杨小虎;朱庆伟;沈宇恒;李航;

• 1:西安科技大学测绘科学与技术学院
• 2:西安科技大学西部矿井开采与灾害防治教育部重点实验室

摘要(Abstract)：

建筑物变形监测数据中存在着随机干扰和不确定性因素,而单一的数学模型预测结果精度较低,制约了变形预测的准确性。针对这一问题,文中采用了一种自适应Kalman滤波的灰色时序组合预测模型。首先,通过自适应Kalman滤波算法对原始数据进行去噪处理,动态的去除数据内部的随机干扰误差;然后,将灰色模型(GM模型)与时间序列分析模型(AR模型)相结合,得到拟合时间序列中的沉降量趋势项和沉降量随机时间序列剩余项,生成一种非线性组合模型;最后,对变形监测数据进行整理预测,并将该预测模型应用于建筑变形工程实例中,与GM(1,1)预测模型、GM(1,1)-AR预测模型通过平均残差、残差的方差和后验差比值进行对比分析。结果表明:该模型后验差比值可达到0.045 1,所得数据结果明显减小,预测精度显著提高,结果更加准确可靠。

关键词(KeyWords)： 动态数据处理;自适应Kalman滤波;GM(1,1)模型;AR模型;预测模型

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(51674195)

作者(Author): 杨小虎;朱庆伟;沈宇恒;李航;

Email:

DOI: 10.13800/j.cnki.xakjdxxb.2020.0522

参考文献(References)：

• [1] 高彩云,高宁.改进极限学习机的不同类型滑坡位移预测[J].西安科技大学学报,2018,38(4):683-689.GAO Cai-yun,GAO Ning.Different types of landslide displacement prediction with improved extreme learning machine[J].Journal of Xi’an University of Science and Technology,2018,38(4):683-689.
• [2] WEN H Y,HU J Y,PAN Y J,et al.Research on a kind of deformation prediction model of wavelet neural network[J].Applied Mechanics & Materials,2014,584-586:1933-1938.
• [3] 刘江,苗昌奇,陈柳阳.小波-时序组合模型在地铁变形监测预测中的应用研究[J].测绘与空间地理信息,2018,41(11):254-256.LIU Jiang,MIAO Chang-qi,CHEN Liu-yang.Application research of wavelet-time series combined model in metro deformation monitoring and prediction[J].Surveying and Spatial Geographic Information,2018,41(11):254-256.
• [4] 谭超.基于小波去噪-灰色预测混合模型在变形监测中的应用研究[D].西安:西北大学,2016.TAN Chao.Application of wavelet denoising-grey prediction hybrid model in deformation monitoring[D].Xi’an:Northwest University,2016.
• [5] 张磊,刘小明,来兴平,等.基于BP神经网络的急倾斜煤层耦合致裂方案优化[J].西安科技大学学报,2018,38(3):367-374.ZHANG Lei,LIU Xiao-ming,LAI Xing-ping,et al.Optimization of the coupling fracture scheme of steeply inclined coal seam based on BP neural network[J].Journal of Xi’an University of Science and Technology,2018,38(3):367-374.
• [6] 王新洲,樊倩,徐成权,等.基于小波变换和支持向量机的大坝变形预测[J].武汉大学学报(信息科学版),2008,33(5):469-471.WANG Xin-zhou,FAN Qian,XU Cheng-quan,et al.Dam deformation prediction based on wavelet transform and support vector machine[J].Geomatics and Information Science of Wuhan University,2008,33(5):469-471.
• [7] SU H,LI X,YANG B,et al.Wavelet support vector machine-based prediction model of dam deformation[J].Mechanical Systems and Signal Processing,2018,110:412-427.
• [8] 周吕,文鸿雁,韩亚坤.灰色预测与Kalman滤波在建筑物沉降变形分析中的应用[J].测绘科学,2014,39(4):149-151.ZHOU Lv,WEN Hong-yan,HAN Ya-kun.Application of grey prediction and Kalman filtering in building settlement deformation analysis[J].Journal of Surveying and Mapping,2014,39(4):149-151.
• [9] LU F M,JIANG T Y.The dangerous rock mass deformation analysis with application of Kalman filter method based on material properties and AR model[J].Advanced Materials Research,2013,738:109-112.
• [10] 钱荣荣.基于经验模态分解的动态变形数据分析模型研究[D].徐州:中国矿业大学,2016.QIAN Rong-rong.Research on dynamic deformation data analysis model based on empirical mode decomposition[D].Xuzhou:China University of Mining and Technology,2016.
• [11] 何君,杨国东.灰色预测理论在建筑物沉降中的应用研究[J].测绘通报,2012(3):63-64.HE Jun,YANG Guo-dong.Application of grey prediction theory in building settlement[J].Bulletin of Surveying and Mapping,2012(3):63-64.
• [12] ZHANG L T,ZHANG H.Application analysis of structure deformation monitoring prediction methods[J].Advanced Materials Research,2010,1067:2674-2677.
• [13] 陈晨,魏冠军,寇瑞雄.基于Kalman滤波的GM(1,1)-AR模型在高层建筑物沉降变形分析中的应用[J].测绘工程,2018,27(10):67-70.CHEN Chen,WEI Guan-jun,KOU Rui-xiong.Application of GM(1,1)-AR model based on Kalman filter in settlement analysis of high-rise buildings[J].Surveying and Mapping Engineering,2018,27(10):67-70.
• [14] LIU H F,REN C,ZHENG Z T,et al.Study of a gray genetic BP neural network model in fault monitoring and a diagnosis system for dam safety[J].ISPRS International Journal of Geo-information,2018,7(1):4.
• [15] 文小勇.基于小波分析的自适应卡尔曼滤波在地铁变形监测中的应用[D].西安:长安大学,2015.WEN Xiao-yong.Application of adaptive Kalman filter based on wavelet analysis in subway deformation monitoring[D].Xi’an:Chang’an University,2015.
• [16] 容静,刘立龙,康昊华,等.基于方差补偿自适应Kalman滤波的ARMA与PSO-SVM模型变形预测[J].大地测量与地球动力学,2018,38(7):689-694.RONG Jing,LIU Li-long,KANG Hao-hua,et al.Deformation prediction of ARMA and PSO-SVM models based on variance-compensated adaptive Kalman filtering[J].Journal of Geodesy and Geodynamics,2018,38(7):689-694.
• [17] 李奕.自适应卡尔曼滤波在变形监测数据处理中的应用研究——以太原万达广场A2区基坑监测为例[D].成都:成都理工大学,2012.LI Yi.Application of adaptive Kalman filter in deformation monitoring data processing:taking the foundation Pit monitoring in A2 area of Taiyuan Wanda plaza as an example[D].Chengdu:Chengdu University of Technology,2012.
• [18] MA X,LIU Z B.The kernel-based nonlinear multivariate grey model[J].Applied Mathematical Modelling,2018,56:217-238.
• [19] 段明旭,邱冬炜,李婉,等.改进灰色人工神经网络模型的超高层建筑变形预测[J].测绘科学,2017,42(4):141-146,183.DUAN Ming-xu,QIU Dong-wei,LI Wan,et al.Deformation prediction of super high-rise buildings based on improved grey artificial neural network model[J].Science of Surveying and Mapping,2017,42(4):141-146,183.
• [20] DUAN H M,XIAO X P.A multimode dynamic short-term traffic flow grey prediction model of high-dimension tensors[J].Complexity,2019:1-18.
• [21] YIN K D,GENG Y,LI X M.Improved grey prediction model based on exponential grey action quantity[J].Systems Engineering and Electronics,2018,29(3):560-570.
• [22] 田振凯,李瑞海,许彦国.动态灰色-时序模型在建筑物沉降预测中的应用[J].测绘与空间地理信息,2018,41(10):238-240,245.TIAN Zhen-kai,LI Rui-hai,XU Yan-guo.Application of dynamic grey-time series model in building settlement prediction[J].Surveying and Spatial Geography Information,2018,41(10):238-240,245.
• [23] 谢荣晖,郑东健.卡尔曼滤波-AR模型在大坝变形预测中的应用[J].人民黄河,2017,39(2):133-135.XIE Rong-hui,ZHENG Dong-jian.Application of Kalman Filter-AR model in dam deformation prediction[J].People’s Yellow River,2017,39(2):133-135.
• [24] 高宁,崔希民,高彩云.灰时序GM-AR模型在建筑物沉降变形预测中的应用[J].测绘科学,2013,38(1):139-141.GAO Ning,CUI Xi-min,GAO Cai-yun.Application of gray time series GM-AR model in prediction of building settlement deformation[J].Surveying and Mapping Science,2013,38(1):139-141.
• [25] GUO N,LI Y B.Application of grey theory in prediction of ground settlement in foundation pit[J].IOP Conference Series:Earth and Environmental Science,2018,170(2):126.