Industrial waste gas detection method and system based on gas chromatography

Abstract

The invention discloses an industrial waste gas detection method and system based on gas chromatography, and the method comprises the following steps: S1, building an environment temperature difference chromatogram correction model in advance, and correcting a chromatogram obtained by a gas chromatograph by the model according to the temperature difference between an actual test environment and a standard test environment; s2, industrial waste gas to be detected is detected through a gas chromatograph, after an original chromatogram is obtained, correction is conducted through the environment temperature difference chromatogram correction model, and a corrected chromatogram is obtained; s3, performing peak searching treatment on the corrected chromatogram through a rapid peak searching model to obtain all chromatographic peaks; and S4, analyzing according to the chromatographic peak obtained in the step S3 to obtain components in the industrial waste gas to be detected and the content of each component. The industrial waste gas can be rapidly and accurately detected, the influence of the environment temperature on the test result can be remarkably reduced, and the industrial waste gas detection device has a very good application prospect.

Description

Technical Field

[0001] The present invention relates to the field of gas detection, and particularly to an industrial waste gas detection method and system based on gas chromatography. Background Art

[0002] The existing detection of industrial waste gas generally determines whether the pollutant content of the industrial waste gas meets the emission standard by taking regular samples of the industrial waste gas or directly introducing it into the detection device for detection, and then determines whether it can be discharged according to the emission standard.

[0003] Gas chromatography detection is widely used in gas detection due to its high precision. Its measurement principle is that if a certain component exists in the gas entering the instrument, it will peak at the corresponding time point, and the concentration of the component can be calculated through the peak area size, thus realizing qualitative and quantitative analysis. For example, an industrial waste gas detection device with remote control function disclosed in Patent CN112798703B, a gas detection device and method based on gas chromatography technology in CN113640439A, a method for detecting gas in public places based on gas chromatography in CN119246711A, etc.

[0004] The chromatographic column is one of the key components in gas chromatography and is greatly affected by changes in environmental temperature. Different environmental temperatures result in deviations in the retention time of substances. The shift of the substance retention time will be directly reflected in the obtained chromatogram, thus affecting the results of qualitative and quantitative analysis using the chromatogram. At the same time, when performing quantitative analysis, a standard curve characterizing the relationship between the chromatographic peak area and the component concentration is usually constructed at the standard environmental temperature. When the environmental temperature is non-standard, using this standard curve for quantitative calculation is likely to further increase the influence of the environmental temperature on the quantitative analysis results. By adjusting the environmental temperature to the standard environmental temperature and then performing the measurement, the above-mentioned influence can be avoided, but in many cases, there are problems such as great implementation difficulty, high cost (such as using temperature control equipment), or inconvenient implementation (such as during on-site testing). For example, the chromatographic analysis method based on peak search technology disclosed in Patent CN115950988A provides a solution to reduce the influence of environmental temperature changes on the test, but its method is relatively complex and is mainly used for peak search and positioning. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an industrial waste gas detection method and system based on gas chromatography in view of the above deficiencies in the prior art.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is: In the first aspect of the present invention, an industrial waste gas detection method based on gas chromatography is provided, including the following steps:

[0007] S1. Pre - construct an environmental temperature difference chromatogram correction model, which corrects the chromatogram obtained by a gas chromatograph according to the temperature difference between the actual test environment and the standard test environment;

[0008] S2. Use a gas chromatograph to detect the industrial waste gas to be detected. After obtaining the original chromatogram, correct it through the environmental temperature difference chromatogram correction model to obtain the corrected chromatogram;

[0009] S3. Perform peak searching on the corrected chromatogram through a fast peak searching model to obtain all chromatographic peaks;

[0010] S4. Analyze the components and the content of each component in the industrial waste gas to be detected based on the chromatographic peaks obtained in step S3.

[0011] Preferably, the environmental temperature difference chromatogram correction model is constructed by the following method:

[0012] S1 - 1. Pre - obtain each waste gas component when the industrial waste gas to be detected contains the most waste gas components in the target scenario;

[0013] S1 - 2. Construct a training data set:

[0014] Perform gas chromatographic detection on a single - component standard sample containing only one kind of waste gas component at the standard environmental temperature, and then perform gas chromatographic detection again after only changing the environmental temperature. Use all the obtained chromatograms to construct a single - component training data set S 1 ;

[0015] Perform gas chromatographic detection on a multi - component standard sample containing at least two kinds of waste gas components at the standard environmental temperature, and then perform gas chromatographic detection again after only changing the environmental temperature. Use all the obtained chromatograms to construct a multi - component training data set S 2 ;

[0016] Perform gas chromatographic detection on a full - component standard sample containing all waste gas components at the standard environmental temperature, and then perform gas chromatographic detection again after only changing the environmental temperature. Use all the obtained chromatograms to construct a full - component training data set S 3 ;

[0017] S1 - 3. Model training:

[0018] Use the single - component training data set S 1 、multi - component training data set S 2 、full - component training data set S 3 ; to train the CNN convolutional neural network in sequence. After the training is completed, the environmental temperature difference chromatogram correction model is obtained.

[0019] Preferably, step S1-2 specifically includes:

[0020] S1-2-1. Construct a single-component training dataset S 1 :

[0021] Under the standard ambient temperature, use a gas chromatograph to obtain a single-component standard sample containing only a single waste gas component for measurement, obtain a chromatogram, denoted as the standard single-component chromatogram; test each waste gas component one by one, and respectively obtain the standard single-component chromatograms of each single-component standard sample;

[0022] For any single-component standard sample, only change the ambient temperature to obtain chromatograms at different actual ambient temperatures, denoted as the measured chromatogram PC BΔT , denote ΔT = actual ambient temperature - standard ambient temperature; for any single-component standard sample C B , combine ΔT, the measured chromatogram PC B corresponding to C at ΔT BΔT and the standard single-component chromatogram PC B of C B into a single-component training data s 1 , combine all the obtained single-component training data s 1 to construct a single-component training dataset S 1 ;

[0023] S1-2-2. Construct a multi-component training dataset S 2 :

[0024] Under the standard ambient temperature, use a gas chromatograph to obtain a multi-component standard sample containing at least two waste gas components for measurement, obtain a chromatogram, denoted as the standard multi-component chromatogram; then change the types and quantities of the waste gas components in the standard sample, and conduct tests respectively to obtain the standard multi-component chromatograms of each multi-component standard sample;

[0025] For any multi-component standard sample, only change the ambient temperature to obtain chromatograms at different actual ambient temperatures, denoted as the measured chromatogram PCd BΔT , denote ΔT = actual ambient temperature - standard ambient temperature; for any multi-component standard sample Cd B , combine ΔT, the measured chromatogram PCd B corresponding to Cd at ΔT BΔT and the standard multi-component chromatogram PCd B of Cd B into a multi-component training data s 2 , combine all the obtained multi-component training data s 2 to construct a multi-component training dataset S 2 ;

[0026] S1-2-3. Construct the full-component training dataset S 3 :

[0027] Under the standard environmental temperature, use a gas chromatograph to obtain a full-component standard sample containing all waste gas components for measurement, obtain a chromatogram, denoted as the standard full-component chromatogram; only change the concentration ratio of the waste gas components in the full-component standard sample, and obtain the standard full-component chromatograms of each full-component standard sample with different concentration ratios;

[0028] For any full-component standard sample, only change the environmental temperature to obtain chromatograms at different actual environmental temperatures, denoted as the measured chromatogram PCa BΔT , denote ΔT = actual environmental temperature - standard environmental temperature; for any full-component standard sample Ca B , combine ΔT, the measured chromatogram PCa of Ca under ΔT B and the standard full-component chromatogram PCa of Ca BΔT into a full-component training data s B B 3 , combine all the obtained full-component training data s 3 to construct the full-component training dataset S 3 ; 1 ;

[0029] Among them, the standard environmental temperature is 20 - 25 °C.

[0030] Preferably, step S1-3 specifically includes:

[0031] S1-3-1. Use the single-component training dataset S 1 , with ΔT, PC BΔT as the input and PC B as the output to train the CNN convolutional neural network. After the training is completed, obtain the first-order correction model;

[0032] S1-3-2. Use the multi-component training dataset S 2 , with ΔT, PCd BΔT as the input and PCd B as the output to train the first-order correction model. After the training is completed, obtain the second-order correction model;

[0033] S1-3-3. Use the full-component training dataset S 3 , with ΔT, PCa BΔT as the input and PCa B as the output to train the second-order correction model. After the training is completed, obtain the final environmental temperature difference chromatogram correction model.

[0034] Preferably, the method for peak searching by the fast peak searching model includes the following steps:

[0035] S3-1. Obtain all peak vertices in the calibrated chromatogram. The set of all peak vertices is denoted as Q p ;

[0036] S3-2. Obtain the peak start point of the peak corresponding to each peak vertex in Q p ;

[0037] S3-3. Obtain the peak end point of the peak corresponding to each peak vertex in Q p ;

[0038] S3-4. According to the obtained peak vertices, the peak start point and peak end point corresponding to each peak vertex, obtain all chromatographic peaks and index them on the chromatographic curve of the calibrated chromatogram, so as to obtain the processed chromatogram with indexed chromatographic peaks.

[0039] Preferably, the method for peak searching by the fast peak searching model includes the following steps:

[0040] S3-1. Obtain peak vertices:

[0041] S3-1-1. On the chromatographic curve of the calibrated chromatogram, starting from the origin, from left to right, take J consecutive data points as a data unit. For the last data unit, take the actual remaining data points as a data unit. Denote the number of data points in the last data unit as J', and the total number of data units as m. Then J'≤J, and the total number of data points N = m*(J - 1)+J';

[0042] S3-1-2. At the same time, adopt the following steps for all data units to obtain peak vertices:

[0043] For a data unit, for any other data point P in the data unit except the left end point and the right end point h , when the following conditions a and b are satisfied, then take the data point P h as a peak vertex:

[0044] a. The signal intensity of the data point P h is simultaneously greater than the signal intensities of the first data point on its left and the first data point on its right;

[0045] b. There are at least n h consecutive data points with decreasing signal intensities from the data point P T to the left, and at least n h consecutive data points with decreasing signal intensities from the data point P T to the right;

[0046] Traverse all the data points in the data unit except the left endpoint and the right endpoint, and construct all the peak vertices obtained into the peak vertex set Q P1 ;

[0047] S3-1-3. On the chromatogram curve of the corrected chromatogram, for the data points of the left endpoint and the right endpoint of all data units, when the above conditions a and b are satisfied simultaneously, take the current data point as the peak vertex, and construct all the peak vertices obtained into the peak vertex set Q P2 ;

[0048] S3-1-4. Combine Q P1 and Q P2 to obtain all the peak vertex sets Q p of the corrected chromatogram, Q p =Q P1 +Q P2 ;

[0049] S3-2. Obtain the peak start points of each peak corresponding to the peak vertices in the peak vertex set Q p :

[0050] For the peak vertex P p in Q n , calculate the slope of the line connecting adjacent two data points sequentially from P n to the left, denoted as: the slope between P n and P n-1 is k n , the slope between P n-1 and P n-2 is k n-1 ,..., the slope between P n-(m-1) and P n-m is k n-(m-1) , the slope between P n-m and P n-(m+1) is k n-m ;

[0051] When the first data point that can satisfy the following condition c or d appears, take this data point P n-m as the peak start point of the peak corresponding to the peak vertex P n :

[0052] c. k n-(m-1) >0, and k n-m ≤0

[0053] d. k n-(m-1) >0, k n-m >0, and k n-(m-1) -k n-m ≤Δk T1 ; Δk T1is a preset slope change threshold value, which is a constant not less than 0;

[0054] S3-3. Obtain the set Q of peak vertices p The peak end points of each peak corresponding to the peak vertices in:

[0055] For Q p the peak vertex P in n , starting from P n calculate the slopes of the lines connecting adjacent two data points in sequence to the right, denoted as: the slope between P n and P n+1 is k n+1 , the slope between P n+1 and P n+2 is k n+2 ,..., the slope between P n+(m'-1) and P n+m' is k n+m' , the slope between P n+m' and P n+(m'+1) is k n+(m'+1) ;

[0056] When the first data point that can meet the following condition e or f appears, take this data point P n+m' as the peak end point of the peak corresponding to the peak vertex P n :

[0057] e. k n+m' <0, and k n+(m'+1) ≥0;

[0058] f. k n+m' <0, k n+(m'+1) <0, and |k n+m' |-|k n+(m'+1) |≤Δk T2 , Δk T2 is a preset slope change threshold value, which is a constant not less than 0;

[0059] S3-4. According to the obtained peak vertices, the peak start points and peak end points corresponding to each peak vertex, obtain all chromatographic peaks and index them on the chromatographic curve of the corrected chromatogram, so as to obtain the processed chromatogram containing chromatographic peaks.

[0060] Preferably, where J = 5 - 50, n T = 2 - 5.

[0061] Preferably, where 0 ≤ Δk T1 ≤ 0.15, 0 ≤ Δk T2 ≤ 0.15.

[0062] Preferably, step S4 is specifically:

[0063] S4-1. Compare the processed chromatogram with the standard full-component chromatogram to obtain the components corresponding to each chromatographic peak in the processed chromatogram:

[0064] For any chromatographic peak R in the processed chromatogram i , denote the peak apex of chromatographic peak R i as P i , obtain the abscissa t i of P i . In the standard full-component chromatogram, within the abscissa range of (t i - 0.5d min ) to (t i + 0.5d min ), obtain the peak apex P Bi corresponding to the minimum absolute value of the difference between the abscissa t i and t Bi . Take the component C Bi where P i is located as the component corresponding to chromatographic peak R i ;

[0065] When there is no chromatographic peak in the standard full-component chromatogram within the abscissa range of (t i - 0.5d min ) to (t i + 0.5d min ), manually confirm the component corresponding to chromatographic peak R i ; where d min is the minimum value of the distance between the peak apices of two adjacent chromatographic peaks on the abscissa in the standard full-component chromatogram;

[0066] S4-2. According to the areas of the chromatographic peaks in the processed chromatogram, combined with the standard curve pre-constructed to represent the relationship between the chromatographic peak areas and component concentrations in the standard full-component chromatogram, calculate the concentrations of the components corresponding to each chromatographic peak.

[0067] In the second aspect of the present invention, there is provided an industrial waste gas detection system based on gas chromatography, which uses the method described above for industrial waste gas detection. The system includes:

[0068] A gas chromatograph, which is used to detect the industrial waste gas to be detected and obtain the original chromatogram;

[0069] An environmental temperature difference chromatogram correction model, which obtains the corrected chromatogram using the method in step S2;

[0070] A fast peak search model, which performs peak search processing on the corrected chromatogram using the method in step S3 to obtain all chromatographic peaks;

[0071] and a chromatographic peak analysis module, which analyzes the obtained chromatographic peaks according to the method in step S4 to obtain the components in the industrial waste gas to be detected and the content of each component.

[0072] The beneficial effects of the present invention are:

[0073] The present invention provides an industrial waste gas detection method based on gas chromatography, which can realize the rapid and accurate detection of industrial waste gas, and can significantly reduce the influence of environmental temperature on the test results, and has good application prospects;

[0074] The present invention uses an environmental temperature difference chromatogram correction model constructed based on a machine learning algorithm to correct the chromatogram collected at a non-standard environmental temperature to the chromatogram at the standard temperature. On the one hand, it reduces the analysis error caused by the shift of the retention time of substances due to temperature changes, and on the other hand, it also reduces the error of using the standard curve constructed at the standard temperature for quantitative analysis at non-standard temperatures, thereby improving the accuracy of subsequent qualitative and quantitative analysis based on the chromatogram;

[0075] The rapid peak search model of the present invention can quickly obtain the peak vertex, peak start point, and peak end point, and can achieve the effect of rapid and accurate peak search; after peak search, the processed chromatogram containing chromatographic peaks is compared with the standard full-component chromatogram obtained during the construction of the environmental temperature difference chromatogram correction model, and the components and concentrations of each chromatographic peak corresponding to the processed chromatogram can be quickly obtained, and finally the rapid qualitative and quantitative analysis of industrial waste gas can be realized. Description of the Drawings

[0076] Figure 1 is a flowchart of the industrial waste gas detection method based on gas chromatography of the present invention;

[0077] Figure 2 is a flowchart of the construction method of the environmental temperature difference chromatogram correction model of the present invention;

[0078] Figure 3 is a specific flowchart of step S1-2 of the present invention;

[0079] Figure 4 is a specific flowchart of step S1-3 of the present invention;

[0080] Figure 5 is a flowchart of the method for peak search processing by the rapid peak search model of the present invention;

[0081] Figure 6 is a specific flowchart of step S4 of the present invention;

[0082] Figure 7 is the standard chromatogram P obtained at an environmental temperature of 25°C in the test example of the present invention 1 ;

[0083] Figure 8 The measured chromatogram P at an environmental temperature of 5°C obtained in the test example of the present invention 2 ;

[0084] Figure 9 The corrected chromatogram P obtained in the test example of the present invention 3 ;

[0085] Figure 10 The schematic of the area represented by S in the test example of the present invention P1 ;

[0086] Figure 11 The schematic of the area represented by ΔS in the test example of the present invention

[0087] Figure 12 The principle block diagram of the industrial waste gas detection system based on gas chromatography in Example 2 of the present invention Detailed implementation manners

[0088] The following further describes the present invention in detail with reference to the embodiments, so that those skilled in the art can implement it according to the description in the specification.

[0089] It should be understood that the terms such as "having", "comprising" and "including" used herein do not exclude the presence or addition of one or more other elements or their combinations.

[0090] Example 1

[0091] An industrial waste gas detection method based on gas chromatography, comprising the following steps:

[0092] S1. Pre-construct an environmental temperature difference chromatogram correction model, which corrects the chromatogram obtained by a gas chromatograph according to the temperature difference between the actual test environment and the standard test environment;

[0093] S2. Detect the industrial waste gas to be detected by a gas chromatograph. After obtaining the original chromatogram, correct it through the environmental temperature difference chromatogram correction model to obtain a corrected chromatogram;

[0094] S3. Perform peak searching on the corrected chromatogram through a fast peak searching model to obtain all chromatographic peaks;

[0095] S4. Analyze the components and the content of each component in the industrial waste gas to be detected based on the chromatographic peaks obtained in step S3.

[0096] In this embodiment, the environmental temperature difference chromatogram correction model is constructed by the following method:

[0097] S1-1. Pre-acquire each exhaust gas component under the condition of the most exhaust gas components contained in the industrial exhaust gas to be detected in the target scenario; that is, for the main application target scenarios, manually obtain in advance all possible components contained in the industrial exhaust gas generated in these target scenarios, and these components can be respectively denoted as C 1 , C 2 , ..., C L , and any one of the exhaust gas components can be denoted as C i . Therefore, the data used to construct the environmental temperature chromatogram correction model should be determined according to the actual detection target.

[0098] S1-2. Construct a training data set:

[0099] S1-2-1. Construct a single-component training data set S 1 :

[0100] At the standard environmental temperature, use a gas chromatograph to obtain a single-component standard sample containing only a single exhaust gas component for determination, and obtain a chromatogram, denoted as the standard single-component chromatogram; test each exhaust gas component one by one, and respectively obtain the standard single-component chromatograms of each single-component standard sample;

[0101] For any single-component standard sample, only change the environmental temperature to obtain chromatograms at different actual environmental temperatures, denoted as the measured chromatogram PC BΔT , and denote ΔT = actual environmental temperature - standard environmental temperature; for any single-component standard sample C B , combine ΔT, the measured chromatogram PC B corresponding to C BΔT and the standard single-component chromatogram PC B of C B into a single-component training data s 1 , and combine all the obtained single-component training data s 1 to construct a single-component training data set S 1 ;

[0102] In this embodiment, in order to ensure the coverage of the training data and improve the model accuracy, the single-component standard samples corresponding to each exhaust gas component need to collect measured chromatograms at multiple different temperatures and construct them into single-component training data.

[0103] S1-2-2. Construct a multi-component training data set S 2 :

[0104] Under standard ambient temperature, a gas chromatograph is used to obtain a multi-component standard sample containing at least two waste gas components for measurement, and a chromatogram is obtained, denoted as the standard multi-component chromatogram; then, the types and quantities of the waste gas components in the standard sample are changed, and tests are carried out separately to obtain the standard multi-component chromatograms of each multi-component standard sample;

[0105] For any multi-component standard sample, only the ambient temperature is changed to obtain chromatograms at different actual ambient temperatures, denoted as the measured chromatogram PCd BΔT , denote ΔT = actual ambient temperature - standard ambient temperature; for any multi-component standard sample Cd B , combine ΔT, the measured chromatogram PCd of Cd under ΔT B and the standard multi-component chromatogram PCd of Cd BΔT into a multi-component training data s B ; combine all the obtained multi-component training data s B to construct a multi-component training data set S 2 ; 2 ; 2 ;

[0106] In this embodiment, in order to ensure the coverage rate of the training data and improve the model accuracy, for each waste gas component, some other components need to be added to it to form a multi-component standard sample, and measured chromatograms are collected at multiple different temperatures as multi-component training data. The richer the combination method of the multi-component standard samples, the more conducive to the enrichment of the training data. For example, it includes multi-component standard samples of several 2-component, 3-component, 4-component,..., (L - 1)-component. In this embodiment, the number of multi-component standard samples used to construct the training data is at least 200*L.

[0107] S1-2-3. Construct the full-component training data set S 3 :

[0108] Under standard ambient temperature, a gas chromatograph is used to obtain a full-component standard sample containing all waste gas components for measurement, and a chromatogram is obtained, denoted as the standard full-component chromatogram; only the concentration ratio of the waste gas components in the full-component standard sample is changed to obtain the standard full-component chromatograms of each different concentration ratio of the full-component standard sample;

[0109] For any full-component standard sample, only the ambient temperature is changed to obtain chromatograms at different actual ambient temperatures, denoted as the measured chromatogram PCa BΔT , denote ΔT = actual ambient temperature - standard ambient temperature; for any full-component standard sample Ca B , combine ΔT, the measured chromatogram PCa of Ca under ΔT B and the standard full-component chromatogram PCa of CaBΔT and Ca B The standard full-component chromatogram PCa B is combined into a full-component training data s 3 . All the obtained full-component training data s 3 are combined to construct a full-component training data set S 3 ;

[0110] Among them, the standard ambient temperature is 25°C. When constructing the training data set, the range of the actual ambient temperature used is 0 to 50°C, specifically including 0°C, 1°C, 2°C, 3°C, 4°C,..., 48°C, 49°C, 50°C. That is, data is collected at temperature intervals of 1°C to provide training data with an appropriate temperature interval span and rich temperature data, so as to ensure the accuracy of the model.

[0111] S1-3. Model training:

[0112] S1-3-1. Using the single-component training data set S 1 , with ΔT and PC BΔT as the input and PC B as the output to train the CNN convolutional neural network. After the training is completed, a first-order calibration model is obtained;

[0113] S1-3-2. Using the multi-component training data set S 2 , with ΔT and PCd BΔT as the input and PCd B as the output to train the first-order calibration model. After the training is completed, a second-order calibration model is obtained;

[0114] S1-3-3. Using the full-component training data set S 3 , with ΔT and PCa BΔT as the input and PCa B as the output to train the second-order calibration model. After the training is completed, the final ambient temperature chromatogram calibration model is obtained.

[0115] The change of ambient temperature will affect the retention time of substances in the chromatograph. The shift of the substance retention time will be directly reflected in the obtained chromatogram, thus affecting the results of qualitative and quantitative analysis using the chromatogram. At the same time, when performing quantitative analysis, a standard curve representing the relationship between the chromatographic peak area and the component concentration is usually constructed at the standard ambient temperature. When the ambient temperature is non-standard, using this standard curve for quantitative calculation is likely to further increase the influence of the ambient temperature on the quantitative analysis results. By adjusting the ambient temperature to the standard ambient temperature and then performing the measurement, the above-mentioned influence can be avoided. However, in many cases, there are problems such as great implementation difficulty, high cost (such as using temperature control equipment), or inconvenient implementation (such as during on-site testing).

[0116] A different approach is adopted in the present invention: by using a model constructed based on a machine learning algorithm, the chromatogram collected at a non-standard ambient temperature is corrected to a chromatogram at the standard temperature. On the one hand, it reduces the analysis error caused by the shift of the retention time of substances due to temperature changes. On the other hand, it also reduces the error of using the standard curve constructed at the standard temperature for quantitative analysis at non-standard temperatures, thereby improving the accuracy of subsequent qualitative and quantitative analysis based on the chromatogram.

[0117] Furthermore, during the model training process of the present invention, first, a single-component training dataset S 1 is used to train the CNN convolutional neural network, enabling the CNN to learn the transformation relationship of the chromatogram under a single exhaust gas component corrected to the standard temperature at different temperatures; then, a multi-component training dataset S 2 is used for training, enabling the initially trained first-order correction model to further learn the transformation relationship of the chromatograms under multiple exhaust gas components corrected to the standard temperature at different temperatures; finally, a full-component training dataset S 3 is used for training, enabling the further trained second-order correction model to further learn the transformation relationship of the chromatograms containing all exhaust gas components but with different concentrations corrected to the standard temperature at different temperatures; by using a training dataset from simple to gradually complex for hierarchical training of the model, it can better ensure the correction ability and correction accuracy of the obtained environmental temperature difference chromatogram correction model.

[0118] In this embodiment, the method for the quick peak search model to perform peak search processing includes the following steps:

[0119] S3-1. Obtain the peak vertex:

[0120] S3-1-1. On the chromatogram curve of the corrected chromatogram, starting from the origin, from left to right, taking J consecutive data points as a data unit, for the last data unit, taking the actual remaining data points as a data unit, recording the number of data points in the last data unit as J', the total number of data units as m, then J'≤J, and the total number of data points N = m*(J - 1)+J';

[0121] Among them, J can be selected according to the actual situation and requirements. When the value of J is appropriate, it has high precision and does not increase too much computational complexity. For example, J = 5 - 50. In this embodiment, J = 10 is selected.

[0122] S3-1-2. Simultaneously use the following steps for all data units to obtain the peak vertex:

[0123] For a data unit, for any other data point P in the data unit excluding the left endpoint and the right endpoint h, when the following conditions a and b are satisfied, then the data point P h is used as the peak vertex:

[0124] a. The signal intensity of the data point P h is greater than the signal intensities of the first data point on its left and the first data point on its right at the same time;

[0125] b. There are at least n consecutive data points with decreasing signal intensities to the left of the data point P h , and there are at least n consecutive data points with decreasing signal intensities to the right of the data point P T ; h From the data point P T to the right, there are at least n consecutive data points with decreasing signal intensities;

[0126] Satisfying condition a indicates that P h is higher than the signal intensities on both the left and the right. Satisfying b indicates that the curve on the left side of P h has an upward trend in a certain range, and the curve on the right side has a downward trend in a certain range. Therefore, when both conditions a and b are satisfied, it can be considered that P h is located at the peak vertex; among them, n T is taken according to the actual situation and requirements. The relatively larger the value of n T , the higher the recognition accuracy. However, when it is too large, there will be a certain risk of missed recognition. Its general value can be n T = 2 - 5. In this embodiment, n T is selected as 3.

[0127] Traverse all data points in the data unit except the left endpoint and the right endpoint, and construct all the obtained peak vertices into a peak vertex set Q P1 ;

[0128] S3-1-3. On the chromatographic curve of the corrected chromatogram, for the data points at the left endpoints and right endpoints of all data units, when both of the above conditions a and b are satisfied, the current data point is used as the peak vertex, and all the obtained peak vertices are constructed into a peak vertex set Q P2 ;

[0129] S3-1-4. Combine Q P1 and Q P2 to obtain all the peak vertex sets Q p of the corrected chromatogram, Q p = Q P1 + Q P2 ;

[0130] S3-2. Obtain the peak start points of each peak corresponding to the peak vertices in the peak vertex set Q p :

[0131] For Q pThe peak vertex P in n , from P n , calculate the slope of the line connecting adjacent two data points in turn to the left, denoted as: the slope between P n and P n-1 is k n , the slope between P n-1 and P n-2 is k n-1 ,..., the slope between P n-(m-1) and P n-m is k n-(m-1) , the slope between P n-m and P n-(m+1) is k n-m ;

[0132] When the first data point that can meet the following condition c or d appears, use this data point P n-m as the peak starting point of the peak corresponding to the peak vertex P n :

[0133] c, k n-(m-1) > 0, and k n-m ≤ 0;

[0134] d, k n-(m-1) > 0, k n-m > 0, and k n-(m-1) - k n-m ≤ Δk T1 ; Δk T1 is a preset slope change threshold, which is a constant not less than 0; for example, 0 ≤ Δk T1 ≤ 0.15, in this embodiment, Δk T1 = 0.12;

[0135] When condition c is met, it indicates that from the peak vertex P n to the left, the slope of the curve gradually decreases from a positive value, and from P n-(m-1) to P n-(m+1) , it changes from positive to negative, indicating that the point P n-(m-1) between P n-(m+1) and P n-m is the starting point where the slope increases, that is, the peak starting point on the left side of the peak;

[0136] When condition d is met, it indicates that from the peak vertex P n to the left, the slope of the curve gradually decreases from a positive value, and from P n-(m-1) to P n-(m+1) , the degree of slope decline can be considered close to zero, so the point P n-(m-1) between P n-(m+1) and P n-mIt can be regarded as the starting point where the slope starts to increase significantly, that is, the peak starting point on the left side of the peak; through the above method, the peak starting point can be quickly and accurately identified;

[0137] S3-3. Obtain the set Q of peak vertices p The peak end points of each peak corresponding to the peak vertices:

[0138] For the peak vertex P in Q p From P n Calculate the slope of the line connecting adjacent two data points in sequence to the right, denoted as: the slope between P n and P n is k n+1 , the slope between P n+1 and P n+1 is k n+2 ,..., the slope between P n+2 and P n+(m'-1) is k n+m' , the slope between P n+m' and P n+m' is k n+(m'+1) , the slope between P n+(m'+1) ;

[0139] When the first data point that can satisfy the following condition e or f appears, take this data point P n+m' as the peak end point of the peak corresponding to the peak vertex P n :

[0140] e. k n+m' <0, and k n+(m'+1) ≥0;

[0141] f. k n+m' <0, k n+(m'+1) <0, and |k n+m' |-|k n+(m'+1) |≤Δk T2 , Δk T2 is a preset slope change threshold, which is a non - negative constant; for example, 0≤Δk T2 ≤0.15, in this embodiment, Δk T2 =0.12;

[0142] When condition e is satisfied, it means that from the peak vertex P n to the right, the slope of the curve gradually increases from a negative value, and changes from negative to positive from P n+(m'-1) to P n+(m'+1) , indicating that the point P n+(m'-1) between P n+(m'+1) and P n+m' is the termination point where the slope increases, that is, the peak end point on the right side of the peak;

[0143] When condition d is satisfied, it means that from the peak vertex Pn To the right, the slope of the curve gradually increases from a negative value and from P n+(m'-1) to P n+(m'+1) When the degree of increase in the slope can be considered close to zero, so P n+(m'-1) and P n+(m'+1) The point P between n+m' can be considered as the termination point of the increasing slope, that is, the peak end point on the right side of the peak; the peak end point can be quickly and accurately identified through the above method;

[0144] S3-4. According to the obtained peak vertices, the peak start points and peak end points corresponding to each peak vertex, all chromatographic peaks are obtained and indexed on the chromatographic curve of the corrected chromatogram, so as to obtain the processed chromatogram containing chromatographic peaks.

[0145] In this embodiment, step S4 is specifically as follows:

[0146] S4-1. Compare the processed chromatogram with the standard full-component chromatogram to obtain the components corresponding to each chromatographic peak in the processed chromatogram:

[0147] For any chromatographic peak R in the processed chromatogram i , denote the peak vertex of chromatographic peak R i as P i , obtain the abscissa t i of P i . In the standard full-component chromatogram, within the abscissa range of (t i -0.5d min )~(t i +0.5d min ), obtain the peak vertex P Bi corresponding to the minimum absolute value of the difference between the abscissa t i and t Bi . Take the component C Bi where P i is located as the component corresponding to chromatographic peak R i ;

[0148] The elution time of any component in the measured chromatogram (i.e., the abscissa in the chromatogram) will be the same as or close to the elution time of this component in the standard full-component chromatogram (especially after correction by the environmental temperature difference chromatogram correction model in the present invention, it will be closer). Therefore, for any chromatographic peak to be determined, by comparing the measured chromatogram with the standard full-component chromatogram, in the standard full-component chromatogram, search for the peak vertex closest to the abscissa of the peak vertex of the chromatographic peak to be determined within a certain range. Then, the component of the chromatographic peak corresponding to this peak vertex is the same component corresponding to the chromatographic peak to be determined, so that the components corresponding to each chromatographic peak in the processed chromatogram can be quickly obtained.

[0149] When in the standard full-component chromatogram, within the horizontal coordinate range of (t i - 0.5d min ) to (t i + 0.5d min ), there is no chromatographic peak, the component corresponding to the chromatographic peak R i is confirmed manually.

[0150] Among them, d min is the minimum value of the distance between the peak vertices of two adjacent chromatographic peaks on the horizontal coordinate in the standard full-component chromatogram; the search range is determined by d min , and the horizontal coordinate search range is located at (t i - 0.5d min ) to (t i + 0.5d min ), which can ensure that there will not be two chromatographic peaks in this range at the same time.

[0151] S4-2. According to the area of the chromatographic peak in the processed chromatogram, combined with the standard curve pre-constructed to represent the relationship between the chromatographic peak area and the component concentration in the standard full-component chromatogram, the concentration of the component corresponding to each chromatographic peak is calculated. Among them, the area of the chromatographic peak is positively correlated with the component concentration. Pre-constructing the standard curve and then analyzing the component concentration using the chromatographic peak area is a common method in the field, and the present invention does not limit it, nor will the specific scheme be elaborated in this embodiment.

[0152] Test Example

[0153] A mixed gas containing benzene, toluene, and styrene is used as the test sample to simulate industrial waste gas, and gas chromatography (Agilent 7890A) is used for chromatographic detection.

[0154] 1. Test on the calibration effect of the environmental temperature difference chromatogram calibration model:

[0155] First, detect the standard chromatogram P 1 at an environmental temperature of 25 °C. As Figure 7 shown, the chromatographic peaks of benzene, toluene, and styrene are successively shown as marked ①, ②, and ③ in the figure;

[0156] Then, detect the measured chromatogram P 2 at an environmental temperature of 5 °C. As Figure 8 shown; the measured chromatogram P 2 is calibrated using the environmental temperature difference chromatogram calibration model to obtain the calibrated chromatogram P 3 . As Figure 9 shown.

[0157] Compare the difference coefficient ε of P 1 and P 2 ​Δs , and P 1 and P 3 The coefficient of difference ε Δs :

[0158] Define the coefficient of difference ε Δs The calculation formula is as follows:

[0159]

[0160] where ε Δs The smaller the value, the smaller the difference; S P1 represents the area of the closed region surrounded by the chromatographic curve in the standard chromatogram, the abscissa, and the vertical perpendicular lines at the left and right endpoints of the chromatographic curve. Refer to Figure 10 ;

[0161] ΔS represents the cross-sectional area. For the coefficient of difference between P 1 and P 2 , when P 1 and P 2 are plotted in the same chromatogram, ΔS is the area of the closed region formed by the intersection of the chromatographic curve in P 1 and the chromatographic curve in P 2 . Refer to Figure 11 ; For the coefficient of difference between P 1 and P 3 , when P 1 and P 3 are plotted in the same chromatogram, ΔS is the area of the closed region between the chromatographic curve in P 1 and the chromatographic curve in P 3 .

[0162] As a further comparison, the measured chromatogram P 2 is corrected using the first-order correction model and the second-order correction model respectively, and P 4 and P 5 are obtained respectively. Then, the coefficient of difference between the corrected chromatogram P 4 and P 1 , and between P 5 and P 1 is calculated.

[0163] The test results are shown in Table 1 below:

[0164] Table 1

[0165] <![CDATA[P 1 associated with P 2 > <![CDATA[P 1 associated with P 3 > <![CDATA[P 1 associated with P 4 > <![CDATA[P 1 associated with P 5 > <![CDATA[Coefficient of difference ε Δs > 1.56％ 0.05％ 0.67％ 0.24％

[0166] It can be seen from the test results that the measured chromatogram P 2 after temperature change and the standard chromatogram P 1There are non-negligible differences. When using the first-order correction model and the second-order correction model for correction, the differences decrease in turn, and when using the environmental temperature difference chromatogram correction model for correction, the difference is the smallest.

[0167] 2. Exhaust gas quantitative analysis result test:

[0168] Using the method in Example 1, chromatographic detection was performed on the test sample at environmental temperatures of 5°C and 15°C respectively; as a control, in the control example, the method in the example was used, but the environmental temperature difference chromatogram correction model was not used for correction. Instead, the original chromatogram was directly subjected to peak searching using the method in step S3, and then analyzed using the method in S4 to calculate the test error coefficient η C :

[0169]

[0170] Among them, C T represents the actual concentration measured at different environmental temperatures (5°C, 15, 25°C), and C 标 represents the actual concentration of the sample in the test sample. The smaller the value of η C , the higher the accuracy.

[0171] Among them, at an environmental temperature of 25°C, the environmental temperature difference chromatogram correction model was also not used for correction. Instead, the original chromatogram was directly subjected to peak searching using the method in step S3, and then analyzed using the method in S4.

[0172] The test results are shown in Table 2 below:

[0173] Table 2

[0174]

[0175] It can be seen from the test results in Table 2 that the test accuracy of Example 1 is very high. When the environmental temperature difference chromatogram correction model is not used for correction, the test results are greatly affected by temperature, and the test accuracy rate is significantly reduced.

[0176] Example 2

[0177] An industrial waste gas detection system based on gas chromatography uses the method of Example 1 for industrial waste gas detection. Referring to Figure 12 , this system includes:

[0178] A filter for filtering the industrial waste gas to be detected;

[0179] A gas chromatograph for detecting the filtered industrial waste gas to obtain an original chromatogram;

[0180] A fast peak search model that uses the method in step S3 to perform peak search on the calibrated chromatogram to obtain all chromatographic peaks;

[0181] And a chromatographic peak analysis module that uses the method in step S4 to analyze the obtained chromatographic peaks to obtain the components in the industrial waste gas to be detected and the content of each component.

[0182] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0183] Although the embodiments of the present invention have been disclosed as above, it is not limited to the applications listed in the specification and embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the field, additional modifications can be easily made. Therefore, without departing from the general concept defined by the claims and the equivalent scope, the present invention is not limited to specific details.

Claims

1. A method for detecting industrial waste gas based on gas chromatography, characterized in that: The following steps are involved: S1. Pre-constructing an environmental temperature difference chromatogram correction model, which corrects the chromatogram obtained by the gas chromatograph according to the temperature difference between the actual test environment and the standard test environment; S2, detecting the industrial waste gas to be detected by a gas chromatograph, obtaining an original chromatogram, and correcting it by using the ambient temperature difference chromatogram correction model to obtain a corrected chromatogram; S3, performing peak searching on the corrected chromatogram using a fast peak searching model to obtain all chromatographic peaks; S4. Analyze the chromatographic peaks obtained in step S3 to obtain the components in the industrial waste gas to be detected and the content of each component.

2. The method for detecting industrial waste gas based on gas chromatography according to claim 1, characterized in that: The environmental temperature difference chromatogram correction model is constructed by the following method: S1-1, pre-acquire each waste gas component in the case where the waste gas component contained in the industrial waste gas to be detected in the target scene is the largest; S1-2. Build training data set: A single-component standard sample containing only one exhaust gas component is subjected to gas chromatography detection at a standard ambient temperature, and then subjected to gas chromatography detection again after only the ambient temperature is changed, and all the obtained chromatograms are used to construct a single-component training data set S1; A multi-component standard sample containing at least two exhaust gas components is subjected to gas chromatography detection at a standard ambient temperature, and then subjected to gas chromatography detection again after only the ambient temperature is changed, and a multi-component training data set S2 is constructed using all the obtained chromatograms; A full-component standard sample containing all exhaust gas components is subjected to gas chromatography detection at a standard ambient temperature, and then gas chromatography detection is performed again after only the ambient temperature is changed, and all the obtained chromatograms are used to construct a full-component training data set S3; S1-3. Model training: A single-component training data set S1, a multi-component training data set S2, and a full-component training data set S3 are used; a CNN convolutional neural network is trained in sequence, and after the training is completed, the environmental temperature difference chromatogram correction model is obtained.

3. The method for detecting industrial waste gas based on gas chromatography according to claim 2, characterized in that: Step S1-2 specifically includes: S1-2-1. Construct a single-component training dataset S1: At a standard ambient temperature, a single-component standard sample containing only one exhaust gas component is obtained by a gas chromatograph for measurement, and a chromatogram is obtained, which is recorded as a standard single-component chromatogram; each exhaust gas component is tested one by one, and a standard single-component chromatogram of each single-component standard sample is obtained respectively; For any single-component standard sample, only the ambient temperature is changed to obtain the chromatograms at different actual ambient temperatures, which are recorded as the measured chromatogram PC. BΔT , record ΔT = actual ambient temperature - standard ambient temperature; for any single-component standard sample C B , set ΔT, ΔT down C B The corresponding measured chromatogram PC BΔT and C B Standard single component chromatogram PC B Combine into a single-component training data s1, combine all the obtained single-component training data s1, and construct a single-component training data set S1; S1-2-2. Construct multi-component training data set S2: At a standard ambient temperature, a multi-component standard sample containing at least two exhaust gas components is obtained by using a gas chromatograph for measurement, and a chromatogram is obtained, which is recorded as a standard multi-component chromatogram; then the types and quantities of exhaust gas components in the standard sample are changed, and the tests are performed separately to obtain a standard multi-component chromatogram of each multi-component standard sample; For any multi-component standard sample, only the ambient temperature is changed to obtain the chromatograms at different actual ambient temperatures, which are recorded as the measured chromatogram PCd BΔT , record ΔT = actual ambient temperature - standard ambient temperature; for any multi-component standard sample Cd B , set ΔT, ΔT under Cd B The measured chromatogram PCd BΔT and Cd B Standard multicomponent chromatogram PCd B Combine into a multi-component training data s2, combine all the obtained multi-component training data s2, and construct a multi-component training data set S2; S1-2-3. Construct the full component training data set S3: At a standard ambient temperature, a full-component standard sample containing all exhaust gas components is obtained by a gas chromatograph for measurement, and a chromatogram is obtained, which is recorded as a standard full-component chromatogram; only the concentration ratio of the exhaust gas components in the full-component standard sample is changed, and a standard full-component chromatogram of the full-component standard sample of each different concentration ratio is obtained; For any full-component standard sample, only the ambient temperature is changed to obtain the chromatograms at different actual ambient temperatures, which are recorded as the measured chromatogram PCa. BΔT , record ΔT = actual ambient temperature - standard ambient temperature; for any full-component standard sample Ca B , set ΔT, ΔT under Ca B The measured chromatogram PCa BΔT and Ca B Standard full component chromatogram PCa B Combine into a full-component training data s3, combine all the obtained full-component training data s3, and construct a full-component training data set S3; Among them, the standard ambient temperature is 20-25℃.

4. The method for detecting industrial waste gas based on gas chromatography according to claim 3, characterized in that: Step S1-3 specifically includes: S1-3-1, using the single-component training data set S1, with ΔT, PC BΔT For input, PC B The CNN convolutional neural network is trained for output, and a first-order correction model is obtained after the training is completed; S1-3-2, using the multi-component training data set S2, with ΔT, PCd BΔT is the input, PCd B The first-order correction model is trained for output, and the second-order correction model is obtained after the training is completed; S1-3-3, using the full component training data set S3, with ΔT, PCa BΔT is the input, PCa B The second-order correction model is trained for output, and the final ambient temperature difference chromatogram correction model is obtained after the training is completed.

5. The method for detecting industrial waste gas based on gas chromatography according to claim 1, characterized in that: The method for performing peak-finding processing using the fast peak-finding model comprises the following steps: S3-1. Obtain all peak vertices in the corrected chromatogram. The set of all peak vertices is recorded as Q p ; S3-2. Get Q p The peak starting point of the peak corresponding to each peak vertex in ; S3-3. Get Q p The peak end point of the peak corresponding to each peak vertex in; S3-4. According to the obtained peak apex, the peak starting point and the peak end point corresponding to each peak apex, all chromatographic peaks are obtained and marked on the chromatographic curve of the corrected chromatogram, so as to obtain a processed chromatogram with marked chromatographic peaks.

6. The method for detecting industrial waste gas based on gas chromatography according to claim 2, characterized in that: The method for performing peak-finding processing using the fast peak-finding model comprises the following steps: S3-1. Get the peak apex: S3-1-1. On the chromatographic curve of the calibrated chromatogram, starting from the origin and from left to right, take J consecutive data points as one data unit. For the last data unit, take the remaining data points as one data unit. The number of data points in the last data unit is J'. The total number of data units is m. Then J'≤J, and the total number of data points N=m*(J-1)+J'; S3-1-2. Use the following steps to obtain the peak vertices for all data units at the same time: For a data unit, remove any other data point P in the data unit except the left endpoint and the right endpoint. h , when the following conditions a and b are met, the data point P h As the peak apex: a. Data point P h The signal strength of is greater than the signal strength of the first data point on its left and the first data point on its right; b. From data point P h There are at least n consecutive T The signal strength of each data point decreases in sequence, from data point P h There are at least n consecutive T The signal strength of each data point decreases successively; Traverse all data points in the data unit except the left and right endpoints, and construct all the peak vertices obtained as the peak vertex set Q P1 ; S3-1-3. On the chromatographic curve of the corrected chromatogram, for the data points at the left and right endpoints of all data units, when the above conditions a and b are met at the same time, the current data point is taken as the peak vertex, and all the obtained peak vertices are constructed as the peak vertex set Q. P2 ; S3-1-4, Q P1 and Q P2 Combine to get the set Q of all peak vertices of the corrected chromatogram p , Q p =Q P1 +Q P2 ; S3-2. Obtaining the peak vertex set Q p The peak starting point of each peak corresponding to the middle peak vertex: For Q p The peak point P n , from P n Calculate the slope of the line between two adjacent data points to the left, recorded as: P n and P n-1 The slope between them is k n , P n-1 and P n-2 The slope between them is k n-1 ,...,P n-(m-1) and P n-m The slope between them is k n-(m-1) , P n-m and P n-(m+1) The slope between them is k n-m ; When the first data point that satisfies the following conditions c or d appears, the data point P n-m As the peak point P n The corresponding peak starting point is: c.k n-(m-1) >0, and k n-m ≤0; d, k n-(m-1) >0,k n-m >0, and k n-(m-1) -k n-m ≤Δk T1 ; Δk T1 is a preset slope change threshold value, which is a constant not less than 0; S3-3. Obtaining the peak vertex set Q p The peak end point of each peak corresponding to the peak apex: For Q p The peak point P n , from P n Calculate the slope of the line between two adjacent data points to the right, recorded as: P n and P n+1 The slope between them is k n+1 , P n+1 and P n+2 The slope between them is k n+2 ,...,P n+(m'-1) and P n+m' The slope between them is k n+m' , P n+m' and P n+(m'+1) The slope between them is k n+(m'+1) ; When the first data point that satisfies the following conditions e or f appears, the data point P n+m' As the peak point P n The corresponding peak end point is: e、k n+m' <0, and k n+(m'+1) ≥0; f、k n+m' <0, k n+(m'+1) <0, and |k n+m' |-|k n+(m'+1) |≤Δk T2 , Δk T2 is a preset slope change threshold value, which is a constant not less than 0; S3-4. According to the obtained peak apex, the peak starting point and the peak end point corresponding to each peak apex, all chromatographic peaks are obtained and marked on the chromatographic curve of the corrected chromatogram, thereby obtaining a processed chromatogram containing chromatographic peaks.

7. The method for detecting industrial waste gas based on gas chromatography according to claim 6, characterized in that: in, J: 5-50,n T 2-5.

8. The method for detecting industrial waste gas based on gas chromatography according to claim 6, characterized in that: in, 0≤Δk T1 ≤0.15,0≤Δk T2 ≤0.15。 9. The method for detecting industrial waste gas based on gas chromatography according to claim 6, characterized in that: Step S4 is specifically as follows: S4-1. Compare the processed chromatogram with the standard full-component chromatogram to obtain the components corresponding to each chromatographic peak in the processed chromatogram: For any chromatographic peak R in the processed chromatogram i , record the chromatographic peak R i The peak point is P i , get P i The horizontal axis t i In the standard full component chromatogram, the horizontal axis range is (t i -0.5d min )~(t i +0.5d min ) range, obtain the horizontal coordinate t of the peak vertex Bi With t i The peak point P corresponding to the minimum absolute value of the difference Bi , P Bi The component C corresponding to the peak i As the chromatographic peak R i The corresponding ingredients; When the standard full component chromatogram is in the horizontal axis range (t i -0.5d min )~(t i +0.5d min ) range, and the chromatographic peak R was manually confirmed. i The corresponding components; among them, d min It is the minimum value of the distance between the peak vertices of two adjacent chromatographic peaks on the horizontal axis in the standard full-component chromatogram; S4-2. According to the area of ​​the chromatographic peak of the processed chromatogram, combined with the pre-constructed standard curve representing the relationship between the chromatographic peak area and the component concentration in the standard full-component chromatogram, the concentration of the component corresponding to each chromatographic peak is calculated.

10. An industrial waste gas detection system based on gas chromatography, characterized in that: The method according to any one of claims 1 to 9 is used to detect industrial waste gas, and the system comprises: A gas chromatograph, which is used to detect the industrial waste gas to be detected and obtain an original chromatogram; A calibration model for the ambient temperature difference chromatogram, which uses the method in step S2 to obtain a calibrated chromatogram; A fast peak-finding model, which uses the method in step S3 to perform peak-finding processing on the corrected chromatogram to obtain all chromatographic peaks; And a chromatographic peak analysis module, which adopts the method in step S4 to obtain the components in the industrial waste gas to be detected and the content of each component according to the obtained chromatographic peak analysis.

Images