一种相关观测的Partial EIV模型求解方法

发布时间：2018-04-22 06:33

本文选题：总体最小二乘 + 相关观测　；参考：《测绘学报》2017年08期

【摘要】：Partial errors-in-variables(Partial-EIV)模型作为EIV模型的扩展形式,其构造方式更有规律,解算方法更为简便,能有效应用于实际情况。针对已有Partial EIV模型方法未考虑观测向量和系数矩阵存在相关性这一情况,通过提取观测向量和系数矩阵组成的增广矩阵中非重复出现的随机元素,构建更具一般适用性的Partial EIV模型,在该模型的基础上,将特殊假定条件扩展到不限定观测数据相关性的一般情况,详细推导了观测向量和系数矩阵元素相关且不等精度情况下的加权总体最小二乘方法,通过算例试验,并与目前已有的解决EIV模型相关观测情况下的方法进行了比较分析,研究表明本文方法可以提高计算效率,更具一般性,特别是对于观测向量和系数矩阵中存在常数元素和重复元素的情况。
[Abstract]:As an extended form of EIV model, the Partial errors-in-variables-Partial-Eve) model is constructed in a more regular manner, and its solution method is simpler and can be effectively applied to the actual situation. In view of the fact that the existing Partial EIV model methods do not consider the correlation between the observation vector and the coefficient matrix, the augmented matrix composed of the observation vector and the coefficient matrix is extracted. A more general applicable Partial EIV model is constructed. Based on the model, the special assumptions are extended to the general case of unqualified correlation of observed data. In this paper, the weighted population least square method with variable precision and correlation between the observation vector and coefficient matrix elements is derived in detail. The method is compared with the existing methods for solving the related observation of EIV model through an example. The results show that the proposed method can improve the computational efficiency and be more general, especially in the case of constant elements and repeated elements in the observation vector and coefficient matrix.
【作者单位】：东华理工大学测绘工程学院;流域生态与地理环境监测国家测绘地理信息局重点实验室;江西省数字国土重点实验室;武汉大学测绘学院;
【基金】：国家自然科学基金(41664001;41204003) 江西省杰出青年人才资助计划项目(20162BCB23050) 国家重点研发计划(2016YFB0501405) 江西省教育厅科技项目(GJJ150595)~~
【分类号】：P207

【相似文献】