Spectral wavelength selection and dimension reduction using Elastic Net in spectroscopy analysis

Zhao Anxin; Tang Xiaojun; Song Ya; Zhang Zhonghua; Liu Junhua

In the use of Fourier transform infrared spectroscopy to build the multi-component gases quantitative analysis model, it is necessary to reduce the dimensions and select characteristics wavelength according to the target gas spectral. Through the regularization algorithm analysis, least absolute shrinkage and selection operator (LASSO) and Elastic Net method were used to do these for seven kinds of mixed gases of methane, ethane, propane, iso-butane, n-butane, iso-pentane and n-pentane. The minimum mean square error (MSE) and prediction deviation were used as the criteria to select LASSO and Elastic Net parameters. Finally, the resolution of 4cm-1 measured spectral data was analyzed. The dimension of spectra were reduced from 2 542 d to 2d and 3d respectively by using LASSO and Elastic Net method under the condition of the MSE of 0.001 9 and 0.002 1. The cross sensitivity of maximum and minimum were 10.271 8% and 1.420 5% by LASSO method. The cross sensitivity of maximum and minimum were 5.494 5% and 0.749 3% by Elastic Net. Results show that the Elastic Net method was better in the characteristic variable selection and the spectral dimension reduction for gas spectral quantitative analysis，and it was foundation to establish the accurate quantitative analysis model.

1. Xi'an University of Science and Technology,Xi'an 710054,China;

2. State Key Laboratory of Electrical Insulation and Power Equipment Xi'an Jiaotong University,Xi'an 710049,China;

3. AVIC Xi'an Aircraft Industry group Company Ltd,Xi'an 710089,China;

4. National Institute of Metrology,Beijing 100013,China

Received Date: 2013-10-12

Rev Recd Date: 2013-11-15

Publish Date: 2014-06-25

Key words:

gases infrared spectroscopy quantitative analysis /
regularization algorithm /
characteristic wavelength selection /
LASSO /
Elastic Net

Abstract: In the use of Fourier transform infrared spectroscopy to build the multi-component gases quantitative analysis model, it is necessary to reduce the dimensions and select characteristics wavelength according to the target gas spectral. Through the regularization algorithm analysis, least absolute shrinkage and selection operator (LASSO) and Elastic Net method were used to do these for seven kinds of mixed gases of methane, ethane, propane, iso-butane, n-butane, iso-pentane and n-pentane. The minimum mean square error (MSE) and prediction deviation were used as the criteria to select LASSO and Elastic Net parameters. Finally, the resolution of 4cm-1 measured spectral data was analyzed. The dimension of spectra were reduced from 2 542 d to 2d and 3d respectively by using LASSO and Elastic Net method under the condition of the MSE of 0.001 9 and 0.002 1. The cross sensitivity of maximum and minimum were 10.271 8% and 1.420 5% by LASSO method. The cross sensitivity of maximum and minimum were 5.494 5% and 0.749 3% by Elastic Net. Results show that the Elastic Net method was better in the characteristic variable selection and the spectral dimension reduction for gas spectral quantitative analysis，and it was foundation to establish the accurate quantitative analysis model.

HTML

Reference (23)

[1]
[2]	Materazzi S, Vecchio S. Recent applications of evolved gas anal ysis by infrared spectroscopy (IR-EGA)[J]. Applied Spectroscopy Reviews, 2013, 48(8): 654-689.
[3]
[4]	Sepman A V, den Blanken R, Schepers R, et al. Quantitative fourier transform infrared diagnostics of the gas-phase composition using the HITRAN database and the equivalent width of the spectral features[J]. Appl Spectrosc, 2009, 63(11): 1211-1222.
[5]	Xu Xiaojing, Huang Wei. Application of spectral imaging in forensic science[J]. Infrared and Laser Engineering, 2012, 41(12): 3280-3284. (in Chinese) 许小京，黄威. 光谱成像技术在物证鉴定领域的应用[J].红外与激光工程, 2012, 41(12): 3280-3284.
[6]
[7]
[8]	Kalivas J H. Multivariate calibration, an overview[J]. Analytical Letters, 2005, 38(14): 2259-2279.
[9]
[10]	Kunz M R, Ottaway J, Kalivas J H, et al. Impact of standardization sample design on Tikhonov regularization variants for spectroscopic calibration maintenance and transfer[J]. Journal of Chemometrics, 2010, 24(3-4SI): 218-229.
[11]	Zeng T, Wen Z, Wen Z, et al. Weighted fusion of multiple models for wavelength selection[J]. Appl Spectrosc, 2013, 67 (7): 718-723.
[12]
[13]
[14]	Zou H, Hastie T. Regularization and variable selection via the elastic net[J]. Journal of the Royal Statistical Society Series B-statistical Methodology, 2005, 67(Part 2): 301-320.
[15]	Tang Xiaojun, Zhang Lei, Wang Erzhen, et al. An improved characteristic spectral selection method for multicomponent gas quantitative analysis based on tikhonov regularization[J]. Spectroscopy and Spectral Analysis, 2012, 32(10): 2730-2734. (in Chinese) 汤晓君，张蕾，王尔珍，等. 一种改进型多组分气体的Tikhonov 正则化特征光谱提取方法[J]. 光谱学与光谱分析, 2012, 32(10): 2730-2734.
[16]
[17]
[18]	Wang Gaofeng, Zhao Yiqiang, Yang Dong. Data acquisition of 1 024-pixel long linear infrared detectors[J]. Infrared and Laser Engineering, 2012, 41(8): 1990-1994. (in Chinese) 王高峰, 赵毅强, 杨栋. 1 024 元长线列红外探测器的数据采集技术[J]. 红外与激光工程, 2012, 41(8): 1990-1994.
[19]
[20]	Friedman J H, Hastie T, Tibshirani R. Regularization paths for generalized linear models via coordinate descent[J]. Journal of Statistical Software, 2010, 33(1): 1-22.
[21]
[22]	Dyar M D, Carmosino M L, Breves E A, et al. Comparison of partial least squares and lasso regression techniques as applied to laser-induced breakdown spectroscopy of geological samples[J]. Spectrochimica Acta Part B-atomic Spectroscopy, 2012, 70: 51-67.
[23]	Tang Xiaojun, Wang Jin, Zhang Lei, et al. Spectral baseline correction by piecewise dividing in fourier transform infrared gas analysis[J]. Spectroscopy and Spectral Analysis, 2013, 33(2): 334-339. (in Chinese) 汤晓君, 王进, 张蕾, 等. 气体光谱分析应用中傅里叶变换红外光谱基线漂移分段比校正方法[J]. 光谱学与光谱分析, 2013, 33(2): 334-339.

Spectral wavelength selection and dimension reduction using Elastic Net in spectroscopy analysis

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Related

Proportional views