CN103884673B - Online infrared spectroscopy monitoring system and method for hard drug taking condition - Google Patents

Abstract

The invention discloses an infrared spectrum online monitoring system and method for drug ingestion. The system includes a plurality of electromagnetic reversing valves, a Fourier transform infrared spectrometer, an air pump and an industrial computer, or a plurality of air pumps and a Fourier transform infrared spectrometer. and industrial computer, or several groups of optical fiber splitters and couplers, Fourier transform infrared spectrometer and industrial computer; the method utilizes the volatility of drugs and drug decomposition products, and the infrared absorption characteristics of volatile components, using Fourier transform infrared The online analysis method of the infrared spectrum gas of the spectrometer can analyze the drug components volatilized in the air online, so as to realize the online monitoring of drug taking in one or more physical spaces in specific places such as KTV and hotels. If any abnormality is found, the industrial computer will Notify the public security organs through the Internet to provide effective investigation methods and evidence for drug control.

Description

An infrared spectrum online monitoring system and method for drug use

【Technical field】

The invention relates to the field of online monitoring of drug components, in particular to an infrared spectrum online monitoring system and method for drug smoking.

【Background technique】

According to the statistics of the United Nations, the annual drug trade volume in the world reaches more than 500 billion U.S. dollars, and the transaction volume continues to rise. Drugs will endanger people's lives, disrupt social order, and cause national instability. Therefore, it can be said that drugs have become an international One of society's greatest public hazards. Since drug taking and drug trafficking are mostly carried out in hotels, hotels, nightclubs and other places, the monitoring of these places is an important means of drug detection and drug control. At present, there is no instrument that can monitor these places online, and the monitoring is mainly carried out by the inspection of the public security department. On the one hand, the current inspection method consumes a lot of police force; on the other hand, it interferes with the normal business of hotels, hotels and other places; third, the efficiency of drug inspection is very limited. Therefore, develop a kind of online monitoring instrument of drug taking situation, by monitoring the composition of drug volatile matter and its decomposition product in the air, come real-time monitoring whether people take drugs in places such as hotels, hotels, will greatly improve the intensity of drug detection and anti-drug, meanwhile, It also reduces the need for drug detection police. Therefore, the development of online drug monitors is of great significance to maintaining social order and stability and protecting public health.

At present, the inspection of drugs is mainly aimed at studying the type, composition, structure, nature and content of drugs, so as to provide scientific basis and important clues for the detection and trial of drug-related crimes. Drug inspection mainly includes appearance inspection, physical inspection, chemical inspection, instrumental analysis, etc. Appearance inspection is to observe and compare the color, smell, physical form and properties of drugs, and make a preliminary judgment on drugs; physical inspection is mainly to test the physical properties of drugs such as solubility, melting point, and boiling point, and conclude that it is related to the processing process. information; chemical testing and instrumental analysis methods are mainly to conduct qualitative and quantitative analysis of the chemical components of drugs, such as drug components, additive components, adulterated components, impurities, solvents, metals and other components, so as to obtain information on the process of drug processing and synthesis. Information on the raw materials used, synthetic methods and routes, and determination of drug sources, etc. In addition, drug testing also includes testing the drug prototype and its metabolites in biological samples such as saliva, urine, blood, and hair to determine whether the suspect has taken drugs and the type and time of drug use. Drug testing materials mainly include drug suspects in various forms, such as lumps, powders, injections and other solid substances, liquids, etc.; biological testing materials, such as urine, blood, nails, etc. of drug addicts. After sampling these samples, the corresponding methods are used for analysis.

The analysis of drugs is a complex and difficult task, and the analytical techniques used in it involve almost all techniques of analytical chemistry. In the initial stage of the development of instrumental analysis technology, drug testing was mainly done by chemical analysis methods, such as chemical chromogenic method. With the development and maturity of instrument analysis technology, molecular spectroscopy, such as fluorescence analysis (Fluorescence analysis), laser Raman technology (Laser Raman Spectroscopy, LRS), Fourier transform infrared spectroscopy (Fourier Transform Infrared, FTIR), chemiluminescence ( Chemoluminescence method), etc.; chromatography, such as thin-layer chromatography (Thin-layer chromatograph), high-performance liquid chromatography (High Performance Liquid Chromatograph, HPLC), gas chromatography (Gas Chromatograph, GC), etc., chromatography-mass spectrometry (such as GC-MS, HPLC-MS, GC-IMS, GC-HRMS, etc.) and capillary electrophoresis (Capillary electrophoresis, CE) and other instrumental analysis methods have become the main means of drug testing and analysis, while chemical analysis methods are used as auxiliary means for drug testing and analysis preliminary screening and pre-experimentation. Commonly used drug analysis methods and their characteristics are as follows:

1) Chemical chromogenic method This is a qualitative analysis method, which usually uses the action of drugs and certain chemical reagents to produce characteristic colors to determine their existence. This method is simple, intuitive, and easy to operate, but requires more samples and has poor specificity;

2) Fluorescence analysis Fluorescence analysis is a method of qualitative analysis based on the molecular fluorescence spectrum of the substance, and quantitative analysis based on the fluorescence intensity. Fluorescence analysis method has high sensitivity, which can reach 10 ^-6 . Environmental factors should be considered when using, different solvents, temperature, pH value and noise will have an impact on the measurement results, but sometimes these factors can also be used in special cases;

3) Laser Raman spectroscopy When monochromatic light is irradiated on a substance, scattered light is formed, and the frequency of most of the scattered light is the same as that of the incident light. This kind of scattering is called Rayleigh scattering; it can be observed in scattered light A fraction of the frequencies differ from the frequency of the incident light, and this scattering is known as Raman scattering. The difference between the frequency of Raman scattering rays and the frequency of incident light is called Raman frequency shift. This frequency shift can characterize the vibration characteristics of different groups in molecules, so it can be used for qualitative and structural analysis of molecules. In addition to determining the molecular structure of various drugs, Raman spectroscopy can also detect whether an unknown item is a drug according to the Raman scattering spectrum;

4) Chemiluminescence analysis Chemiluminescence is the excitation energy provided by chemical reactions, which excites product molecules or other coexisting molecules to produce light radiation. Chemiluminescence analysis is to detect and analyze different substances according to the difference of light radiation. There are two types of chemiluminescence analysis: liquid-phase chemiluminescence and gas-phase chemiluminescence. Commonly used chemiluminescence analysis instruments include discrete sampling and flow injection. Experiments have proved that chemiluminescence detection is generally more sensitive and accurate than ultraviolet spectrometry and fluorescence analysis, and it has become an important trace analysis method;

5) Chromatography, chromatography-mass spectrometry Chromatography is an instrument that separates the components of the analyte through a capillary to analyze the composition of the substance, and is especially suitable for the quantitative analysis of organic matter. Mass spectrometer is an analytical instrument for molecular identification, especially suitable for qualitative analysis. The two analytical instruments are connected through a special interface (molecular separator), and the computer is used to automatically control the operating parameters after the connection, so that it can be integrated and provide analysis information. Chromatography-mass spectrometry can perform rapid qualitative and quantitative analysis of components in complex mixture samples, and has the characteristics of accurate qualitative, fast analysis, and easy operation. Chromatography-mass spectrometry currently used mainly includes GC/MS and HPLC/MS, and this method has become one of the most effective and commonly used standard verification methods for qualitative and quantitative analysis of drugs. However, chromatographs and mass spectrometers are relatively expensive, and the combination of the two instruments is even more expensive;

6) Capillary Electrophoresis Capillary electrophoresis (CE), also known as high-performance capillary electrophoresis (HPCE), is one of the fastest-growing analytical chemistry research fields in recent years. As far as drug analysis is concerned, the chemical properties of drug components may vary greatly, or the molecular structure may be very similar, and impurities and diluents often exist in trace amounts. Capillary electrophoresis technology has high separation efficiency and high mass sensitivity. The demand for samples and solvents is small, and a variety of separation modes can be used, combined with a variety of detection methods, and the analysis range is wide, so it is suitable for effective qualitative and quantitative analysis of complex drugs. At present, it has been applied to the analysis and detection of several common drugs such as morphine hydrochloride, heroin, and thebaine. In addition, compared with other technologies applied to drug testing, CE has a unique separation mechanism, which can be used as an important method for drug system analysis. It can be mutually authenticated with other methods to complete the description of various drugs, establish a fingerprint library, and further provide information for drug analysis. Provide strong technical support for source deduction;

7) Fourier Transform Infrared Spectroscopy In 1999, the University of St. Andrews in the United Kingdom cooperated with the industry to develop a new type of spectrometer that can quickly analyze various toxic gas components, which has appeared on the market. The instrument is an improved Fourier transform infrared spectrophotometer, which can determine the type of poisonous gas through the absorption of different wavelengths of poisonous gas.

From the above analysis, it can be seen that most of the current detection of drugs at home and abroad is qualitative and quantitative analysis of the original drug in various samples and drug metabolites in biological samples. The detection of metabolites requires sampling of individuals or samples. It is a post-drug test rather than a pre-event or ongoing test, so it is not suitable for online monitoring. The analysis of drug components also requires procedures such as sampling and inspection, and does not test the discovery of drugs in specific areas. Therefore, from existing narcotics inspection instrument and method, still do not have the narcotics on-line monitoring instrument that can be cheap at present. Conventional inspection instruments basically involve procedures such as sampling and manual analysis. Some drug testing instruments are quite expensive and complicated to operate, such as chromatography-mass spectrometry, chromatographs and mass spectrometers are relatively expensive instruments, and the price of a single instrument is as high as hundreds of thousands of yuan. Moreover, the operation of these instruments All personnel need to undergo strict training to achieve correct analysis, so it cannot meet the needs of automatic virus detection.

【Content of invention】

The object of the present invention is to provide an infrared spectrum online monitoring system and method for drug abuse. It provides effective special reconnaissance means and drug detection evidence for drug control.

In order to achieve the above object, the first kind of parallel technical scheme that the present invention adopts is:

An infrared spectrum on-line monitoring system for drug abuse, including several electromagnetic reversing valves, Fourier transform infrared spectrometers, air pumps and industrial computers; wherein, the air inlets of several electromagnetic reversing valves are set at corresponding physical locations to be monitored. In the space, the air outlets of several electromagnetic reversing valves are respectively connected with the air chamber inlets of the Fourier transform infrared spectrometer. The industrial computer is used to select one of the electromagnetic reversing valves to work at the same time, and the air pump is used to monitor The gas in the physical space is pumped into the gas chamber of the Fourier transform infrared spectrometer, and the Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the monitored physical space gas to the industrial computer.

The method based on the infrared spectrum online monitoring system of above-mentioned drug taking situation comprises the following steps:

1) At the same time, the industrial computer selects one of the electromagnetic reversing valves to work, and the air pump pumps the air in a monitored physical space corresponding to the electromagnetic reversing valve to the Fourier transform infrared spectrometer through the electromagnetic reversing valve The gas chamber conducts online analysis to obtain the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space;

2) The Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space to the industrial computer, and the industrial computer will compare the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space with the preset number Compared with the absorption spectrum data of a drug gas, if the absorbance spectrum data of the gas in the monitored physical space is greater than or equal to the preset value of the absorption spectrum of one or more preset drug gases, or if the transmittance spectrum data of the gas in the monitored physical space If it is less than or equal to the preset value of the absorption spectrum of one or more preset drug gases, the industrial computer will send out a warning signal; Absorption spectrum preset value, then the industrial computer selects another electromagnetic reversing valve to work.

The second parallel technical solution adopted by the present invention is: an infrared spectrum on-line monitoring system for drug ingestion, including several air pumps, Fourier transform infrared spectrometers and industrial computers; wherein, the air inlets of several air pumps are arranged in In the corresponding monitored physical spaces, the air outlets of several air pumps are respectively connected with the air chamber inlets of the Fourier transform infrared spectrometer. The industrial computer is used to select one of the air pumps to work at the same time, and the Fourier transform infrared The spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the monitored physical space gas to the industrial computer.

The method based on the infrared spectrum online monitoring system of above-mentioned drug taking situation comprises the following steps:

1) At the same time, the industrial computer selects one of the air pumps to work, and the air pump pumps the air in the monitored physical space to the gas chamber of the Fourier transform infrared spectrometer for online analysis, and obtains the absorbance spectral data of the gas in the monitored physical space and transmittance spectral data;

2) The Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space to the industrial computer, and the industrial computer will compare the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space with the preset number Compared with the absorption spectrum data of a drug gas, if the absorbance spectrum data of the gas in the monitored physical space is greater than or equal to the preset value of the absorption spectrum of one or more preset drug gases, or if the transmittance spectrum data of the gas in the monitored physical space If it is less than or equal to the preset value of the absorption spectrum of one or more preset drug gases, the industrial computer will send out a warning signal; Absorption spectrum preset value, then the industrial computer selects another pump to work.

The third parallel technical solution adopted by the present invention is: an infrared spectrum online monitoring system for drug abuse, including several groups of optical fiber splitters and couplers, Fourier transform infrared spectrometers and industrial computers; wherein, several groups of optical fiber splitters The circuits and couplers are arranged in corresponding physical spaces to be monitored, and each group of optical fiber branches and couplers couples the interference light absorbed in the corresponding monitored physical spaces to the light source of the Fourier transform infrared spectrometer, industrial control The computer is used to select a group of optical fiber splitters and couplers to work at the same time, and the Fourier transform infrared spectrometer transmits the absorbance spectral data and transmittance spectral data of the monitored physical space to the industrial computer.

The method based on the infrared spectrum online monitoring system of above-mentioned drug taking situation comprises the following steps:

1) At the same time, the industrial computer selects a group of optical fiber splitters and couplers to work, and the group of optical fiber splitters and couplers couples the absorbed interference light in the monitored physical space to the light source of the Fourier transform infrared spectrometer for online analysis , to obtain the absorbance spectrum data and transmittance spectrum data of the monitored physical space gas;

2) The Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space to the industrial computer, and the industrial computer will compare the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space with the preset number Compared with the absorption spectrum data of a drug gas, if the absorbance spectrum data of the gas in the monitored physical space is greater than or equal to the preset value of the absorption spectrum of one or more preset drug gases, or if the transmittance spectrum data of the gas in the monitored physical space If it is less than or equal to the preset value of the absorption spectrum of one or more preset drug gases, the industrial computer will send out a warning signal; Absorption spectrum preset value, then the industrial computer selects another group of optical fiber branch and coupler to work.

Compared with the prior art, the present invention utilizes the volatility of drugs and the decomposition products of drugs in the absorption process, and the infrared absorption characteristics of the active ingredients in the volatile matter, and adopts the method of infrared spectrum gas analysis to analyze the drug components volatilized in the air. Online analysis is carried out to realize online monitoring of drug use in specific places such as KTVs and hotels, and to provide effective investigation methods and evidence for drug control.

【Description of drawings】

Fig. 1 is the schematic diagram of the infrared spectrum online monitoring system of drug taking situation of the present invention;

Fig. 2 is a schematic diagram of optical path switching of a drug online monitoring system in multiple physical spaces;

Fig. 3 is heroin base and heroin hydrochloride infrared absorption spectrogram;

Figure 4(a) is a single-component 10% methane, 2% ethane, 1% propane infrared absorption spectrum, and Figure 4(b) is a single-component mid-infrared spectrum of a mixture of 10% methane and 2% ethane High-wavelength spectrum;

Figure 5(a) is the three spectra obtained during the long-term on-line monitoring of methane, ethane and propane gases with baseline distortion;

Fig. 5 (b) is the spectrogram after baseline correction of the three spectrograms in Fig. 5 (a);

Fig. 5(c) is the initial spectrogram of data1 in Fig. 5(a), and its baseline corrected spectrogram and reconstructed spectrogram;

Fig. 5(d) is the initial spectrogram of data2 in Fig. 5(a), and its baseline corrected spectrogram and reconstructed spectrogram;

Fig. 5(e) is the initial spectrogram of data3 in Fig. 5(a), and its baseline corrected spectrogram and reconstructed spectrogram.

【detailed description】

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Referring to Fig. 1 and Fig. 2, a kind of drug taking situation infrared spectrum online monitoring system and method of the present invention, this system comprises several electromagnetic reversing valves, Fourier transform infrared spectrometer, air pump and industrial computer, or several air pumps, Fu Fourier transform infrared spectrometer and industrial computer, or several groups of optical fiber splitters and couplers, Fourier transform infrared spectrometer and industrial computer; the method utilizes the volatility of drugs and the decomposition products of drugs in the absorption process, and the volatile matter Infrared absorption characteristics of the active ingredients in the drug, using the infrared spectrum gas analysis method to conduct online analysis of the drug components volatilized in the air, so as to realize online monitoring of drug abuse in specific places such as KTVs and hotels. Notify the public security organs to provide effective investigation methods and evidence for drug control. The principle diagram of infrared spectrum online monitoring of drug abuse is shown in Figure 1. For places with multiple physical spaces, such as hotels, a set of optical platforms is used in one location, one optical path/air path for each independent physical space, and multiple optical paths/air paths share a set of optical platforms according to the switching method of optical path/air path , and then scan the air spectrum in each physical space one by one to determine whether there are drugs taken, so as to save system costs. The optical path/gas path switching of the infrared spectrum online monitoring system for drug abuse is shown in Figure 2. For the drug online testing system with optical path switching, the optical fiber is first used to send the optical signal to each physical space, and after the air in the physical space absorbs the optical signal, the optical signal is transmitted back to the optical platform with the optical fiber; for the air path switching The test system uses an electromagnetic reversing valve to switch the air path, and there is also an exhaust device to take samples from the air in the physical space and transport them to the sample chamber of the optical platform. As for the instrument that needs to be moved for drug testing, there is only one optical path, a set of optical platform, and a miniature exhaust fan in the gas path, which pumps the sample gas into the sample gas chamber of the optical platform for spectral analysis.

The spectral online analysis of drugs is carried out according to the method of gas online analysis. First obtain the drug standard sample and its spectrum, and then use the feature extraction method to extract the characteristic variables for each drug, and use these characteristic variables as input, and use the drug species and concentration as the output, using partial least squares method, neural The network method establishes its analysis model. In the actual on-line analysis process, the spectral baseline is first corrected, and then the characteristic variable values are calculated according to the calculation formula of the characteristic variables, and the concentration of each component of the drug gas is calculated according to the established model parameters. If the concentration of a certain drug gas component reaches a certain set concentration value, a warning signal will be given.

Furthermore, the infrared spectrum online monitoring of drug use in multiple physical spaces of the present invention is carried out by switching the optical path or the gas path, and using a Fourier transform infrared spectrometer to scan each monitored physical space one by one. For analysis, if the gas circuit is switched, the gas in the monitored physical space is pumped to the gas chamber of the spectrometer by an air pump; The light is coupled to the spectrometer light source.

If the infrared spectroscopy online monitoring system for drug abuse is of the air circuit switching type, each monitored physical space has a pipeline connected to the air chamber of the Fourier transform infrared spectrometer through a solenoid valve, and only one of the air circuits is connected to it. The state of the connected solenoid valve is in the open state. When the system is monitoring the i-th physical space and needs to switch to the i+1-th physical space, the closing signal is sent to the solenoid valve connected to the i-th physical space pipeline, and at the same time , send the opening signal to the solenoid valve connected to the pipeline in the i+1th physical space; if the infrared spectrum online monitoring system for drug abuse is an optical path switching type, then each monitored physical space has two optical fibers through the optical fiber branch The / coupler is connected to the Fourier transform infrared spectrometer, one is used to transmit the coupled light source interference light to the monitored physical space, and the other is used to transmit the interference light absorbed by the volatiles of drugs and their decomposition products For the detector, the system sends the switching signal to the fiber splitter/coupler. The switching method is the same as the gas path switching method. At any time, only one optical path is in the optical path of the Fourier transform infrared spectrometer.

Among them, finding anomalies refers to monitoring the possibility of taking drugs in a certain physical space, and its varieties can include the drugs currently legalized by the state. Notifying the public security organs through the Internet refers to notifying the public security organs through text messages and the Internet, including the contents of the drugs taken. Variety, the location of the place, including the address, name, and room number of the hotel/KTV.

The infrared spectrum on-line monitoring system for drug abuse of the present invention is composed of hardware and software. The present invention will be further described in detail below in conjunction with the accompanying drawings.

1. Hardware part

The hardware part of the infrared spectrum online monitoring system for drug abuse includes connecting pipelines, pipeline taps, industrial computers, and Fourier transform infrared spectrometers. For optical fiber switching monitoring systems, there are also optical fiber splitters/couplers, infrared optical fiber , and for the gas circuit switching monitoring system, there are also multi-channel electromagnetic reversing valves and exhaust pumps.

The Fourier transform infrared spectrometer is used to obtain the absorption spectrum of drug gas, which can be Bruker’s Tensor27, Tensor37, or PE’s Spectrum Two, as long as it can provide a resolution of 0.5cm ^-1 or more, stable performance, wavenumber The upper limit of the range can reach more than 6000cm ^-1 and the lower limit is less than 600cm ^-1 ;

The connecting pipeline can be glass tube, rubber tube, stainless steel tube and other closed connecting tubes that are not corroded by multi-component gas to be measured. For this implementation example, as long as the connecting pipeline is not corroded by organic matter, the diameter of the pipeline is enough to be buried in the wall, and it can withstand 0.9-1.2 atmospheric pressure;

The industrial computer is used to obtain the absorption spectrum of the drug gas from the spectrometer, and realize the quantitative analysis of the gas, and control the switching of the optical fiber/gas circuit. As long as there is a network port, it can perform Internet communication and send binary signals. 500M memory, 1G master frequency or above, the model is not limited, it can be an ordinary industrial computer, it can also be an ordinary personal desktop computer, or even a notebook computer;

The electromagnetic reversing valve can be multi-way or two-way. If it is the former, he itself has played the role of a pipeline tap, and no longer needs a pipeline tap. If it is the latter, a pipeline tap is also needed. The electromagnetic reversing valve is controlled by the industrial computer to determine which physical space the air is pumped into the air chamber of the Fourier transform infrared spectrometer to monitor which physical space for drugs;

The pipeline tap is used to connect the pipeline for transporting gas in each physical space and the gas pipeline for the gas chamber of the Fourier transform infrared spectrometer;

The exhaust pump is used to draw the gas in each monitored physical space into the air chamber of the Fourier transform infrared spectrometer, while maintaining the air pressure in the air chamber within a certain range;

The optical fiber is used to transmit optical signals. On the one hand, it transmits the optical signal of the Fourier transform infrared spectrometer to the physical space where drug online monitoring is required. On the other hand, it receives and transmits the optical signal to the photoelectric detection of the Fourier transform infrared spectrometer. device. As long as the optical fiber can transmit infrared light with a wavelength above 1 μm, the model is not required;

As long as the optical fiber splitter/coupler can couple the interference light of the Fourier transform infrared spectrometer into the optical fiber, it can couple the absorbed infrared light in the monitored physical space into the optical fiber.

2. Software part

The software runs in the industrial computer, and its functions mainly include the control of the exhaust pump, optical path/gas path switching, spectral analysis of drug volatiles or decomposition gas, and signal transmission.

1) Control of the suction pump

If the infrared spectrum on-line monitoring system for drug abuse is of the gas path switching type, a suction pump is installed. The exhaust pump is used to draw the gas in each monitored physical space into the air chamber of the Fourier transform infrared spectrometer, while maintaining the air pressure in the air chamber within a certain range. The pressure is determined according to the actual application. The determined principle is that the pressure in the gas chamber is consistent with the pressure when the sample gas is produced during the system development process, and the deviation is determined by the user. The control algorithm can adopt PID control algorithm:

Δ=FS-F(1a)

In the formula, C(n) is the current control quantity; F is the current test pressure value; F _S is the pressure reference value that will not affect the production or safety of the front end; Δ(n), Δ(n-1) are the reference values respectively The difference from the current pressure value and the pressure value at the previous moment; p _p , pI and p _D are the coefficients of the proportional link, integral link and differential link in PID control, respectively, and they can be obtained by adjusting the PID control parameters. If the infrared spectrum online monitoring system for drug abuse is of the optical path switching type, this step is not required;

2) Switching of optical path/gas path

If the infrared spectrum on-line monitoring system for drug abuse is of the gas path switching type, only one of all the gas paths is in an open state. When the system is monitoring the i-th physical space and needs to switch to the i+1-th physical space, the closing signal is sent to the solenoid valve connected to the i-th physical space pipeline, and at the same time, the opening signal is sent to the i-th physical space +1 Solenoid valve connected by physical space pipeline. If the infrared spectrum on-line monitoring system for drug abuse is an optical path switching type, the switching signal is sent to the optical fiber splitter/coupler, and the switching method is the same as the gas path switching method;

3) Spectral analysis of gas composition and concentration of drug volatiles and decomposition products

Spectral analysis of drug volatiles and decomposition gas components and concentrations is the key point of the present invention. The present invention uses heroin as an example to illustrate the spectral analysis method of drugs. For illustration, only heroin base and heroin hydrochloride are considered. The absorption spectra of heroin base and heroin hydrochloride are shown in Figure 3. The conventional spectral analysis of gas composition and concentration is realized through an analysis model. The input of the analysis model is the characteristic variable formed by some spectral data after certain operations, and the output is the concentration of the target gas. Since the volatile content of the drug and its decomposition products in the air is low, the concentration is relatively small, and its absorption spectrum will not tend to be saturated. The nonlinear absorbance of the spectral line is basically not considered, and the characteristic variable can be used as the drug component. Analysis model. Due to the long-term working process of the spectrometer, the spectrum is prone to distortion, so it is necessary to identify and process the distortion to obtain reliable and accurate analysis results. Therefore, spectral analysis mainly includes two parts: feature variable extraction of each drug component, spectral distortion identification and processing. The first step is completed in the system development process, and the latter step is a step that must be completed for each spectral analysis.

(1) Analysis model establishment

· Sample gas spectrum acquisition

To establish an analysis model for multi-component gases, a certain calibration sample must be made first. In this implementation example, it is necessary to analyze heroin base and heroin hydrochloride. In a given sample, fill it into the air chamber of a Fourier transform infrared spectrometer, first obtain its infrared absorption spectrum, and then take a small amount of sample gas from the air chamber for use Analyzed by gas chromatography to obtain its concentration.

· Feature variable extraction

In the process of feature variable extraction, a certain spectral line can be extracted as the characteristic variable of a certain gas, or the area of a certain spectrum can be used as the characteristic variable, and the combination of multiple spectral line values can also be extracted as the characteristic variable of different gases. For different applications, different extraction methods are used, and the selected feature variables are also different. For example, observing accompanying drawing 3, we can know that, for the spectral quantitative analysis of the mixed gas of heroin base and heroin hydrochloride two gaseous drugs, heroin hydrochloride has a stronger and wider absorption peak at 2630cm ^-1 , while heroin base No, on the contrary, heroin base has a relatively sharp absorption peak at 3060cm ^-1 , but heroin hydrochloride does not. Therefore, these two absorption peaks can be regarded as characteristic absorption peaks to distinguish the two drugs. Feature variable selection methods include forward selection method, Tikhonov method, least squares method, principal component analysis method, etc. For simplicity, the forward selection method is used here, and two spectral line values with wavenumbers of 2630cm ^-1 and 2135cm ^-1 are selected The difference of the natural logarithms of is used as the characteristic variable of heroin hydrochloride:

v _m =ln(val ₂₁₃₅ )-ln(val ₂₆₃₀ )(2a)

The natural logarithm difference of the spectral line values of 3060cm ^-1 , 3000 cm ^-1 and _3140cm ^-1 is used as the characteristic variable of heroin base:

v _e =ln(val ₃₀₀₀ )+ln(val ₃₁₄₀ )-2×ln(val ₃₀₆₀ )(2b)

In the formula, v _m and v _e represent the characteristic variables of heroin hydrochloride and heroin base respectively, and ln(val _n ) represents the natural logarithm of the spectral line value with wave number n. The forward selection method used here is to compare the characteristic variable formed by the difference between the two spectral lines by observing the attached figure 3, which has a higher sensitivity to a certain gas and a lower sensitivity to other gases. The natural logarithm is used because the spectral line value is the spectral transmittance, and the linearity of the characteristic variable formed after taking the natural logarithm is relatively high. The spectral line difference is taken as the characteristic variable because the characteristic variable formed by this method is beneficial to eliminate the influence of spectral baseline shift. If the ordinate of the spectrogram is in the form of absorbance, then there is no need to calculate the natural logarithm in formula (2), just use the spectral line value directly;

(2) Spectral analysis and spectral distortion identification and processing

After the spectrometer works for a long time, due to certain changes in environmental parameters and spectrometer device characteristics, the spectrum will be distorted, which will seriously affect the analysis results. Spectral distortion identification and processing are identified and processed according to the spectral distortion, so as to minimize the impact of spectral distortion on the analysis results. At the same time, the identification and processing of spectral distortion must be combined with spectral analysis to obtain ideal results. In this application example, the spectral distortion is divided into baseline regular distortion and irregular distortion. For regular distortion, spectrum correction is realized through spectral translation and rotation. For irregular distortion, this analysis is abandoned and an error is prompted. Or give the credibility of the analysis results. If continuous irregular distortion occurs, rescan the background. In view of the fact that the long-term drug online monitoring experiment is likely to cause space pollution and is not general, this embodiment uses the online monitoring of methane, ethane and propane as an example to illustrate the spectral distortion identification and processing method. The spectrogram of the single-component gas methane, ethane and propane that the concentration that obtains is 1% is as shown in accompanying drawing 4 (a), the single-component gas 10% methane, 1% ethane, 1% propane that obtains, And the mixed gas spectrum of 10% methane and 2% ethane is shown in accompanying drawing 4 (b). Corresponding to formula (2), the characteristic variables of methane, ethane and propane are:

v _m =ln(val ₃₀₂₂ . ₃ )-ln(val ₃₀₁₆ . ₅ )(3a)

v _e =ln(val _3081.2 )-ln(val _3029.2 )(3b)

v _p =ln(val _3028.0 )-ln(val _3002.2 )(3c)

The calculation expressions of the concentration of methane, ethane and propane are respectively:

C ₁ =a ₁ v _m +a ₂ v _e +a ₃ v _p (4a)

C ₂ =b ₁ v _m +b ₂ v _e +b ₃ v _p (4b)

C ₃ =c ₁ v _m +c ₂ v _e +c ₃ v _p (4c)

The three spectrograms data1, data2 and data3 shown in Figure 5(a) are the spectra obtained during the long-term online monitoring of the three-component alkane gas by the Fourier transform infrared spectrometer, and their spectral analysis and spectral distortion identification and processing are respectively Use the following steps to achieve:

·Gas composition and non-sensitive area search

Observing the accompanying drawing 4(a), it can be seen that the sensitivity of methane, ethane and propane is very small near wavenumbers 600, 1100, 2000, 2500, 3400, etc., which are called non-sensitive regions. For the sensitivity Si _j of the i-th gas in the j-th non-sensitive area, it can also be determined by formula (5):

s _ij =inv(Y _i *Y _i ')*Y _i *(1-V _ij )'(5)

In the formula, Y _i represents the concentration value vector of the i-th gas single-component sample; Y _i ' represents the transposition of the vector Yi; V _ij =[mean(v _{ij1 ),mean(v ij2 )} ,...,mean( v _ijN )] represents the spectral line mean vector of the jth non-sensitive area in the single-component sample spectrum of the i-th gas, and mean(v _ijk )(k=1,2,...,N) represents the i-th gas The average value of several spectral lines in the jth non-sensitive area of the kth single-component sample spectrum; inv(·) represents the matrix inversion operation. Using the formula (3) and the spectral data of the single-component samples of methane, ethane, and propane in the above 5 wavenumber bands, and selecting 5 consecutive spectral lines for each band to calculate the average value, their sensitivity coefficients can be calculated as:

S ₆₀₀ = [00.0000420.000076]; S ₁₁₀₀ = [0.0000290.0000960.000549]

S ₂₀₀₀ = [0.0000220.0001510.000331]; S ₂₅₀₀ = [0.0000240.0001710.000352]

S ₃₄₀₀ = [0.0000060.0001300.000281]; (6)

Between all two adjacent non-sensitive areas, the spectral segment is translated and rotated to perform baseline correction

For the above 5 non-sensitive areas, the entire spectrum can be divided into 4 intervals, and each interval can be translated and rotated to correct the regular distortion of the spectrum. Since each spectrogram consists of two columns, the first column is the wave value, the second column is the spectral line value corresponding to the first column, and the spectral line numbers corresponding to the wave numbers around 3400, 2500, 2000, 1100 and 600 are in order For 273, 753, 1023, 1503, 1758. Therefore, take 5 spectral line values in each non-sensitive area to find the average deviation, assuming that the concentration vector of the three gases to be analyzed is C=[C ₁ ,C ₂ ,C ₃ ], the following source code can be used to realize the initial spectrum Translation and rotation:

Baseline3400=1-mean(data(271:275,2))-S ₃₄₀₀ C';% Find the deviation at wave number 3400

Baseline2500=1-mean(data(751:755,2))-S ₂₅₀₀ C';% Find the deviation at wave number 2500

Baserate2500=(Baseline2500-Baseline3400)/(753-273);% Find the slope between wavenumber 2500 and 3400

data(1:753,2)=data(1:753,2)+Baseline3400+([1:753]'-271)*Baserate2500;% correct the spectrum between wavenumber 2500 and 3400

Baseline2000=1-mean(data(1021:1025,2))-S ₂₀₀₀ C';% Find the deviation at wave number 2000

Baserate2000=(Baserate2000-Baserate2500)/(1023-753);% Find the slope between wavenumber 2000 and 2500

data(754:1023,2)=data(754:1023,2)+Baseline2500+([754:1023]'-753)*Baserate2000;% correct the spectrum between wavenumber 2500 and 3400

Baseline1100=1-mean(data(1501:1505,2))–S ₁₁₀₀ C';% Find the deviation at wave number 1100

Baserate1100=(Baseline1100-Baseline2000)/(1503-1023);% Find the slope between wavenumber 1100 and 2000

data(1024:1503,2)=data(1024:1503,2)+Baseline2000+([1024:1503]'-1023)*Baserate1100;% correct the spectrum between wavenumber 1100 and 2000

Baseline600=1-mean(data(1756:1760,2))–S ₆₀₀ C';% Find the deviation at wave number 1100

Baserate600=(Baseline600-Baseline1100)/(1758-1503);% Find the slope between wavenumber 600 and 1100

data(1504:1866,2)=data(1504:1866,2)+Baseline1100+([1504:1866]'-1503)*Baserate600;% correct the spectrum between wavenumber 600 and 1100

During the continuous online spectral analysis process, the gas concentration vector C in the above source code is set as the last analysis result. If it is the first analysis, set it to 0. For the accompanying drawing 5(a) in this implementation example, there are three spectrograms: data1, data2 and data3. Comparing accompanying drawing 5(a) with accompanying drawing 4(a), since there is no strong absorption peak near the wave number 2900, it shows that in the gases represented by these three spectra, the concentration of various alkanes is very small, therefore, in From wavenumber 800 to wavenumber 1100, and in the range from 2500 wavenumber to 3400 wavenumber, it is almost a straight line with an amplitude of 1. But in Figure 5(a), the spectral value of the spectral segment from wavenumber 800 to wavenumber 1100 is obviously greater than 1, and it is slightly inclined. The spectral segment to wavenumber 1100 is slightly larger, so there is a baseline regular distortion that needs to be corrected. Set the gas concentration vector C to 0, and use the source code of this step to correct the spectrogram obtained after correction as shown in Figure 5 (b);

Use the calibrated analysis model to analyze the corrected spectrum to obtain gas components and their concentrations

Substitute the spectra of data1, data2 and data3 shown in Figure 5(b) into formula (3) respectively, calculate the respective characteristic variable values v _m , v _e and v _p , and calculate the respective concentrations according to formula (4) .

_C1 = [0.01960.00320.0017];

_C2 = [0.00710.00120.0003]; (7)

C3 ₌ [0.04730.00210.0023];

If part of the gas concentration is relatively large, use the latest calculated gas concentration value to substitute into the previous spectral correction step, and re-correct the spectrum until the difference between the two adjacent deviation values of any non-sensitive area in this step is less than a certain threshold . The threshold of each non-sensitive area is set as the noise amplitude of the spectrum in this wavenumber band. Since the gas concentration of each component in this implementation example is very small, it is enough to correct once by this step, and the result of formula (4) can be regarded as the final analysis result.

Reconstruct the spectrum from the analyzed gas components and their concentrations

In order to reconstruct the spectrum, it is first necessary to estimate the converted absorbance of each spectral line, that is, the product of the absorbance and the optical path of the spectrometer. The converted absorbance of the j-th single-component sample of the i-th gas at the k-th spectral line is:

δ _ik,j =-log(v _ikj )/c _i , _j (8)

In the formula, v _ikj represents the spectral value of the j-th single-component sample of the i-th gas at the k-th spectral line; log( ) means the natural logarithm operation; c _i,j means the j-th spectral value of the i-th gas One-component sample concentration. If the calculated concentration value of the i-th gas is ci,x, and ci, _j <ci, _x ≤ci _,j+1 , j=1,2,...,N-1, then the k-th At the spectral line, the converted absorbance of the gas is:

rate=(c _i,x -c _i,j )/(c _i,j+1 -c _i,j )(9a)

δ _ikx =(1-rate)×δ _ik,j +rate×δ _ik,j+1 (9b)

In order to shorten the calculation time, only the target gas suction peak and its nearby spectral segments in the spectrum can be reconstructed. In this embodiment, only the wavenumber segments 700-1300 and 2800-3200 need to be reconstructed. For the sake of simplicity, in this embodiment, the spectral line reconstruction method at the 579th spectral line (corresponding to wavenumber 2881.5) is used to illustrate the spectral line reconstruction method. Since in the case of small concentrations, the converted absorbance of each component gas hardly changes with the change of gas concentration, so the converted absorbance of each component gas in this embodiment can directly use the converted absorbance at 0.1% concentration. For example, the value of the spectral line at the 579th spectral line (corresponding to wave number 2881.5) of a single-component sample of 0.1% n-pentane is 0.9462, so its converted absorbance is calculated by formula (10):

δ _7,579,4 =-log(0.9462)/0.1=0.5530(10)

By analogy, the converted absorbance of the three-component gas at 579 spectral lines can be obtained and the vector Δ is formed to obtain:

Δ=[0.02000.14100.5393](11)

From the formulas (7) and (11) according to the Lambert-Beer theorem, the reconstructed spectral values of the 579th spectral line of data1, data2 and data3 can be obtained respectively as follows:

v _1,579 =exp(-C ₁ Δ')=0.9982

v _2,579 =exp(-C ₂ Δ')=0.9995

v _3,579 =exp(-C ₃ Δ')=0.9974

By analogy, the reconstructed spectral line value of each spectral line can be obtained. Therefore, for data1, data2 and data3 in Fig. 5(a), the spectra before, after correction and reconstruction are shown in Figs. 5(c), 5(d) and 5(e) respectively.

· Spectral distortion identification and processing

For the regular distortion of the spectrum, the previous steps have actually corrected the spectrum. This step is used to identify the irregular local distortion of the spectrum, and make reasonable processing according to the identification results.

Observing the spectrum in Figure 5(c), it can be found that the corrected data1 spectrum value is significantly higher than 1.0020 near the wavenumber 1060. In essence, the noise amplitude here is only about 0.0015, so there may be local distortion in the data1 spectrum. This analysis The deviation of the result may be relatively large. If possible, the background needs to be re-scanned to obtain a better analysis result; observing the accompanying drawing 5(d), it can be found that the spectral value of the corrected data2 spectrum has almost reached 1.0020 near the wave number 1045 , so there may also be local distortion in the data2 spectrum; observing Figure 5(e), it can be found that the maximum spectral value of the corrected data3 spectrum is only about 1.0010, so data3 is a good spectrum without local distortion, and the calibration model is used for this The reliability of the spectrum analysis is very high. In essence, it can be seen from Fig. 5(b) that within the wavenumber range of 1000 to 1200, the spectra of data1 and data2 are ripples with different frequencies, which itself is a manifestation of local spectral distortion.

Of course, in computer-based intelligent identification, it is impossible to identify by manual observation, and it needs to be identified by software. For this step, it is only necessary to compare the difference between the reconstructed spectrum and the corrected spectrum within the wavenumber range of interest. If the value of the corrected spectrum is greater than the value of the corresponding reconstructed spectrum and greater than the noise level, there may be local distortion. The larger the difference, the greater the distortion. If the difference is too large, the current analysis result will be discarded, and an error will be prompted; if the difference is within a certain allowable range, not enough to have a large impact, continue to work; if there are continuous local irregular distortions in the spectrum, re-scan the background .

4) Decision making

If it can be judged from the spectral analysis results that there may be one or more drugs in a certain monitored physical space, the public security agency will be notified through text messages or the Internet. No.

The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments. It cannot be determined that the specific embodiments of the present invention are limited thereto. Under the circumstances, some simple deduction or replacement can also be made, all of which should be regarded as belonging to the scope of patent protection determined by the submitted claims of the present invention.

Claims (6)

1. An infrared spectrum online monitoring system for taking drugs is characterized in that it comprises some electromagnetic reversing valves, a Fourier transform infrared spectrometer, an air pump and an industrial computer; wherein, the air inlets of some electromagnetic reversing valves are arranged on In the corresponding monitored physical spaces, the air outlets of several electromagnetic reversing valves are respectively connected with the inlets of the gas chambers of the Fourier transform infrared spectrometer. The industrial computer is used to select one of the electromagnetic reversing valves to work at the same time. The air pump is used to pump the gas in the monitored physical space to the gas chamber of the Fourier transform infrared spectrometer, and the Fourier transform infrared spectrometer transmits the absorbance spectral data and transmittance spectral data of the gas in the monitored physical space to the industrial computer; The leaf-transform infrared spectrometer is also used for background scanning of the gas chamber of the spectrometer, and through spectral distortion identification and processing combined with gas chamber switching, the problem of serious interference is solved. 2. An infrared spectrum online monitoring method for drug taking, characterized in that, the method is based on the infrared spectrum online monitoring system for drug taking according to claim 1, comprising the following steps: 1) At the same time, the industrial computer selects one of the electromagnetic reversing valves to work, and the air pump pumps the air in a monitored physical space corresponding to the electromagnetic reversing valve to the Fourier transform infrared spectrometer through the electromagnetic reversing valve The gas chamber conducts online analysis to obtain the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space; 2) The Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space to the industrial computer, and the industrial computer compares the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space with the preset number Compared with the absorption spectrum data of a drug gas, if the absorbance spectrum data of the gas in the monitored physical space is greater than or equal to the preset value of the absorption spectrum of one or more preset drug gases, or if the transmittance spectrum data of the gas in the monitored physical space If it is less than or equal to the preset value of the absorption spectrum of one or more preset drug gases, the industrial computer will send out a warning signal; Absorption spectrum preset value, then the industrial computer selects another electromagnetic reversing valve to work; Among them, Fourier transform infrared spectrometer performs spectral analysis, including two parts: characteristic variable extraction of each drug component and spectral distortion identification and processing. In the process of characteristic variable extraction, a certain spectral line is extracted as a characteristic variable of a certain gas, and a certain The area of a spectrum is used as a characteristic variable, or the combination of multiple spectral line values is extracted as a characteristic variable of different gases. The identification and processing of spectral distortion is combined with spectral analysis, and is identified and processed according to the distortion of the spectrum to reduce the spectral distortion. The impact of distortion on the analysis results. Spectral distortion is divided into baseline regular distortion and irregular distortion. For regular distortion, the spectrum is corrected by shifting and rotating the spectrum. For irregular distortion, this analysis is abandoned and Prompt an error, or give the reliability of the analysis result, if continuous irregular distortion occurs, re-scan the background. 3. An infrared spectrum online monitoring system for taking drugs is characterized in that it includes several air pumps, Fourier transform infrared spectrometers and industrial computers; wherein, the air inlets of some air pumps are arranged in corresponding physical spaces to be monitored Inside, the gas outlets of several air pumps are respectively connected with the gas chamber inlets of the Fourier transform infrared spectrometer. The industrial computer is used to select one of the air pumps to work at the same time. The absorbance spectral data and transmittance spectral data are transmitted to the industrial computer; the Fourier transform infrared spectrometer is also used for background scanning of the gas chamber of the spectrometer, and through spectral distortion identification and processing combined with gas chamber switching, the problem of serious interference is solved. 4. An infrared spectrum online monitoring method for drug abuse, characterized in that the method is based on the infrared spectrum online monitoring system for drug abuse according to claim 3, comprising the following steps: 1) At the same time, the industrial computer selects one of the air pumps to work, and the air pump pumps the air in the monitored physical space to the gas chamber of the Fourier transform infrared spectrometer for online analysis, and obtains the absorbance spectral data of the gas in the monitored physical space and transmittance spectral data; 2) The Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space to the industrial computer, and the industrial computer compares the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space with the preset number Compared with the absorption spectrum data of a drug gas, if the absorbance spectrum data of the gas in the monitored physical space is greater than or equal to the preset value of the absorption spectrum of one or more preset drug gases, or if the transmittance spectrum data of the gas in the monitored physical space If it is less than or equal to the preset value of the absorption spectrum of one or more preset drug gases, the industrial computer will send out a warning signal; Absorption spectrum preset value, then the industrial computer selects another pump to work; Among them, Fourier transform infrared spectrometer performs spectral analysis, including two parts: characteristic variable extraction of each drug component and spectral distortion identification and processing. In the process of characteristic variable extraction, a certain spectral line is extracted as a characteristic variable of a certain gas, and a certain The area of a spectrum is used as a characteristic variable, or the combination of multiple spectral line values is extracted as a characteristic variable of different gases. The identification and processing of spectral distortion is combined with spectral analysis, and is identified and processed according to the distortion of the spectrum to reduce the spectral distortion. The impact of distortion on the analysis results. Spectral distortion is divided into baseline regular distortion and irregular distortion. For regular distortion, the spectrum is corrected by shifting and rotating the spectrum. For irregular distortion, this analysis is abandoned and Prompt an error, or give the reliability of the analysis result, if continuous irregular distortion occurs, re-scan the background. 5. An infrared spectrum on-line monitoring system for taking drugs is characterized in that it includes several groups of optical fiber shunts and couplers, Fourier transform infrared spectrometers and industrial computers; wherein, several groups of optical fiber shunts and couplers are arranged in corresponding In several monitored physical spaces, each group of optical fiber splitters and couplers couples the interference light absorbed in the corresponding monitored physical space to the light source of the Fourier transform infrared spectrometer, and the industrial computer is used at the same time Select one of the optical fiber splitters to work with the coupler, and the Fourier transform infrared spectrometer will transmit the absorbance spectral data and transmittance spectral data of the monitored physical space to the industrial computer; the Fourier transform infrared spectrometer is also used for the gas chamber background of the spectrometer Scanning, and through spectral distortion identification and processing combined with gas chamber switching, solves the problem of serious interference. 6. An infrared spectrum online monitoring method for drug abuse, characterized in that the method is based on the infrared spectrum online monitoring system for drug abuse according to claim 5, comprising the following steps: 1) At the same time, the industrial computer selects a group of optical fiber splitters and couplers to work, and the group of optical fiber splitters and couplers couples the absorbed interference light in the monitored physical space to the light source of the Fourier transform infrared spectrometer for online analysis , to obtain the absorbance spectrum data and transmittance spectrum data of the monitored physical space gas; 2) The Fourier transform infrared spectrometer transmits the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space to the industrial computer, and the industrial computer compares the absorbance spectrum data and transmittance spectrum data of the gas in the monitored physical space with the preset number Compared with the absorption spectrum data of a drug gas, if the absorbance spectrum data of the gas in the monitored physical space is greater than or equal to the preset value of the absorption spectrum of one or more preset drug gases, or if the transmittance spectrum data of the gas in the monitored physical space If it is less than or equal to the preset value of the absorption spectrum of one or more preset drug gases, the industrial computer will send out a warning signal; Absorption spectrum preset value, then the industrial computer selects another group of optical fiber branch and coupler to work; Among them, Fourier transform infrared spectrometer performs spectral analysis, including two parts: characteristic variable extraction of each drug component and spectral distortion identification and processing. In the process of characteristic variable extraction, a certain spectral line is extracted as a characteristic variable of a certain gas, and a certain The area of a spectrum is used as a characteristic variable, or the combination of multiple spectral line values is extracted as a characteristic variable of different gases. The identification and processing of spectral distortion is combined with spectral analysis, and is identified and processed according to the distortion of the spectrum to reduce the spectral distortion. The impact of distortion on the analysis results. Spectral distortion is divided into baseline regular distortion and irregular distortion. For regular distortion, the spectrum is corrected by shifting and rotating the spectrum. For irregular distortion, this analysis is abandoned and Prompt an error, or give the reliability of the analysis result, if continuous irregular distortion occurs, re-scan the background.