基于素数分解排序的水汽层析代数重构算法
发布时间：2019-06-28 10:49
【摘要】：该文首先介绍了解决层析方程组病态问题的代数重构算法,并对影响该算法迭代结果的松弛因子、投影次序、停止规则进行了讨论,提出了基于素数分解排序的水汽层析代数重构算法。其次,详细陈述了该算法的3个核心内容:分组排序、素数分解排序、非负约束。分组排序可提高反演结果精度;素数分解排序保证了解的无偏性;非负约束则使迭代结果符合水汽值的非零特性。最后,利用2015年8月13日UTC 0时的香港参考站的观测数据对6组实验方案进行了分析,证明了基于素数分解的投影排序对水汽层析结果具有重要影响,且通过该方法可以有效提高代数重构算法解算GNSS水汽层析方程组的精度。
[Abstract]:In this paper, an algebra reconstruction algorithm for solving ill-conditioned problems of tomographic equations is introduced, and the relaxation factors, projection order and stop rules that affect the iterative results of the algorithm are discussed, and a steam chromatography algebra reconstruction algorithm based on prime number decomposition and sorting is proposed. Secondly, the three core contents of the algorithm are described in detail: grouping sorting, prime decomposition sorting and non-negative constraints. Grouping sorting can improve the accuracy of inversion results, prime decomposition sorting ensures the unbiased solution, and non-negative constraints make the iterative results conform to the non-zero characteristics of water vapor value. Finally, six groups of experimental schemes are analyzed by using the observation data of Hong Kong reference station at UTC 00 on August 13, 2015. It is proved that the projection ranking based on prime decomposition has an important influence on the results of water vapor chromatography, and the accuracy of solving GNSS water vapor chromatography equations can be effectively improved by using this method.
【作者单位】： 中国矿业大学环境与测绘学院;
【基金】：国家自然科学基金项目(41504032) 江苏省自然科学基金项目(BK20150175) 高等学校博士学科点专项科研基金项目(20130095110022)
【分类号】：P228.4;P407
本文编号：2507242
[Abstract]:In this paper, an algebra reconstruction algorithm for solving ill-conditioned problems of tomographic equations is introduced, and the relaxation factors, projection order and stop rules that affect the iterative results of the algorithm are discussed, and a steam chromatography algebra reconstruction algorithm based on prime number decomposition and sorting is proposed. Secondly, the three core contents of the algorithm are described in detail: grouping sorting, prime decomposition sorting and non-negative constraints. Grouping sorting can improve the accuracy of inversion results, prime decomposition sorting ensures the unbiased solution, and non-negative constraints make the iterative results conform to the non-zero characteristics of water vapor value. Finally, six groups of experimental schemes are analyzed by using the observation data of Hong Kong reference station at UTC 00 on August 13, 2015. It is proved that the projection ranking based on prime decomposition has an important influence on the results of water vapor chromatography, and the accuracy of solving GNSS water vapor chromatography equations can be effectively improved by using this method.
【作者单位】： 中国矿业大学环境与测绘学院;
【基金】：国家自然科学基金项目(41504032) 江苏省自然科学基金项目(BK20150175) 高等学校博士学科点专项科研基金项目(20130095110022)
【分类号】：P228.4;P407
【相似文献】
相关硕士学位论文 前1条
1 张尊良;GPS三维水汽层析的算法研究与实现[D];中国石油大学(华东);2014年
，本文编号：2507242
本文链接：https://www.wllwen.com/kejilunwen/dizhicehuilunwen/2507242.html
