Study On The Qualitative Method And Application Of Passive FTIR Telemetry Of Gas

Dangerous gas,endangering people's safety,is released into the atmosphere in the case of sudden leakage of dangerous gas,emission of polluting gas from industrial park and release of chemical warfare agent.At this time,a flexible and online remote monitoring and early warning technology is urgently needed.The passive Fourier transform infrared(p-FTIR)spectrum monitoring technology has the advantages of real-time,non-contact,wide band,high signal-to-noise ratio,and has been widely used in various measurement fields.However,in the actual telemetry application of FTIR system,the key technical problems,such as changing background,gas temperature inversion and multi-component identification,are the stumbling blocks of passive FTIR early warning and monitoringThis paper systematically solves the above problems in the application of p-FTIR system in telemetry recognition.First of all,in the process of passive telemetry of FTIR spectrometer,the constant changes of the instrument state such as the temperature of the light source,the performance of the beam splitter,and the external background environment lead to a certain drift of the spectral baseline.The spectrum with the baseline drift will introduce the pseudo spectral signal in the inversion analysis process,which will seriously affect the subsequent qualitative identification and quantitative analysis of the spectrum.An iterative averaging automatic baseline correction(IA)method is proposed based on the background change characteristics of infrared spectrum to solve this problem.And it was compared with other methods using simulation and experimental data.The results show that IA algorithm has the best effect of baseline correction with the least root mean square error.In order to further study the applicability of baseline correction method,IA baseline correction method was applied to spectra with different signal-noise-ratio,and using the established performance metrics to evaluate the effect of baseline correction.The experimental results show that IA algorithm is less affected by spectral SNR,which shows that this method can solve the problem of background baseline drift well and it provides the key technology in the preprocessing process for the passive recognition analysis method.Then,due to the influence of water vapor and carbon dioxide in the radiometric calibration process of the instrument,a radiometric calibration method is proposed to deduct water vapor and carbon dioxide.Different blackbody radiometric calibration methods are used in the FTIR telemetry system for experimental study.The results show that the calibration curve after the gas absorption is deducted is closer to the ideal measurement state and the RMSE is reduced from 2.5585 X 10'8 to 2.3368 × 10-8,providing accurate telemetry spectrum.It improves the accuracy of infrared spectrum signal measurement and provides guarantee for qualitative and quantitative analysis of pollutant gas composition.A spectral identification algorithm based on variable selection technology is proposed to solve the problem of spectral component identification in p-FTIR telemetry.The main implementation process of the algorithm is to initially screen the component variables by the different variables(sequence forward selection method,lasso and elastic Net)combined with the general cross validation method,and then the component variables with concentration less than 0 were continuously eliminated by the iterative least square method.Finally,the target components were screened according to the weight of the screened variables in the fitting vector.The experimental results show that VSM_GCV_IM method can identify the target components quickly and effectively,with the recognition rate up to 99%,and can accurately identify the interference component H2O at the meantime.This method provides technical support for the application of p-FTIR spectroscopy in emergency monitoring and early warning of dangerous gas leakage.Finally,the p-FTIR spectral analysis method is established by the key technologies proposed in this paper,such as automatic baseline correction,radiation calibration with gas absorption deducted,inversion temperature of vibration rotation spectrum and improved qualitative identification analysis based on variable technology.This spectral analysis method is analyzed and verified by CH4 simulation emission experiment,SF6 continuous intermittent simulation emission experiment and aircraft exhaust telemetry experiment,and it achieved a satisfactory experimental result.