岩移沉降观测中数据缺失的一种估算方法  

An estimation method of missing data in rock movement and subsidence observation

作　　者：赖礼泉 Lai Liquan(Geological Brigade of Southeast Fujian Province,Quanzhou 362000,China)

机构地区：[1]福建省闽东南地质大队,福建泉州362000

出　　处：《矿山测量》2021年第4期46-49,共4页Mine Surveying

摘　　要：在矿区的岩移沉降观测过程中,由于环境、人为等原因导致观测数据缺失问题,损害了整体观测数据的完整性,影响后续沉降趋势预测。文中提出了一种最大期望算法,通过计算极大似然估计,交替执行E步和M步,最终得到观测缺失数据填补值。利用某矿区实测数据进行实验,对比分析发现最大期望算法对缺失数据填补更接近实际测量值,在实际应用中有一定参考价值。During the observation of the rock movement and subsidence in the mining area, the problem of missing observational data due to environmental and human factors had damaged the integrity of the overall observation data and affected the subsequent subsidence trend prediction. In this paper, a maximum expectation algorithm was proposed, which calculated the maximum likelihood estimation, E steps and M steps were executed alternately, finally, the fill value of the missing data was obtained. Actual measured data was used to experiment to compara and analyze in a certain mining area, which found that the maximum expectation algorithm for filling the missing data was closer to the actual measured value, which had the certain reference value in practical applications.

关 键 词：岩移沉降观测 数据缺失 最大期望算法 填补值 

分 类 号：P258[天文地球—测绘科学与技术]