利用成长模型的自适应垂直坐标序列分析  

Analysis of adaptive vertical coordinate series based on growth model

作　　者：王梦婷[1] 聂建亮[2] 郭春喜[2] 田婕[2] 王斌[2] WANG Mengting;NIE Jianliang;GUO chunxi;TIAN Jie;WANG Bin(School of Science,Xi′an Univ.of Arch.&Tech.,Xi′an 710055,China;Geodetic Data Processing Centre,National Administration of Surveying,Mapping and Geoinformation of China,Xi′an 710054,China)

机构地区：[1]西安建筑科技大学理学院,陕西西安710055 [2]国家测绘地理信息局大地测量数据处理中心,陕西西安710054

出　　处：《西安建筑科技大学学报（自然科学版）》2018年第5期761-765,共5页Journal of Xi'an University of Architecture & Technology(Natural Science Edition)

基　　金：国家自然科学基金项目(41574003;41474015;41604001);国家测绘地理信息局2015年公益性行业科研专项(201512004);科技基础性工作专项重点项目(2015FY210400)

摘　　要：针对监测点的垂直运动不具有均匀性,采用Weibull等地面沉降成长模型为运动模型,提高动力学模型的可靠性;利用预测残差构造自适应因子,调整动力学模型与观测方程的对滤波解贡献.以沉降区域实测点垂向坐标序列为例,该方法能够提高状态方程预报值与滤波解的可靠性.Growth models as Weibull Model are used as the dynamic equation to improve the accuracy for the character of vertical movement in view of the fact that the monitored points has no uniformity.The adaptive factor is obtained with predicted residuals to adjust the contribution between the dynamic and observation equation.Finally,actual data of monitored points are given to test the new adaptive filtering.It is shown that the new algorithm can improve the accuracy of predicted state,and further enhance the reliability of the filtering.

关 键 词：Gompertz模型 LOGISTIC模型 Weibull模型 自适应滤波 

分 类 号：TU478[建筑科学—结构工程]