气溶胶单粒子光谱的PLS聚类分析

J4 ›› 2012, Vol. 29 ›› Issue (1): 106-113.

气溶胶单粒子光谱的PLS聚类分析

张子良,提汝芳,黄腾,王颖萍,丁蕾,郑海洋,方黎

中国科学院安徽光学精密机械研究所环境光谱学研究室，安徽合肥 230031

收稿日期:2011-02-17 修回日期:2011-03-15 出版日期:2012-01-28 发布日期:2012-01-28
通讯作者: 方黎（1964-），研究员，博士生导师. E-mail:fangli@aiofm.ac.cn
作者简介:张子良（1986-）研究生，主要研究方向为光谱数据处理，E-mail: ashuang@mail.ustc.edu.cn
基金资助:
国家自然科学基金资助（20677056）

PLS cluster analysis of Individual particle spectrum

ZHANG Ziliang, TI Rufang, HUANG Teng, WANG Yingping, DING Lei, ZHENG Haiyang, FANG Li

Laboratory of Environmental Spectroscopy, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China

Received:2011-02-17 Revised:2011-03-15 Published:2012-01-28 Online:2012-01-28

摘要/Abstract

摘要：

PLS聚类法是一种全新的气溶胶单粒子光谱数据处理方法，是利用具有“自组织机制”的PLS(Partial Least Square)回归算法去完成数据的聚类。文中首先阐述了PLS聚类对模拟数据集的运用以展示这种方法的一般特征及有效性，然后应用到气溶胶激光飞行时间质谱数据以展示PLS聚类的正确性及成功运用，最后我们将PLS聚类应用到氯化钙、氯化镁、氯化钠及氯化钾四种气溶胶单粒子激光击穿光谱混合数据集，通过分析聚类获得的树形图和图中节点的统计特性，我们剖析了正确聚类及发生错误划分的原因，表明了PLS聚类方法在气溶胶单粒子谱分析方面的应用潜力。

关键词: 光谱学, 偏最小二乘回归, 聚类分析, 气溶胶单粒子

Abstract:

PLS(Partial Least Squares) cluster analysis, based on PLS regression procedure with a self-organizing mechanism, is a novel data analysis method of spectrum and is. PLS Cluster was first used to classify simulated dataset, and present the general properties and validity. Then it was used to aerosol laser time-of-flight mass spectra data, showing correctness and successful application in practice. Finally, this mothod was applied to mixed spectra dataset of aerosol particles of CaCl2, MgCl2, NaCl and KCl. By examing the statistical properties of obtained dendrogram plots and nodes, we got the reason of clustering and misclassifying. It demonstrates the potency of application of PLS_Cluster in individual aerosol particle spectra.

Key words: spectroscopy, PLS-regression, cluster analysis, individual aerosol particle

中图分类号:

O433

张子良提汝芳黄腾王颖萍丁蕾郑海洋方黎. 气溶胶单粒子光谱的PLS聚类分析[J]. J4, 2012, 29(1): 106-113.

ZHANG Ziliang, TI Rufang, HUANG Teng, WANG Yingping, DING Lei, ZHENG Haiyang, FANG Li. PLS cluster analysis of Individual particle spectrum[J]. J4, 2012, 29(1): 106-113.

参考文献

[1] National Research Council, Research priorities for airborne particulate matter-immediate priorities and a long-range research portfolio, Washington, D.C.: National Academy Press, 1998, Chapter 1.
[2] Wang Mingxing. Atmospheric Chemistry(大气化学), 2nd Edition, Beijing: Weather Press, 1999(in Chinese).
[3] Suess D T, Prather K A. Mass Spectrometry of Aerosols. Chem. Rev., 1999, 99: 3007-3035.
[4] Xia Zhuhong, Feng Sumin, Fang Li, et al. Advances of Atmospheric Aerosol Mass Spectrometry. Chinese Journal of Quantum Electronics(量子电子学报), 2003, 20(2):147-151(in Chinese).
[5] Han Jiawei, Kamber Micheline. Data Mining:Concepts and Techniques(数据挖掘：概念与技术), 2nd Edition, Beijing: China Machine Press, 2007, Chapter 7(in Chinese).
[6] Xhoffer C, Bernard P, Van G R, et al. Chemical characterization and source apportionment of individual aerosol particles over the North Sea and the English Channel using multivariate techniques. Environ. Sci. Technol., 1991, 25: 1470-1478.
[7] Bondarenko I, Treiger B, et al. IDAS: a Windows based software package for cluster analysis. Spectrochim. Acta, Part B, 1996, 51(4): 441-456.
[8] Guo Xiaoyong, Zhu Yuan, et al. Real-time data acquisition and processing of atmosphere aerosol time-of-flight laser mass spectrometry. Chinese Journal of Quantum Electronics(量子电子学报), 2006, 23(2): 217- 221(in Chinese).
[9] Hinz K P, Kaufmann R, et al. Simultaneous Detection of Positive and Negative Ions From Single Airborne Particles by Real-time Laser Mass Spectrometry. Aerosol Sci. Technol., 1996, 24: 233-242.
[10] Kohonen T. The Self-Organizing Map. Proceedings of the Ieee, 1990, 78: 1464-1480.
[11] Kohonen T. Self-Organizing Maps, 2nd Edition, Berlin: Springer, 1997.
[12] Lin Ying, Guo Xiaoyong, et al. Data Analysis of Laser Desorption /Ionization Mass Spectrum of Individual Particle Using Adaptive Resonance Theory Based Neural Network. Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2009, 29(3): 580-584(in Chinese).
[13] Murphy D M, Middlebrook A M, et al. Cluster analysis of data from the Particle Analysis by Laser Mass Spectrometry (PALMS) instrument. Aerosol Sci. Technol., 2003, 37(4): 382-391.
[14] Barros A S, Rutledge D N, PLS_Cluster: a novel technique for cluster analysis. Chemometr. Intell. Lab., 2004, 70: 99-112.
[15] Wang Huiwen, Partial Least-Squares Regression Method and Application(偏最小二乘回归方法及其应用), Beijing: National Defense Industry Press, 1999(in Chinese).
[16] Lin Sheng, Bai Yin, Improvement of Hierarchical Cluster analysis method based on Partial Least Squares. Statistics and Decision(统计与决策), 2006, 14: 16-18(in Chinese).
[17] Helland I S. Some theoretical aspects of partial least squares regression. Chemometr. Intell. Lab., 2001, 58: 97-107.
[18] Jong S D, SIMPLS:An Alternative Approach to Partial Least Squares Regression. Chemometr. Intell. Lab., 1993, 18: 251-263.
[19] Ding Huilin, Gao Li-xin, et al. Experimental Investigation of Atmosphere and Water Vapor with Laser Induced Breakdown Spectroscopy. Spectroscopy and Spectral Analysis(光谱学与光谱分析),2010, 30(1):1-5(in Chinese).