Modeling of oil near-infrared spectroscopy based on similarity and transfer learning algorithm

doi:10.1007/s11705-019-1807-2

Front. Chem. Sci. Eng.

2019, Vol. 13

Issue (3) : 599-607 https://doi.org/10.1007/s11705-019-1807-2

RESEARCH ARTICLE

Modeling of oil near-infrared spectroscopy based on similarity and transfer learning algorithm

Yifei Wang^1,², Kai Wang^1,², Zhao Zhou^1,²(

), Wenli Du^1,²(

)

¹. Key Laboratory of Advanced Control and Optimization for Chemical Processes (Ministry of Education), East China University of Science and Technology, Shanghai 200237, China
². School of information science and engineering, East China University of Science and Technology, Shanghai 200237, China

Download: PDF(1118 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Near-infrared spectroscopy mainly reflects the frequency-doubled and total-frequency absorption information of hydrogen-containing groups (O‒H, C‒H, N‒H, S‒H) in organic molecules for near-infrared lights with different wavelengths, so it is applicable to testing of most raw materials and products in the field of petrochemicals. However, the modeling process needs to collect a large number of laboratory analysis data. There are many oil sources in China, and oil properties change frequently. Modeling of each raw material is not only unfeasible but also will affect its engineering application efficiency. In order to achieve rapid modeling of near-infrared spectroscopy and based on historical data of different crude oils under different detection conditions, this paper discusses about the feasibility of the application of transfer learning algorithm and makes it possible that transfer learning can assist in rapid modeling using certain historical data under similar distributions under a small quantity of new data. In consideration of the requirement of transfer learning for certain similarity of different datasets, a transfer learning method based on local similarity feature selection is proposed. The simulation verification of spectral data of 13 crude oils measured by three different probe detection methods is performed. The effectiveness and application scope of the transfer modeling method under different similarity conditions are analyzed.

Keywords near-infrared spectroscopy transfer learning similarity modeling

Corresponding Author(s): Zhao Zhou,Wenli Du

Just Accepted Date: 18 March 2019 Online First Date: 22 April 2019 Issue Date: 22 August 2019

Cite this article:

Yifei Wang,Kai Wang,Zhao Zhou, et al. Modeling of oil near-infrared spectroscopy based on similarity and transfer learning algorithm[J]. Front. Chem. Sci. Eng., 2019, 13(3): 599-607.

URL:

https://academic.hep.com.cn/fcse/EN/10.1007/s11705-019-1807-2
https://academic.hep.com.cn/fcse/EN/Y2019/V13/I3/599

Fig.1 The spectrum data measured by the transmittance probe, reflectance probe and transreflectance probe. (a) Transmittance spectrum, (b) reflectance spectrum, (c) transreflectance spectrum

Dataset	$cos ? θ$	$\| ρ X Y \|$
TF-TM	0.4237	0.4503
TF-RF	0.3593	0.3680

Tab.1 Similarity between two data sets

No.	$cos ? θ$	$\| ρ X Y \|$
1	0.0895	0.1378
2	?0.1330	0.4379
3	0.4081	0.2376
4	?0.0987	0.2717
5	0.3281	0.2366
6	0.0144	0.2303
7	?0.1757	0.2398
8	0.0170	0.3421
9	0.0382	0.5063
10	0.6067	0.5993
11	0.3612	0.3563
12	0.9569	0.9838
13	0.5610	0.7221

Tab.2 Local similarity of transreflectance and transmittance probe data set

No.	$cos ? θ$	$\| ρ X Y \|$
1	0.1278	0.1466
2	0.1325	0.2183
3	?0.0627	0.2014
4	?0.1562	0.1974
5	?0.3917	0.2048
6	0.0132	0.2562
7	?0.5256	0.2534
8	0.2285	0.2522
9	0.0404	0.4213
10	0.8141	0.7394
11	0.1341	0.2823
12	0.7107	0.7553
13	0.4245	0.5454

Tab.3 Local similarity of transreflectance and reflectance probe data set

Fig.2 Local wavenumber extraction. (a) Transmittance spectrum, (b) Transreflectance spectrum

Fig.3 A flowchart of STBB

Tab.4 MAPE of BP1, BP2, TrAdaBoost, TCA and PCA/%

Fig.4 Error rate curves on transreflectance and transmittance probe data set for BP1, BP2, TrAdaBoost, TCA and PCA/%

Fig.5 Error rate curves on transreflectance and reflectance probe data set for BP1, BP2, TrAdaBoost, TCA and PCA/%

Dataset	Selection range of feature data	$cos ? θ$	$\| ρ X Y \|$
TF-TM	410?489	0.7027	0.7337
TF-RF	321?400	0.7100	0.7479

Tab.5 Similarity of part with high similarity between source data set and target data set

Fig.6 The extracted part of the transreflectance and transmittance probe data. (a) Transreflectance spectrum, (b) Transmittance spectrum

Fig.7 The extracted part of the transreflectance probe and reflectance probe data. (a) Transreflectance spectrum, (b) Reflectance spectrum

Fig.8 Error rate curves on transreflectance and transmittance probe data set for BP1, S-TrAdaBoost, S-TCA and STBB/%

Fig.9 Error rate curves on transreflectance and reflectance probe data set for BP1, S-TrAdaBoost, S-TCA and STBB/%

Tab.6 MAPE of BP1, S-TrAdaBoost, S-TCA and STBB/%

1	Y L Yan. The Basis and Application of Near Infrared Spectroscopy. Beijing: China Light Industry Press, 2005, 286–564 (in Chinese)
2	W Z Lu. Modern Near Infrared Spectroscopy Analysis Technology. Beijing: China Petrochemical Press, 2007, 14–26 (in Chinese)
3	J Workman Jr. A brief review of near infrared in petroleum product analysis. Journal of Near Infrared Spectroscopy, 1996, 4(1): 69 https://doi.org/10.1255/jnirs.77
4	H Oja. Multivariate Linear Regression. New York: Springer, 2010, 183–200
5	N Tormod, M Harald. Principal component regression in NIR analysis: Viewpoint, background details and selection of components. Journal of Chemometrics, 1988, 2(2): 155–167 https://doi.org/10.1002/cem.1180020207
6	P Geladi, B R Kowalski. Partial least-squares regression: a tutorial. Analytica Chimica Acta, 1985, 185(86): 1–17
7	Y He, X Li, X Deng. Discrimination of varieties of tea using near infrared spectroscopy by principal component analysis and BP model. Journal of Food Engineering, 2007, 79(4): 1238–1242 https://doi.org/10.1016/j.jfoodeng.2006.04.042
8	H Shimodaira. Improving predictive inference under covariate shift by weighting the log-likelihood function. Journal of Statistical Planning and Inference, 2000, 90(2): 227–244 https://doi.org/10.1016/S0378-3758(00)00115-4
9	Y He. Modelling of near-infrared spectroscopy based on semi-supervised learning and transfer learning. Dissertation for the Doctor Degree. Shandong: Ocean University of China, 2012
10	S J Pan, Q Yang. A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345–1359 https://doi.org/10.1109/TKDE.2009.191
11	K Weiss, T M Khoshgoftaar, D D Wang. A survey of transfer learning. Journal of Big Data, 2016, 3(1): 9 https://doi.org/10.1186/s40537-016-0043-6
12	B Tan, Y Song, E Zhong, Q Yang. Transitive transfer learning. Acm Sigkdd International Conference on Knowledge Discovery & Data Mining, 2015, 1155–1164
13	B Tan, Y Zhang, S J Pan, Q Yang. Distant domain transfer learning. Association for the Advance of Artificial Intelligence, 2017, 2604–2610
14	J Gao. The application of near infrared spectroscopy in oil quality analysis. Dissertation for the Master Degree. Jiangsu: Nanjing Tech University, 2005, 11–12
15	T V Karstang, K Valheim. Multivariate prediction and background correction using local modeling and derivative spectroscopy. Analytical Chemistry, 1996, 63(8): 767–772 https://doi.org/10.1021/ac00008a006
16	C H Zhao, M H Tian, J W Li. Research progress on spectral similarity metrics. Journal of Harbin Engineering University, 2017, 38(8): 1179–1189 (in Chinese)
17	C Wang, M Gong, M Zhang, Y Chan. Unsupervised hyperspectral image band selection via column subset selection. IEEE Geoscience and Remote Sensing Letters, 2015, 12(7): 1411–1415 https://doi.org/10.1109/LGRS.2015.2404772
18	A Schlamm, D Messinger. Improved detection clustering of hyperspectral image date by preprocessing with a euclidean distance transformation. WHISPERS, 2011, 1(2): 1–4
19	Y Zhong, X Lin, L Zhang. A support vector conditional random fields classifier with a Mahalanobis distance boundary constraint for high spatial resolution remote sensing imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(4): 1314–1330 https://doi.org/10.1109/JSTARS.2013.2290296
20	F A Kruse, A B Lefkoff, J W Boardman, K B Heidebrecht, A T Shapiro, P J Barloon. The spectral image processing systems (SIPS)-interactive visualization and analysis of imaging spectrometer data. Aip Conference, 1993, 283(1): 192–201
21	C I Chang. Spectral information divergence for hyperspectral image analysis. IEEE International Geoscience & Remote Sensing Symposium, 1999, 509–511
22	S J Pan, J T Kwok, Q Yang, J J Pan. Adaptive localization in a dynamic WiFi environment through multi-view learning. Association for the Advance of Artificial Intelligence, 2007, 1108–1113
23	J C Granahan, J N Sweet. An evaluation of atmospheric correction techniques using the spectral similarity scale. IEEE International Geoscience & Remote Sensing Symposium, 2001, 2022–2024
24	S J Pan, I W Tsang, J T Kwok, Q Yang. Domain adaptation via transfer component analysis. IEEE Transactions on Neural Networks, 2011, 22(2): 199–210 https://doi.org/10.1109/TNN.2010.2091281
25	L Breiman. Bagging predictors. Machine Learning, 1996, 24(2): 123–140 https://doi.org/10.1007/BF00058655
26	W Y Dai, Q Yang, G R Xue, Y Yu. Boosting for transfer learning. International Conference on Machine Learning, Corvalis, 2007, 238(6): 193–200
27	Y Freund, R E Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 1997, 55(1): 119–139 https://doi.org/10.1006/jcss.1997.1504
28	S H Zhou, W L Du. Modeling of ethylene cracking furnace yields based on transfer learning. CIESC Journal, 2014, 65(12): 4921–4928

[1]	Kasra Pirzadeh, Ali Asghar Ghoreyshi, Mostafa Rahimnejad, Maedeh Mohammadi. Optimization of electrochemically synthesized Cu₃(BTC)₂ by Taguchi method for CO₂/N₂ separation and data validation through artificial neural network modeling[J]. Front. Chem. Sci. Eng., 2020, 14(2): 233-247.
[2]	Michael Bonitz, Alexey Filinov, Jan-Willem Abraham, Karsten Balzer, Hanno Kählert, Eckhard Pehlke, Franz X. Bronold, Matthias Pamperin, Markus Becker, Dettlef Loffhagen, Holger Fehske. Towards an integrated modeling of the plasma-solid interface[J]. Front. Chem. Sci. Eng., 2019, 13(2): 201-237.
[3]	Ismael Matino, Valentina Colla, Teresa A. Branca. Extension of pilot tests of cyanide elimination by ozone from blast furnace gas washing water through Aspen Plus^® based model[J]. Front. Chem. Sci. Eng., 2018, 12(4): 718-730.
[4]	Bo Chen, Yan Dai, Xuehua Ruan, Yuan Xi, Gaohong He. Integration of molecular dynamic simulation and free volume theory for modeling membrane VOC/gas separation[J]. Front. Chem. Sci. Eng., 2018, 12(2): 296-305.
[5]	Mehdi SEDIGHI,Kamyar KEYVANLOO. Kinetic study of the methanol to olefin process on a SAPO-34 catalyst[J]. Front. Chem. Sci. Eng., 2014, 8(3): 306-311.
[6]	Anton ALVAREZ-MAJMUTOV,Jinwen CHEN. Analyzing the energy intensity and greenhouse gas emission of Canadian oil sands crude upgrading through process modeling and simulation[J]. Front. Chem. Sci. Eng., 2014, 8(2): 212-218.
[7]	Zhikai LI, Zhangfeng QIN, Yagang ZHANG, Zhiwei WU, Hui WANG, Shuna LI, Mei DONG, Weibin FAN, Jianguo WANG. A logic-based controller for the mitigation of ventilation air methane in a catalytic flow reversal reactor[J]. Front Chem Sci Eng, 2013, 7(3): 347-356.
[8]	Swati Mukhopadhyay. Chemically reactive solute transfer in a boundary layer slip flow along a stretching cylinder[J]. Front Chem Sci Eng, 2011, 5(3): 385-391.
[9]	WANG Lijun, CHENG Youwei, WANG Qinbo, LI Xi. Progress in the research and development of p-xylene liquid phase oxidation process[J]. Front. Chem. Sci. Eng., 2007, 1(3): 317-326.

Viewed

Full text

Abstract

Cited

Shared

Discussed