稀疏矩阵压缩技术在CPⅢ网平差中的应用研究  

Research on application of sparse matrix to the compression technology in the adjustment of CPⅢ network

作　　者：周凌焱 刘成龙[1] 张强[1] 孙维亚[1] 杨雪峰[1] 

机构地区：[1]西南交通大学地球科学与环境工程学院,四川成都611756

出　　处：《铁道科学与工程学报》2014年第6期142-147,共6页Journal of Railway Science and Engineering

基　　金：中央高校基本科研业务费专项资金资助(SWJTU12BR014/SWJTU12ZT07);国家自然科学基金资助项目(51178404);"2011计划"轨道交通安全协同创新中心经费资助项目

摘　　要：在高铁测量领域,由于引入了智能型全站仪、电子水准仪等先进仪器,导致控制网规模越来越大,网平差速度越来越慢,给高速铁路的测量与数据处理带来一定困扰。基于矩阵的行、列压缩的存储原理,利用压缩矩阵与稀疏矩阵的乘法代替矩阵间的乘法,实现矩阵的快速乘法计算;同时采用"循环变量重新编号法"进行矩阵的求逆运算,从而大幅度提高轨道控制网等大型控制网平差计算速度。与其他优化方法相比,新方法具有计算速度快、便于理解、易于实现等特点,为目前国内高铁测量软件的维护升级提供一条可行的思路。In high-speed railway surveying,with the application of the robotic total station,electronic level me-ter and other advanced instruments,the scale of control network becomes increasingly bigger and the speed of ad-justment gets slower and slower,which brings some trouble to the survey and data processing of high -speed rail.Based on the storage principle of matrix compressed in row and column,the multiplication of compressed matrix and sparse matrix was adopted to replace the multiplication between matrixes,to achieve rapid matrix mul-tiplication.Meanwhile,employed the method of‘loop variable reordering’to conduct inverse matrix,the adjust-ment speed of huge control network,such as track control network,was dramatically improved.Compared with the other optimization methods,this new method is fast,easy to understand and implement,which provides a feasible idea for maintaining and upgrading the current surveying software used in high-speed railway of China.

关 键 词：稀疏矩阵 矩阵压缩 网平差 高铁测量 

分 类 号：P207[天文地球—测绘科学与技术]