| 小波变换-偏最小二乘法用于柴油近红外光谱分析 |
| |
| 引用本文： | 田高友,褚小立,袁洪福.小波变换-偏最小二乘法用于柴油近红外光谱分析[J].计算机与应用化学,2006,23(10):971-974. |
| |
| 作者姓名： | 田高友 褚小立 袁洪福 |
| |
| 作者单位： | 1. 总后油料研究所,北京,102300
2. 石油化工科学研究院,北京,100083 |
| |
| 摘 要： | 为了提高近红外光谱分析精度和速度,需要分别对近红外光谱扣除背景、降低噪音和优选变量等预处理,其过程稍嫌烦琐,本文提出用小波变换-偏最小二乘法(PLS)。该方法利用小波变换提取光谱变量,利用PLS将光谱变量和柴油性质进行关联,建立模型。利用该方法能够分析柴油的诸多性质,分析精度与传统处理方法(微分-PLS)基本一致。该方法具有预处理简单、优选参数少、建模变量少等特点,能够大大简化建模过程、提高建模和分析速度。
|
| 关 键 词： | 近红外光谱 小波变换 偏最小二乘法 柴油 性质分析 |
| 文章编号： | 1001-4160（2006）10-971-974 |
| 收稿时间： | 2005-10-18 |
| 修稿时间： | 2005-10-182006-01-06 |
Near infrared spectra analysis of diesel by wavelet transform combined with partial least square regression method |
| |
| Tian Gaoyou,Chu Xiaoli,Yuan Hongfu.Near infrared spectra analysis of diesel by wavelet transform combined with partial least square regression method[J].Computers and Applied Chemistry,2006,23(10):971-974. |
| |
| Authors: | Tian Gaoyou Chu Xiaoli Yuan Hongfu |
| |
| Affiliation: | 1. Beijing POL Institute, Beijing, 102300, China; 2. Research Institute of Petroleum Processing, Beijing, 100083, China |
| |
| Abstract: | In order to improve analysis precision and analytical speed,redundant preproeessings were conducted traditionally to sub- tract background,denoise and select variables.In this paper,WT-PLS (Wavelet Transform combined with Partial Least Square Re- gression) method was proposed.WT was firstly conducted to extract variables and then PLS was applied to relate variables with proper- ties of diesel.This method can be used to analyze many properties of diesel with the same precision as that from traditional derivative algorithm-PLS.This method has many advantages,such as simple processing,few parameters to optimize,small size of variables, which has greatly simplify model establishment and reduce the time of building model and analyzing. |
| |
| Keywords: | wavelets transform(WT) NIR diesel partial least square regression |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
