SVM Analysis Method for Infrared Spectra of Mixed Gas

Peng Bai; Yan Li; Peng Liu

doi:10.4028/www.scientific.net/AMM.58-60.1681

Paper Titles

A Research on Circular Localization Algorithm of Wireless Sensor Network
p.1657

Using Formal Concept Analysis and Rough Set Constructing the System of Semantic Web
p.1664

An Algorithm of Constraint Frequent Neighboring Class Set Mining Based on Interval Mapping
p.1671

An Algorithm of Constraint Frequent Neighboring Class Set Mining Based on Filling Class Set
p.1676

SVM Analysis Method for Infrared Spectra of Mixed Gas
p.1681

Improved Tabu Search for Solving Order Delivery Problem under B2C Electronic Commerce
p.1685

Validation and Verification of Life Prediction Technology for Electromechanical Products
p.1690

Research and Application of a FPGA-Based Dynamic Distributed Algorithm
p.1696

The Study of Port Logistics Capabilities Based on Random Forest
p.1701

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 58-60SVM Analysis Method for Infrared Spectra of Mixed...

SVM Analysis Method for Infrared Spectra of Mixed Gas

Abstract:

In order to solving the problem that the mass samples of mixed gas spectra data samples being unable to obtain, characteristic absorption spectrum line of the component gas for mixed gas being overlap, and the problem of randomness of component concentration distribution for mixed gas and so on, support vector machine is introduced for the infrared spectra analysis for the mixed gas. Key technologies as feature selection of spectra data samples, data preprocessing, SVM calibration model parameters optimization and level structure for spectrum analysis of a mixed gas is proposed in the paper. The influence of above-mentioned four key technologies to the analysis results is discussed by using experimental means. The experimental result shows that with adoption of the key technologies, the maximum absolute error of component concentration analysis for the mixed gas is 2.93%, and the maximum average absolute error is of 0.73%. The method can also be used for infrared spectra analysis for other mixed gas, and it has practical application value.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 58-60)

Pages:

1681-1684

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.58-60.1681

Citation:

Cite this paper

Online since:

June 2011

Authors:

Peng Bai, Yan Li, Peng Liu

Keywords:

Analysis, Infrared, Model, Spectra, Support Vector Machine (SVM)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] XU Guang-tong, YUAN Hong-fu, LU Wan-Zhen. Development of modern near infrared spectroscopic techniques and its applications. Spectroscopy and Spectral Analysis, 2000, 20(2): 134-135.

Google Scholar

[2] V.N. Vapnik. The Nature of Statistical Learning[M]. New York: Springer, (1995).

Google Scholar

[3] Chih-Chung Chang, Chih-Jen Lin. LIBSVM. a library for support vector machines. Software available at http: /www. csie. ntu. edu. tw/~cjlin/libsvm, (2001).

Google Scholar

[4] LIN Ji-peng, LIU Jun-hua. A New Technology Study on Overlapped Absorbed Peak of Infrared Hybrid Gas[J]. Acta Photonica Sinic, 2006, 35 (3): 408-412.

Google Scholar

[5] XIONG Yu-hong, WEN Zhi-yu, CHEN Gang, et a1. Spectral Multicomponent Analysis Based on Wavelet Transform and Support Vector Machine[J]. Acta Photonica Sinic, 2005, 34(10): 1514-1517.

Google Scholar

[6] CHU Xiao-li, YUAN Hong-fu, LU Wan-Zhen. Progress and Application of Spectral Data Pretreatment and Wavelength Selection Methods in NIR Analytical Technique[J]. Progress in Chemistry, 2004, 16(4): 528-542.

Google Scholar