Acta Metrologica Sinica  2023, Vol. 44 Issue (4): 622-629    DOI: 10.3969/j.issn.1000-1158.2023.04.20
Current Issue | Archive | Adv Search |
An Adaptive Lorentz Peak Fitting Method Based on GA-NSVM to Identify Overlapping Oeaks of Infrared Spectrum Absorption
LI Zhong-bing1,PANG Wei1,2,LIANG Hai-bo3,JIANG Chuan-dong1,DUAN Hong-ming1,LUO Yi4
1. School of Electrical and Information Engineering, Southwest Petroleum University, Chengdu, Sichuan 610500, China
2. Engineering Technology Research Institute, CNPC West Drilling Engineering Co. LTD, Urumqi, Xinjiang 830026, China
3. School of Mechanical and Electrical Engineering, Southwest Petroleum University, Chengdu, Sichuan 610500, China
4. Development Company of Xinjiang Oilfield Company, PetroChina, Karamay, Xinjiang 834000, China
Download: PDF (84472 KB)   HTML (1 KB) 
Export: BibTeX | EndNote (RIS)      
Abstract  A genetic algorithm-nonlinear support vector machine (GA-NSVM)-based adaptive Lorentz peak fitting method for infrared spectral overlapping peak identification is proposed. Based on the fundamental difference of the characteristic absorption line patterns of the elemental substance, the mixture spectrum is decomposed into multiple Lorentz single peaks, and the nonlinear support vector machine is used to perform multi-class screening on the multiple fitted single peaks, so as to determine absorption peaks for specific components. The feasibility of this method in the spectral identification and classification of highly similar molecular structures was demonstrated in the collected infrared spectrum data set of 400 alkane mixture samples. The experiment results show that the proposed method can effectively separate the infrared absorption peaks of methane, ethane, and propane in alkanes with good accuracy and robustness, and a stronger parameters interpretation ability. It can accelerate the application of spectral detection technology in the fields of biopharmaceuticals, food chemical industry, oil and gas exploration, etc., especially in analysis and applications containing homologous organic compounds.
Key wordsmetrology      IR spectrum      overlapping oeaks of absorption      genetic algorithm      NSVM      peak fitting     
Received: 03 December 2021      Published: 18 April 2023
PACS:  TB99  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
LI Zhong-bing
PANG Wei
LIANG Hai-bo
JIANG Chuan-dong
DUAN Hong-ming
LUO Yi
Cite this article:   
LI Zhong-bing,PANG Wei,LIANG Hai-bo, et al. An Adaptive Lorentz Peak Fitting Method Based on GA-NSVM to Identify Overlapping Oeaks of Infrared Spectrum Absorption[J]. Acta Metrologica Sinica, 2023, 44(4): 622-629.
URL:  
http://jlxb.china-csm.org:81/Jwk_jlxb/EN/10.3969/j.issn.1000-1158.2023.04.20     OR     http://jlxb.china-csm.org:81/Jwk_jlxb/EN/Y2023/V44/I4/622
Copyright © Editorial Board of Acta Metrologica Sinica
Supported by:Beijing Magtech