| 模式识别在火灾调查中的汽油分类问题的应用研究 |
| |
| 引用本文： | 支有冉,宗若雯,王荣辉,李松阳.模式识别在火灾调查中的汽油分类问题的应用研究[J].火灾科学,2009,18(2):108-114. |
| |
| 作者姓名： | 支有冉 宗若雯 王荣辉 李松阳 |
| |
| 作者单位： | 1. 中国科学技术大学,火灾科学国家重点实验室,安徽,合肥,230026;中国科学技术大学苏州研究院,苏州市城市公共安全重点实验室,江苏,苏州,215123
2. 北京市消防局,北京,102308 |
| |
| 摘 要： | 在火灾调查中,检测汽油成分并对其进行正确分类尤为重要.运用GC-MS对90#和93#两种普通汽油的共50个样本进行检测,所得的GC-MS原始数据通过PCA方法进行处理,以提取有用信息,避免冗余变量进入后续计算.在此基础上应用KNN方法对这两种汽油助燃剂进行分类.结果表明, KNN方法对这两种汽油的分类准确率达到100%,且当初始数据未经标准化预处理时也能达到同样准确的分类效果.研究表明:将模式识别方法正确地运用到助燃剂鉴定和分类工作中有助于火灾调查.
|
| 关 键 词： | 助燃剂 主成分分析 模式识别 |
| 收稿时间： | 2009/2/17 0:00:00 |
| 修稿时间： | 2009/3/12 0:00:00 |
The application of pattern recognition to the classification of regular gasoline in fire investigation |
| |
| ZHI You-ran,ZONG Ruo-wen,WANG Rong-hui and LI Song-yang.The application of pattern recognition to the classification of regular gasoline in fire investigation[J].Fire Safety Science,2009,18(2):108-114. |
| |
| Authors: | ZHI You-ran ZONG Ruo-wen WANG Rong-hui and LI Song-yang |
| |
| Institution: | 1.State Key Laboratory of Fire Science;University of Science and Technology of China;Hefei;Anhui;230026 China;2.Suzhou Key Laboratory of Urban Public Safety;Suzhou Institute for Advanced Study;Suzhou;Jiangsu;215123 China;3.Beijing Fire Bureau;Beijing;102308 China |
| |
| Abstract: | Detection and accurate classification of gasoline is very important in fire investigation.In this paper,a total of 50 samples of regular gasoline,covering two different grades(90# and 93 #),were examined by gas chromatography-mass spectrometry(GC-MS).The GC-MS data were treated by Principal Component Analysis(PCA) to distill the information from the original dataset in order to avoid the redundant variables to be calculated.And k-nearest neighbors algorithm(KNN) was further applied to classify the two types... |
| |
| Keywords: | KNN GC-MS Accelerant PCA KNN GC-MS Pattern recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《火灾科学》浏览原始摘要信息 |
| 点击此处可从《火灾科学》下载免费的PDF全文 |
|
