| 多源、多类、多时态非线性数据处理的整体降维解算 |
| |
| 引用本文： | 陶华学,郭金运.多源、多类、多时态非线性数据处理的整体降维解算[J].大地测量与地球动力学,2003,23(1):1-4. |
| |
| 作者姓名： | 陶华学 郭金运 |
| |
| 作者单位： | 山东科技大学地球信息科学与工程学院,泰安,271019 |
| |
| 基金项目： | 国家自然科学基金资助项目 (批准号 40 1740 0 3) |
| |
| 摘 要： | 针对当今各国十分关注、大力倡导的“数字地球”、“数字国家”、“数字矿山”等科学工程构建中的多源，多维、多类型，多时态，多糖度并具有非线性特征的数据处理，及其函数模型中同时包含有非随机参数和随机参数，而随机参数又常常是随时间动态变化的情况，如仍采用经典的最小二乘处理方法或一般的非线性最小二乘数据处理方法，是不准确、不科学的。为此，提出了一种新的数据处理方法，即广州非线性动态最小二乘数据处理方法。针对广州非线最小二乘问题维数高的特殊结构，在已研究提出的分离迭代求解模型的基础上，提出了另一种新的整体降维解算的模型和算法，使原问题庞大的高维方程组的解算得以简化，将待求参数分离求解，大大减少了计算工作量，为多源，多维，多类型，多时态，多精度的非线性数据处理开辟了另一新途径。
|
| 关 键 词： | 数据处理 广义非线性动态最小二乘法 降维解算 随机参数 空间数据 数字地球 |
| 文章编号： | 1671-5942(2003)01-0001-04 |
| 修稿时间： | 2002年9月2日 |
INTEGRAL DETRACTED-DIMENSIONAL SOLUTION METHOD FOR NONLINEAR PROCESSING OF DATA WITH MULTI-SOURCES,DIFFERENT TYPES AND MULTI-STATES |
| |
| Tao Huaxue and,Guo Jinyun.INTEGRAL DETRACTED-DIMENSIONAL SOLUTION METHOD FOR NONLINEAR PROCESSING OF DATA WITH MULTI-SOURCES,DIFFERENT TYPES AND MULTI-STATES[J].Journal of Geodesy and Geodynamics,2003,23(1):1-4. |
| |
| Authors: | Tao Huaxue and Guo Jinyun |
| |
| Abstract: | Data are very important to build the digital earth. Data come from many sources, have different types and temporal states. These data also have many dimensions and precisions. Relations between one type of data and another one, or between data and unknown parameters are more nonlinear. The unknown parameters are non-random or random, among which the random parameters often dynamically vary with time. Therefore it is not accurate and reliable to process the data with the classical least squares method or the method of the common nonlinear least squares. So a generalized nonlinear dynamic least squares method to process these data is put forward in the paper. Based on the separated iterative solution model, a new integral detracted-dimensional model and solution are put forward to process the data in the paper. First the searching direction in the non-constraint optimal problem can be solved with the separated solution method. So the first order derivative must be solved. Then the integral searching direction should be obtained. The method can simplify the solution of the complex, high-dimensional equation group and decease the loading work for it can decompose the unknown parameters based on the parameters' characteristics. It opens up a new method for the nonlinear processing of data with many sources, many dimensions, different types, temporal states and many precisions. |
| |
| Keywords: | method for generalized nonlinear dynamic least squares many sources different types temporal states detracted-dimensional solution |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
