2020， 硕士

【摘要】 煤炭开采会引起地表移动甚至地面坍塌,给生态环境和人类生产带来危害,因此在煤炭开采工程中进行地表移动规律研究具有一定的现实意义。所以在开采过程中,监测数据的真实有效性和对数据进一步预测分析是非常重要的。针对卡尔曼滤波在地表移动监测过程精度低和稳定性差等缺陷,对卡尔曼滤波进行改进并提出极大后验自适应扩展卡尔曼滤波。以山西岳城矿区Ⅲ1301工作面数据为基础数据进行开采沉陷预计研究。基于以上背景,本文的主要工作和成果如下:1.针对卡尔曼滤波和扩展卡尔曼滤波在非线性系统中的不足,利用极大后验估计原理,提出极大后验自适应扩展卡尔曼滤波。通过极大后验和自适应性来简化扩展卡尔曼滤波的计算复杂度和扩展卡尔曼滤波依赖局部非线性的缺陷,提升了算法计算精准度,改善了滤波依赖局部非线性的缺陷。2.利用山西岳城矿区Ⅲ1301工作面地表移动监测站实测数据,对卡尔曼滤波、扩展卡尔曼滤波和极大后验自适应扩展卡尔曼滤波与实测数据进行对比分析并选取最优滤波进行预测。从对比分析结果上可以看出极大后验自适应扩展卡尔曼滤波精准度为80%,扩展卡尔曼滤波精准度50%,标准卡尔曼滤波精准度30%。使用极大后验自适应扩展卡尔曼滤波进行实测数据预测,结果表明预测值与实际值大部分在10mm内,最大差值为25mm,滤波稳定性好,精准度高。3.针对开采过后发生的残余变形,本文结合残余变形预计理论对本矿开采后进行预测分析,验证极大后验自适应扩展卡尔曼滤波在残余变形中的可行性。结果表明,滤波预计的残余变形和地表移动变形期内的变化规律是一致的,满足在衰退期的下沉速度界限要求。且将预测效果与残余变形预计理论相对比发现,90%预计值与理论值相差在1mm范围内,表明极大后验自适应扩展卡尔曼滤波可以应用于该地区的残余变形预计分析中。更多还原

【Abstract】 Coal mining will cause surface movement and even surface collapse,which will bring harm to the ecological environment and human’s normal life.Therefore,it is of certain practical significance to study the law of surface movement in coal mining projects.Hence,in the mining process,the true validity of monitoring data and the further prediction and analysis of data are very important.Aiming at the defects such as low precision and poor stability of KF(Kalman Filter)in the surface movement monitoring process,the KF is improved and the MPAEKF(Maximum A Posteriori Adaptive Extended Kalman Filter)is proposed.Shanxi yuecheng Ⅲ 1301 face data mining area based data mining subsidence is expected to study.Based on the above background,the main work and achievements of this paper are as follows:1.Aiming at the deficiency of KF and EKF(Extended Kalman Filtering)in nonlinear systems,the MPAEKF is proposed by using the maximum posterior estimation principle.The computational complexity of the KEF and the defect that the EKF depends on local nonlinearity are simplified by means of the great posterior and the self-adaptability,and the computational accuracy of the algorithm is improved and the defect that the filter depends on local nonlinearity is improved.2.The use of surface movement monitoring station of shanxi yuecheng mining Ⅲ 1301 working face of the measured data,the KF and EKF and MPAEKF and the measured data were analyzed and selected the optimal filter.From the comparative analysis results,it can be seen that the precision of MPAEKF is 80%,the precision of EKF is 50%,and the precision of standard kf is 30%.MPAEKF was used to predict the measured data.The results showed that the predicted value and the actual value were mostly within 10 mm,and the maximum difference was 25 mm.The filter had good stability and high accuracy.3.In view of the residual deformation after mining,this paper combined with the residual deformation prediction theory to conduct prediction analysis after mining and verify the feasibility of MPAEKF in residual deformation.The results show that the variation law of residual deformation predicted by filtering is consistent with that of surface movement deformation,and meets the requirements of the limit of subsidence velocity in the decline period.Moreover,by comparing the prediction effect with the prediction theory of residual deformation,it is found that 90% of the predicted value is within 1mm of the theoretical value,indicating that the MPAEKF can be applied to the prediction analysis of residual deformation in this region.更多还原