基于光谱的天然气组分定量分析方法研究  

Study on the Method of Quantitative Analysis Method Based on Spectral-based Natural Gas Groups

作　　者：周德 张霖 何小宇 唐伟晔 ZHOU De;ZHANG Lin;HE Xiaoyu;TANG Weihua(School of Electrical Engineering,Chongqing University of Science and Technology,Chongqing 401331,China;Southwest Sichuan Gas Field,Mianyang 621700,China;School of Automation,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]重庆科技学院电气工程学院,重庆401331 [2]西南油气田川西北气矿,绵阳621700 [3]北京理工大学自动化学院,北京100081

出　　处：《自动化与仪表》2023年第6期115-119,共5页Automation & Instrumentation

基　　金：重庆市技术创新与应用发展专项项目(cstc2019jscx-msxmX0054)。

摘　　要：对构成组分多、相对浓度差大、化学键相似的混合光谱进行定量分析时,存在各组分光谱谱峰交错重叠严重、高浓度占比组分光谱大量掩盖低浓度占比组分光谱等现象,导致单一组分光谱难以从混合光谱中有效分离解析。针对以上问题,该文以天然气红外光谱作为实验对象,提出一种光谱叠加性原理结合GA-SA算法的定量分析方法对天然气光谱进行解析。算法在全局搜索的基础上结合局部邻域寻优,并通过参数及流程优化,可以有效克服混合光谱求解过程中微量组分光谱寻优性差、易陷入局部极值、不具有重复性等缺点。通过对比实验证明该方法相较于其他光谱分析方法,用于天然气光谱定量分析具备更好的稳定性与重复性,预测光谱与真实光谱相关系数R2大于0.98823。When quantitatively analyzing mixed spectra with multiple components,large relative concentration differences,and similar chemical bonds,there are phenomena such as severe overlapping of spectral peaks of each component,and a large number of high concentration components masking the spectrum of low concentration components,which makes it difficult to effectively separate a single component spectrum from the mixed spectrum.To solve the above problems,this paper proposes a quantitative analysis method based on the spectral superposition principle and GA-SA algorithm to analyze the natural gas spectrum.Based on global search and local neighborhood optimization,the algorithm can effectively overcome the shortcomings of poor spectral optimization of trace components,easy to fall into local extreme values,and non repeatability in the hybrid spectral solution process.Comparative experiments have shown that this method has better stability and repeatability compared to other spectral analysis methods for quantitative analysis of natural gas,and the correlation coefficient R2 between predicted and true spectra is greater than 0.98823.

关 键 词：天然气组分 红外光谱 GA-SA算法 定量分析 

分 类 号：TE81[石油与天然气工程—油气储运工程]