光谱学与光谱分析 |
|
|
|
|
|
Quantitative Analysis of Multi-Component Gas Mixture Based on KPCA and SVR |
HAO Hui-min1,2,TANG Xiao-jun1,BAI Peng1,3,LIU Jun-hua1,ZHU Chang-chun1 |
1. School of Electrical Engineering, Xi’an Jiaotong University, Xi’an 710049, China 2. Taiyuan Iron and Steel Co., Automatic Company, Taiyuan 030003, China 3. Engineering Institute, Air Force Engineering University, Xi’an 710038, China |
|
|
Abstract In the present paper, the authors present a new quantitative analysis method of mid-infrared spectrum. The method combines the kernel principal component analysis (KPCA) technique with support vector regress machine (SVR) to createa quantitative analysis model of multi-component gas mixtures. Firstly,the spectra of multi-component gas mixtures samples were mapped nonlinearly into a high-dimensional feature space through the use of Gaussian kernels. And then, PCA technique was employed to compute efficiently the principal components in the high-dimensional feature spaces. After determining the optimal numbers of principal components, the extracted features (principal components) were used as the inputs of SVR to create the quantitative analysis model of seven-component gas mixtures. The prediction RMSE (φ×10-6) of seven-component gases of prediction set samples by use of KPCA-SVR model were respectively 124.37, 72.44, 136.51, 87.29, 153.01, 57.12, and 81.72, ten times less than that by use of SVR model. The elapsed time of modeling and prediction by using KPCA-SVR were respectively 46.59 (s) and 4.94 (s), which was consumedly less than 752.52 (s) and 26.21 (s) by using only SVR. These results show that KPCA has an excellent ability of nonlinear feature extraction. It can make the most of the information of entire spectra range and effectively reduce noise and the dimension of the spectra. The KPCA combined with SVR can improve the model’s analysis precision and cut the elapsed time of modeling and analysis. From our research and experiments, we conclude that KPCA-SVR is an effective new method for infrared spectroscopic quantitative analysis.
|
Received: 2007-08-22
Accepted: 2007-11-28
|
|
Corresponding Authors:
HAO Hui-min
E-mail: helenwangmin@gmail.com
|
|
[1] ZHANG Lin, ZHANG Li-ming, LI Yan, et al(张 琳,张黎明,李 燕,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2006, 26(4): 620. [2] LI Hong-lei, LIU Xian-yong, ZHOU Fang-jie, et al. Proceedings of the SPIE, 2005, 5640: 692. [3] SUN Xiu-yun, LI Yan, WANG Jun-de(孙秀云,李 燕,王俊德). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2003, 23(4): 739. [4] Evans Wayne F J, Puckrin Eldon, McMaster D. Proceedings of the SPIE, 2002, 4574;44. [5] Bacsik Z, McGregor J, Mink J. Food and Chemical Toxicology, 2007, 45(2): 266. [6] DENG Nai-yang, TIAN Ying-jie(邓乃扬,田英杰). New Method of Data Mining—Support Vector Machine(数据挖掘中的新方法——支持向量机). Beijing: Science Press(北京:科学出版社), 2004. [7] Hoegaerts L, Suykens J A K, Vandewalle J, et al. Neurocomputing, 2005, 63: 293. [8] Vapnik V N. The Nature of Statistical Learning Theory. New York: Springer-Verlag, 1995. [9] Vapnik V N. Statistical Learning Theory, New York: Wiley, 1998. [10] Duan Kaibo, Keerthi S Sathiya, Poo Aun Neow. Neurocomputing, 2003, 51: 41. [11] BAI Peng, LIU Jun-hua(白 鹏,刘君华). Control and Instruments in Chemical Industry(化工自动化及仪表), 2005, 32(5): 47. [12] LIN Ji-peng, LIU Jun-hua(林继鹏,刘君华). Journal of Xi’an Jiaotong University(西安交通大学学报), 2005, 39(6): 586. [13] Kim K L, Franz M O, Schlkopf B. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27: 1351. [14] Rosipal Roman, Girolami Mark, Trejo Leonard J, et al. Neural Computing & Applications, 2001, 10(3): 231. [15] Müller Klaus-Robert, Mika Sebastian, Ratsch Gunnar, et al. IEEE Transactions on Neural Networks, 2001, 12(2): 181. [16] Schlkopf B, Smola A,Müller Klaus-Robert. Neural Computation, 1998, 10(5): 1299. [17] John Shawe-Taylor, Nello Cristianini. Kernel Methods for Pattern Analysis. Cambridge: Cambridge University Press, 2004. 143. [18] Mika Sebastian, Schlkopf Bernhard, Smola Alex, et al. Advances in Neural Information Processing System Ⅱ. Cambridge, MA: MIT Press, 1999. 536. [19] Mika S. Advances in Neural Information Processing System Ⅱ. Cambridge, MA: MIT Press, 1999. 536. [20] Schlkopf B, Smola A. Advances in Kernel Methods-Support Vector Learning. Cambridge, MA: MIT Press, 1999. 327.
|
[1] |
WANG Gan-lin1, LIU Qian1, LI Ding-ming1, YANG Su-liang1*, TIAN Guo-xin1, 2*. Quantitative Analysis of NO-3,SO2-4,ClO-4 With Water as Internal Standard by Raman Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1855-1861. |
[2] |
YU Zhi-rong, HONG Ming-jian*. Near-Infrared Spectral Quantitative Analysis Network Based on Grouped Fully Connection[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1735-1740. |
[3] |
MA Fang1, HUANG An-min2, ZHANG Qiu-hui1*. Discrimination of Four Black Heartwoods Using FTIR Spectroscopy and
Clustering Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1915-1921. |
[4] |
ZHANG Dian-kai1, LI Yan-hong1*, ZI Chang-yu1, ZHANG Yuan-qin1, YANG Rong1, TIAN Guo-cai2, ZHAO Wen-bo1. Molecular Structure and Molecular Simulation of Eshan Lignite[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1293-1298. |
[5] |
SUN Zhi-xing, ZHAO Zhong-gai*, LIU Fei. Near-Infrared Spectral Modeling Based on Stacked Supervised Auto-Encoder[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 749-756. |
[6] |
ZHOU Jun1, 2, YANG Yang2, YAO Yao2, LI Zi-wen3, WANG Jian3, HOU Chang-jun1*. Application of Mid-Infrared Spectroscopy in the Analysis of Key Indexes of Strong Flavour Chinese Spirits Base Liquor[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 764-768. |
[7] |
WANG Fang-fang1, ZHANG Xiao-dong1, 2*, PING Xiao-duo1, ZHANG Shuo1, LIU Xiao1, 2. Effect of Acidification Pretreatment on the Composition and Structure of Soluble Organic Matter in Coking Coal[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 896-903. |
[8] |
HU Chao-shuai1, XU Yun-liang1, CHU Hong-yu1, CHENG Jun-xia1, GAO Li-juan1, ZHU Ya-ming1, 2*, ZHAO Xue-fei1, 2*. FTIR Analysis of the Correlation Between the Pyrolysis Characteristics and Molecular Structure of Ultrasonic Extraction Derived From Mid-Temperature Pitch[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 889-895. |
[9] |
YANG Jiong1, 2, QIU Zhi-li1, 4*, SUN Bo3, GU Xian-zi5, ZHANG Yue-feng1, GAO Ming-kui3, BAI Dong-zhou1, CHEN Ming-jia1. Nondestructive Testing and Origin Traceability of Serpentine Jade From Dawenkou Culture Based on p-FTIR and p-XRF[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(02): 446-453. |
[10] |
LI Yuan1, 2, SHI Yao2*, LI Shao-yuan1*, HE Ming-xing3, ZHANG Chen-mu2, LI Qiang2, LI Hui-quan2, 4. Accurate Quantitative Analysis of Valuable Components in Zinc Leaching Residue Based on XRF and RBF Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(02): 490-497. |
[11] |
FU Juan, MO Jia-mei, YU Yi-song, ZHANG Qing-zong, CHEN Xiao-li, CHEN Pei-li, ZHANG Shao-hong, SU Qiu-cheng*. Experimental Study on the Structure Characteristics of CO2 in Gas Hydrate by Solid-State Nuclear Magnetic Resonance and Raman Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(02): 464-469. |
[12] |
YU Feng-ping1, LIN Jing-jun1*, LIN Xiao-mei1, 3*, LI Lei1,2*. Detection of C Element in Alloy Steel by Double Pulse Laser Induced Breakdown Spectroscopy With a Multivariable GA-BP-ANN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(01): 197-202. |
[13] |
WANG Xiao-bin1, 2, 3, ZHANG Xi1, GUAN Chen-zhi1, HONG Hua-xiu1, HUANG Shuang-gen2*, ZHAO Chun-jiang3. Quantitative Detection of Ascorbic Acid Additive in Flour Based on Raman Imaging Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(12): 3765-3770. |
[14] |
HE Xiong-fei1, 2, HUANG Wei3, TANG Gang3, ZHANG Hao3*. Mechanism Investigation of Cement-Based Permeable Crystalline Waterproof Material Based on Spectral Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(12): 3909-3914. |
[15] |
LIU Jia-cheng1, 2, HU Bing-liang1, YU Tao1*, WANG Xue-ji1, DU Jian1, LIU Hong1, LIU Xiao1, HUANG Qi-xing3. Nonlinear Full-Spectrum Quantitative Analysis Algorithm of Complex Water Based on IERT[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(12): 3922-3930. |
|
|
|
|