Method and system for remote classification and identification of metallurgical gases based on mid-long wave infrared cooperation
By employing a mid-to-long-wave infrared collaborative detection strategy and a decision tree model, the problem of identifying metallurgical gas types over long distances was solved, enabling rapid and accurate gas type identification, particularly the distinction between coke oven gas and blast furnace gas.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- UNIV OF SCI & TECH BEIJING
- Filing Date
- 2026-01-21
- Publication Date
- 2026-06-05
AI Technical Summary
Existing technologies struggle to quickly and accurately detect and identify the types of metallurgical gas leaks under long-distance, non-contact, and passive conditions, especially since the low methane content in coke oven gas, converter gas, and blast furnace gas makes identification difficult.
A dual-band collaborative detection strategy is adopted, using a mid-wave analysis window to detect carbon monoxide leaks and a long-wave analysis window to detect methane leaks. By combining nonlinear least squares method and decision tree model, the concentration inversion of carbon monoxide and methane and the identification of gas types are realized.
It enables rapid detection and accurate identification of metallurgical gas types under long-distance, non-contact, and passive conditions, improves radiation contrast and detection sensitivity, and simplifies the identification process.
Smart Images
Figure CN122150166A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of metallurgical gas classification and identification technology, and in particular to a method and system for metallurgical gas remote sensing classification and identification based on mid- and long-wave infrared synergy. Background Technology
[0002] By-product gases produced in the metallurgical industry mainly include blast furnace gas, converter gas, and coke oven gas. Their main components are all carbon monoxide, carbon dioxide, and methane, but in different proportions. In particular, coke oven gas contains approximately 22%-26% methane, while blast furnace and converter gases contain extremely low levels of methane. Existing technological solutions include:
[0003] (1) Extraction-type FTIR monitoring: Gas is extracted and analyzed through a sampling tube, but it has disadvantages such as response lag, inability to monitor fugitive leaks, and high maintenance costs.
[0004] (2) Active open-path FTIR: It requires infrared light sources and detectors to be set up at both ends of the monitoring path, but it has disadvantages such as system complexity, difficulty in optical path alignment, susceptibility to field vibration, and unsuitability for flexible inspection.
[0005] (3) Single-band passive FTIR (medium wave only): It uses the 3-5μm band to simultaneously invert carbon monoxide and methane, but it has the disadvantage of being easily submerged by noise, which makes it impossible to accurately identify methane and thus unable to distinguish the types of gases.
[0006] Meanwhile, for unorganized emissions (leaks) of metallurgical gases, existing technologies are insufficient to simultaneously achieve rapid leak detection and accurate gas identification under long-distance, non-contact, and passive conditions. Summary of the Invention
[0007] To address the problems in the existing technology, this invention provides a method and system for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy. This invention first employs a dual-band collaborative detection strategy, using a mid-wave analysis window as the trigger signal for detecting carbon monoxide leaks and a long-wave analysis window as the trigger signal for detecting methane leaks, accurately measuring the leakage of both carbon monoxide and methane. Second, through inversion, the concentrations of carbon monoxide and methane are obtained, resulting in higher radiation contrast and detection sensitivity. Finally, a decision tree model is used to identify the types of metallurgical gases, making the identification results simpler and more accurate. This enables rapid leak detection and accurate gas identification under long-distance, non-contact, and passive conditions. To achieve the above objectives, the technical solution is as follows:
[0008] On the one hand, this invention provides a method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy, the method comprising:
[0009] S1. Scan the area to be measured using a wideband passive Fourier transform infrared spectrometer telemetry system to obtain the infrared radiation spectrum of the area to be measured.
[0010] S2. Based on the infrared radiation spectrum of the area to be measured, perform spectral preprocessing and band separation to obtain the mid-wave analysis window and the long-wave analysis window;
[0011] S3. Based on the mid-wave analysis window, the carbon monoxide leakage situation is obtained through inversion;
[0012] S4. Based on this long-wavelength analysis window, the leakage situation of methane is obtained through inversion;
[0013] S5. Based on the leakage of carbon monoxide and methane, the metallurgical gas types in the area to be tested are identified using a decision tree model.
[0014] Optionally, the area to be tested includes: pipes or outlets.
[0015] Optionally, in step S2, based on the infrared radiation spectrum of the region to be measured, spectral preprocessing and band separation are performed to obtain a mid-wave analysis window and a long-wave analysis window, including:
[0016] S21. Based on the infrared radiation spectrum of the area to be measured, perform a Fourier transform to obtain the brightness temperature spectrum and radiance spectrum of the area to be measured.
[0017] S22. Based on the brightness temperature spectrum and radiance spectrum of the area to be measured, the mid-wave analysis window and long-wave analysis window are obtained by separating the wavebands.
[0018] Optionally, in S3, based on the mid-wave analysis window, the carbon monoxide leakage situation is obtained through inversion, including:
[0019] S31. Based on the mid-wave analysis window, the absorbance spectrum of carbon monoxide is obtained by fitting using the nonlinear least squares method.
[0020] S32. Based on the absorbance spectrum of the carbon monoxide, the column concentration of carbon monoxide is obtained through inversion.
[0021] S33. Based on the column concentration of carbon monoxide, the leakage status of carbon monoxide is obtained through the carbon monoxide judgment rules.
[0022] Optionally, the carbon monoxide determination rule includes:
[0023] Rule 1: If the column concentration of carbon monoxide is greater than the set threshold for carbon monoxide, it is determined that there is a carbon monoxide gas leak.
[0024] Rule 2: If the column concentration of carbon monoxide is less than or equal to the set threshold for carbon monoxide, it is determined that there is no carbon monoxide gas leak.
[0025] Optionally, in S4, based on the long-wavelength analysis window, the methane leakage situation is obtained through inversion, including:
[0026] S41. Based on the long-wavelength analysis window, water vapor is introduced as an interference component to participate in the fitting, and the band signal after deducting the continuous absorption of water vapor in the long-wavelength band is obtained.
[0027] S42. Based on the band signal that has been deducted from the continuous absorption of long-wavelength water vapor, the column concentration of methane is obtained through inversion.
[0028] S43. Based on the column concentration of methane, the leakage situation of methane is obtained.
[0029] Optionally, in step S5, based on the leakage status of carbon monoxide and methane, a decision tree model is used to identify the types of metallurgical gases in the area to be tested, including:
[0030] S51. Based on the carbon monoxide leakage situation, the leakage situation of metallurgical gas is obtained;
[0031] S52. Based on the leakage of the metallurgical gas and the leakage of the methane, the type of metallurgical gas in the area to be tested is obtained.
[0032] Optionally, the metallurgical gas types in the area to be tested include: coke oven gas, blast furnace gas, or converter gas.
[0033] On the other hand, the present invention provides a metallurgical gas telemetry classification and identification system based on mid- and long-wave infrared synergy. This system is applied to a metallurgical gas telemetry classification and identification method based on mid- and long-wave infrared synergy. The system includes:
[0034] The spectral scanning module is used to scan the area to be measured using a broadband passive Fourier transform infrared spectrometer telemetry system to obtain the infrared radiation spectrum of the area to be measured.
[0035] The analysis window separation module is used to perform spectral preprocessing and band separation based on the infrared radiation spectrum of the region to be measured, to obtain the mid-wave analysis window and the long-wave analysis window;
[0036] The carbon monoxide leakage analysis module is used to obtain the carbon monoxide leakage situation through inversion based on the mid-wave analysis window.
[0037] The methane leakage analysis module is used to obtain the methane leakage situation through inversion based on the long-wavelength analysis window;
[0038] The gas classification module is used to identify the types of metallurgical gases in the area to be tested by using a decision tree model based on the leakage status of carbon monoxide and methane.
[0039] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:
[0040] The above-mentioned scheme employs a dual-band collaborative detection strategy, using a mid-wave analysis window as a trigger signal for detecting carbon monoxide leaks and a long-wave analysis window as a trigger signal for detecting methane leaks, to accurately measure the leakage of both carbon monoxide and methane. Secondly, it obtains the concentrations of carbon monoxide and methane through inversion, resulting in higher radiation contrast and detection sensitivity. Thirdly, it identifies the types of metallurgical gases through a decision tree model, making the identification results simpler and more accurate. This enables rapid leak detection and accurate gas identification under long-distance, non-contact, and passive conditions. Attached Figure Description
[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0042] Figure 1 This is a flowchart of an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention;
[0043] Figure 2 This is a flowchart illustrating the mid-wave analysis window and long-wave analysis window obtained in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared coordination of the present invention.
[0044] Figure 3 This is a flowchart illustrating the carbon monoxide leakage situation obtained in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0045] Figure 4 This is a flowchart illustrating the methane leakage situation obtained in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0046] Figure 5 This is a flowchart illustrating the metallurgical gas types in the area to be measured in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0047] Figure 6 This is a comparison of the brightness temperature spectra of the mid-wave analysis window and the long-wave analysis window in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0048] Figure 7This is a schematic diagram of the result of spectral fitting of carbon monoxide in coke oven gas by the mid-wave analysis window in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0049] Figure 8 This is a comparison of the spectral fitting effects of methane in the mid-wave and long-wave analysis windows in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0050] Figure 9 This is a system block diagram of an embodiment of the metallurgical gas remote sensing and classification system based on mid- and long-wave infrared synergy of the present invention. Detailed Implementation
[0051] The technical solution of the present invention will now be described with reference to the accompanying drawings.
[0052] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.
[0053] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.
[0054] like Figure 1 The flowchart shown is an embodiment of the metallurgical gas telemetry classification and identification method based on mid- and long-wave infrared synergy of the present invention. The present invention provides a metallurgical gas telemetry classification and identification method based on mid- and long-wave infrared synergy, which is implemented by a metallurgical gas telemetry classification and identification system based on mid- and long-wave infrared synergy. The method includes:
[0055] S1. Scan the area to be measured using a wideband passive Fourier transform infrared spectrometer telemetry system to obtain the infrared radiation spectrum of the area to be measured.
[0056] Specifically, the area to be tested includes: pipes or outlets.
[0057] S2. Based on the infrared radiation spectrum of the area to be measured, perform spectral preprocessing and band separation to obtain the mid-wave analysis window and the long-wave analysis window;
[0058] Specifically, such as Figure 2The flowchart shown in the embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention obtains the mid-wave analysis window and the long-wave analysis window. In step S2, based on the infrared radiation spectrum of the region to be measured, spectral preprocessing and band separation are performed to obtain the mid-wave analysis window and the long-wave analysis window, including:
[0059] S21. Based on the infrared radiation spectrum of the area to be measured, perform a Fourier transform to obtain the brightness temperature spectrum and radiance spectrum of the area to be measured.
[0060] S22. Based on the brightness temperature spectrum and radiance spectrum of the area to be measured, the mid-wave analysis window and long-wave analysis window are obtained by separating the wavebands.
[0061] Furthermore, such as Figure 6 The brightness temperature spectrum comparison diagram of the mid-wave analysis window and the long-wave analysis window in the embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention is shown. Figure 6 Analysis shows that, Figure 6 (a) shows the methane signal near 3.3 μm (weak, submerged in noise) and the carbon monoxide signal near 4.6 μm (stronger). Figure 6 (b) shows the spectrum in the 7-13 μm range. The methane signal near 7.6 μm is strong and distinctive even in the presence of water vapor.
[0062] Therefore, the mid-wave analysis window is 2050–2200 cm⁻¹. -1 (Approximately 4.6 μm), covering the characteristic absorption peaks of CO; long-wavelength analysis window: 1200–1400 cm⁻¹ 1 (Approximately 7.6 μm), covering methane Bending vibration belt.
[0063] S3. Based on the mid-wave analysis window, the carbon monoxide leakage situation is obtained through inversion;
[0064] Specifically, such as Figure 3 The flowchart shown in this embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention obtains the carbon monoxide leakage situation. In step S3, the carbon monoxide leakage situation is obtained through inversion based on the mid-wave analysis window, including:
[0065] S31. Based on the mid-wave analysis window, the absorbance spectrum of carbon monoxide is obtained by fitting using the nonlinear least squares method.
[0066] Furthermore, we obtain, as Figure 7 The diagram shows the result of spectral fitting of carbon monoxide in coke oven gas using a mid-wave analysis window in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention.
[0067] S32. Based on the absorbance spectrum of the carbon monoxide, the column concentration of carbon monoxide is obtained through inversion.
[0068] S33. Based on the column concentration of carbon monoxide, the leakage status of carbon monoxide is obtained through the carbon monoxide judgment rules.
[0069] Specifically, the carbon monoxide determination rule includes:
[0070] Rule 1: If the column concentration of carbon monoxide is greater than the set threshold for carbon monoxide, it is determined that there is a carbon monoxide gas leak.
[0071] Rule 2: If the column concentration of carbon monoxide is less than or equal to the set threshold for carbon monoxide, it is determined that there is no carbon monoxide gas leak.
[0072] Furthermore, all metallurgical gases (blast furnace, converter, coke oven) contain carbon monoxide, making carbon monoxide the best leak indicator. If a carbon monoxide gas leak is detected, proceed to the next classification step; otherwise, continue monitoring.
[0073] S4. Based on this long-wavelength analysis window, the leakage situation of methane is obtained through inversion;
[0074] Specifically, such as Figure 4 This is a flowchart illustrating the methane leakage situation obtained in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention. In step S4, the methane leakage situation is obtained through inversion based on the long-wave analysis window, including:
[0075] S41. Based on the long-wavelength analysis window, water vapor is introduced as an interference component to participate in the fitting, and the band signal after deducting the continuous absorption of water vapor in the long-wavelength band is obtained.
[0076] S42. Based on the band signal that has been deducted from the continuous absorption of long-wavelength water vapor, the column concentration of methane is obtained through inversion.
[0077] S43. Based on the column concentration of methane, the leakage situation of methane is obtained.
[0078] S5. Based on the leakage of carbon monoxide and methane, the metallurgical gas types in the area to be tested are identified using a decision tree model.
[0079] Specifically, such as Figure 5 The flowchart shown in this embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention obtains the metallurgical gas types in the area to be measured. In step S5, based on the leakage status of carbon monoxide and methane, the metallurgical gas types in the area to be measured are identified through a decision tree model, including:
[0080] S51. Based on the carbon monoxide leakage situation, the leakage situation of metallurgical gas is obtained;
[0081] S52. Based on the leakage of the metallurgical gas and the leakage of the methane, the type of metallurgical gas in the area to be tested is obtained.
[0082] Specifically, the types of metallurgical gases in the area to be tested include: coke oven gas, blast furnace gas, or converter gas.
[0083] Furthermore, such as Figure 8 The figure shown is a comparison of the spectral fitting effects of methane in the mid-wave and long-wave analysis windows in an embodiment of the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy of the present invention. Figure 8 (a) shows the spectral fitting effect of methane in the mid-wave analysis window. Figure 8 (b) is a spectral fitting result of methane in the long-wavelength analysis window, used to determine the methane content: a methane threshold is set. ,
[0084] Branch A: If the column concentration of methane > (Significant methane is present): Conclusion: The leak source is determined to be coke oven gas.
[0085] Branch B: If the column concentration of methane is ≈0 or below the detection limit (showing no / very low methane): Conclusion: The leak source is determined to be blast furnace gas or converter gas.
[0086] like Figure 9 The diagram shown is a system block diagram of an embodiment of the metallurgical gas remote sensing classification system based on mid- and long-wave infrared synergy of the present invention. The present invention provides a metallurgical gas remote sensing classification and identification system based on mid- and long-wave infrared synergy. This system is applied to a metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy. The system includes: a spectral scanning module, an analysis window separation module, a carbon monoxide leakage analysis module, a methane leakage analysis module, and a gas classification module. Specifically,
[0087] The spectral scanning module is used to scan the area to be measured using a broadband passive Fourier transform infrared spectrometer telemetry system to obtain the infrared radiation spectrum of the area to be measured.
[0088] The analysis window separation module is used to perform spectral preprocessing and band separation based on the infrared radiation spectrum of the region to be measured, to obtain the mid-wave analysis window and the long-wave analysis window;
[0089] The carbon monoxide leakage analysis module is used to obtain the carbon monoxide leakage situation through inversion based on the mid-wave analysis window.
[0090] The methane leakage analysis module is used to obtain the methane leakage situation through inversion based on the long-wavelength analysis window;
[0091] The gas classification module is used to identify the types of metallurgical gases in the area to be tested by using a decision tree model based on the leakage status of carbon monoxide and methane.
[0092] This invention provides a method and system for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy. Firstly, it employs a dual-band collaborative detection strategy, using a mid-wave analysis window as a trigger signal for detecting carbon monoxide leaks and a long-wave analysis window as a trigger signal for detecting methane leaks, accurately measuring the leakage of both carbon monoxide and methane. Secondly, it obtains the concentrations of carbon monoxide and methane through inversion, achieving higher radiation contrast and detection sensitivity. Finally, it identifies the types of metallurgical gases using a decision tree model, making the identification results simpler and more accurate. This enables rapid leak detection and accurate gas identification under long-distance, non-contact, and passive conditions.
[0093] It is understood that the present invention has been described through the above embodiments and should not be construed as limiting the implementation and scope of the present invention. Those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the present invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.
Claims
1. A method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy, characterized in that, The method includes: S1. Scan the area to be measured using a wideband passive Fourier transform infrared spectrometer telemetry system to obtain the infrared radiation spectrum of the area to be measured. S2. Based on the infrared radiation spectrum of the area to be measured, perform spectral preprocessing and band separation to obtain the mid-wave analysis window and the long-wave analysis window; S3. Based on the mid-wave analysis window, the carbon monoxide leakage situation is obtained through inversion; S4. Based on the long-wavelength analysis window, the leakage situation of methane is obtained through inversion; S5. Based on the leakage of carbon monoxide and methane, the metallurgical gas types in the area to be tested are identified using a decision tree model.
2. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 1, characterized in that, The area to be tested includes: pipes or outlets.
3. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 1, characterized in that, In step S2, based on the infrared radiation spectrum of the region to be measured, spectral preprocessing and band separation are performed to obtain a mid-wave analysis window and a long-wave analysis window, including: S21. Based on the infrared radiation spectrum of the area to be measured, perform a Fourier transform to obtain the brightness temperature spectrum and radiance spectrum of the area to be measured. S22. Based on the brightness temperature spectrum and radiance spectrum of the area to be measured, the mid-wave analysis window and long-wave analysis window are obtained by separating the wavebands.
4. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 1, characterized in that, In step S3, based on the mid-wave analysis window, the carbon monoxide leakage situation is obtained through inversion, including: S31. Based on the mid-wave analysis window, the absorbance spectrum of carbon monoxide is obtained by fitting using the nonlinear least squares method. S32. Based on the absorbance spectrum of carbon monoxide, the column concentration of carbon monoxide is obtained by inversion; S33. Based on the column concentration of carbon monoxide, the leakage status of carbon monoxide is determined by the carbon monoxide judgment rules.
5. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 4, characterized in that, The carbon monoxide determination rules include: Rule 1: If the column concentration of carbon monoxide is greater than the set threshold for carbon monoxide, it is determined that there is a carbon monoxide gas leak. Rule 2: If the column concentration of carbon monoxide is less than or equal to the set threshold for carbon monoxide, it is determined that there is no carbon monoxide gas leak.
6. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 1, characterized in that, In step S4, based on the long-wavelength analysis window, the methane leakage situation is obtained through inversion, including: S41. Based on the long-wavelength analysis window, water vapor is introduced as an interference component to participate in the fitting, and a band signal after deducting the continuous absorption of water vapor in the long-wavelength band is obtained. S42. Based on the band signal after deducting the continuous absorption of long-wavelength water vapor, the column concentration of methane is obtained through inversion. S43. Based on the column concentration of methane, the leakage situation of methane is obtained.
7. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 1, characterized in that, In step S5, based on the leakage status of carbon monoxide and methane, a decision tree model is used to identify the types of metallurgical gases in the area to be tested, including: S51. Based on the carbon monoxide leakage situation, the leakage situation of metallurgical gas is obtained; S52. Based on the leakage of the metallurgical gas and the leakage of methane, the type of metallurgical gas in the area to be tested is obtained.
8. The method for remote sensing classification and identification of metallurgical gases based on mid- and long-wave infrared synergy according to claim 1, characterized in that, The types of metallurgical gases in the area to be tested include: coke oven gas, blast furnace gas, or converter gas.
9. A metallurgical gas remote sensing classification and identification system based on mid- and long-wave infrared synergy, used to implement the metallurgical gas remote sensing classification and identification method based on mid- and long-wave infrared synergy as described in any one of claims 1-8, characterized in that, The system includes: The spectral scanning module is used to scan the area to be measured using a broadband passive Fourier transform infrared spectrometer telemetry system to obtain the infrared radiation spectrum of the area to be measured. The analysis window separation module is used to perform spectral preprocessing and band separation based on the infrared radiation spectrum of the region to be measured, to obtain the mid-wave analysis window and the long-wave analysis window; The carbon monoxide leakage analysis module is used to obtain the carbon monoxide leakage situation through inversion based on the mid-wave analysis window. The methane leakage analysis module is used to obtain the methane leakage situation through inversion based on the long-wavelength analysis window. The gas classification module is used to identify the types of metallurgical gases in the area to be tested by using a decision tree model based on the leakage status of carbon monoxide and methane.