均值GM(1,1)模型在变形预测中的应用  

Application of mean GM(1,1)model in deformation prediction

作　　者：杨怀义[1] 邱利军[2] 张波[2] Yang Huaiyi;Qiu Lijun;Zhang Bo(Shanxi Branch of China National Geological Exploration Center of Building Materials Industry,Taiyuan 030031,China;Hebei University of Architecture,Zhangjiakou 075000,China)

机构地区：[1]中国建筑材料工业地质勘查中心山西总队,山西太原030031 [2]河北建筑工程学院,河北张家口075000

出　　处：《山西建筑》2021年第4期158-159,共2页Shanxi Architecture

基　　金：河北省教育厅2020年度河北省高等学校科学研究项目(Z2020222);张家口科技计划项目(2021028D)。

摘　　要：以变形监测实测20期高程数据为依据,选择前15期建立均值GM(1,1)模型,对后5期数据进行预测,并对建模拟合结果和预测结果进行分析,认为以均值GM(1,1)模型进行拟合预测精度满足要求。并且与以荷载和时间为影响因素建立的多元线性回归预测结果进行比较,相较于多元线性回归预测精度高。该研究对建筑物变形预测有一定的实际应用价值和参考意义。Based on the 20 periods of elevation data measured in deformation monitoring,the first 15 periods are selected to establish the mean GM(1,1)model,and the data of the last 5 periods are predicted.The modeling fitting results and prediction results are analyzed.It is considered that the fitting prediction accuracy of the mean GM(1,1)model meets the requirements.Compared with the prediction results of multiple linear regression established by load and time,the prediction accuracy of multiple linear regression is higher.This study has a certain practical application value and reference significance for building deformation prediction.

关 键 词：均值GM(1 1)模型 多元线性回归 变形监测 预测 

分 类 号：TU196.1[建筑科学—建筑理论]