| 基于核函数变换的NLPCC分析在洪水分类中的应用 |
| |
| 引用本文： | 刘玉邦,梁川.基于核函数变换的NLPCC分析在洪水分类中的应用[J].武汉理工大学学报(交通科学与工程版),2011,35(1). |
| |
| 作者姓名： | 刘玉邦 梁川 |
| |
| 作者单位： | 1. 四川大学水电学院,成都,610041;成都理工大学学术期刊编辑中心,成都,610059
2. 四川大学水电学院,成都,610041 |
| |
| 基金项目： | “十一五”国家科技支撑计划项目资助(批准号:2007BAD88B08) |
| |
| 摘 要： | 通过拟线性化变换和降维去噪,得出多维分类指标的低维主成分值,然后通过对每一待分类样本的低维主成分值进行聚类分析,最终得出洪水的自然分类结果.洪水分类实例结果表明,该计算方法不需要复杂的计算机专业知识和优化算法知识,原理清楚,计算简单,结果客观有效,不失为一种洪水分类评价的新途径.选择适宜的非线性变换函数是正确应用该方法的关键,同时对于能够预先给出分类指标值且数值范围较小时,需要指标标准数值判定和聚类效果判定相结合.
|
| 关 键 词： | 洪水分类 非线性主成分-聚类(NLPCC)分析 核函数 |
Kernel-Based Nonlinear Principal Component Transform-Cluster Analysis and Its Application in the Flood Classification |
| |
| Liu Yubang,Liang Chuan.Kernel-Based Nonlinear Principal Component Transform-Cluster Analysis and Its Application in the Flood Classification[J].journal of wuhan university of technology(transportation science&engineering),2011,35(1). |
| |
| Authors: | Liu Yubang Liang Chuan |
| |
| Institution: | Liu Yubang1,2) Liang Chuan1)(College of Water Resource & Hydropower,Sichuan University,Chengdu 610041,China)1)(Academic Journal Editors Center,Chengdu University of Technology,Chengdu 610059,China)2) |
| |
| Abstract: | A natural classification of flood will be got through the pseudo-linear transformation and dimensionality reduction de-noising and the cluster analysis to low-dimensional and principal component values of each sample to be classified.The results of the instance of flood classification show that this method is regarded as a classification and evaluation of new approaches to flood,for it is easy to compute and the results got by this method is objective and effective without any computer expertise and knowled... |
| |
| Keywords: | flood categories Nonlinear Principal Component-Cluster Analysis Gaussian kernel function |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
