Development And Application Of Online Raman Analyzer For Complex Gas Mixtures

As a new detection technology,Raman spectroscopy has the advantages of non-invasive measurement and fast analysis speed.It is suitable for continuous online detection in industry.As the key to the application of Raman technology,the quantitative analysis algorithm of Raman spectroscopy has always been one of the hotspots of research.When the Raman spectroscopy of most components for a mixture are unknown,how to obtain the component concentration is a challenging problem.For this reason,this article takes natural gas as an example to conduct in-depth study on the quantitative analysis algorithms of Raman spectroscopy and its industrial field applications.The main study content of the paper is:(1)A preprocessing algorithm for gas spectrum is proposed to reduce baseline drift and background noise.And three common baseline correction methods are analyzed and compared.(2)A Raman quantitative analysis algorithm based on genetic algorithm(GA)and linear additive model is proposed.This algorithm only needs to know the Raman spectra of one main component.Firstly,the characteristic peak positions of each component were obtained by correlation analysis.Then the spectrum matrix of pure components corresponding to the mixture was fitted and optimized by genetic algorithm.Finally,the Raman spectroscopy quantitative analysis model of the components in the mixture was established by using the spectral matrix.Taking natural gas as an example,when only the Raman spectroscopy of the main component methane is known,a quantitative analysis model for the main components(CH4,C2H6,C3H8,CO2,N2,H2,CO)in natural gas is established.For the test sample set,the corresponding prediction root mean square error(Root Mean Square Error,RMSE)is 0.488%,0.103%,0.358%,0.079%,0.087%,0.126%,0.022%,and the multi-correlation coefficient(Multi-correlation coefficient,R2)are 0.995,0.991,0.990,0.991,0.998,0.999,0.999,respectively.Although this result is slightly inferior to the result obtained when the Raman spectroscopy of all pure substances are known,it provides a solution for most cases where the spectra of pure substances are unknown.(3)The calorific value information of natural gas is very important for the application of natural gas.Based on the quantitative analysis method of natural gas,the calorific value of natural gas is predicted and analyzed.In addition,a prototype system of an online analyzer for natural gas calorific value based on Raman spectroscopy was built.