High-sensitivity natural gas leakage laser detection method

Abstract

The invention discloses a high-sensitivity natural gas leakage laser detection method, and particularly relates to the technical field of natural gas detection. A plurality of laser emission wavelengths are set and respectively correspond to characteristic absorption spectrum sections of natural gas and interference gas, a plurality of tunable lasers are utilized to synchronously emit laser beams, high-resolution multi-wavelength absorption spectrum data are acquired through a multi-channel optical receiving system, and spectrum preprocessing and spectrum abnormal characteristic extraction technologies are combined, so that the characteristic absorption spectrum sections of the natural gas and the interference gas are obtained. Data are input into the fused gas absorption characteristic prediction model for comprehensive analysis, accurate identification and concentration quantification of a target natural gas absorption spectral line are realized, an identification result and concentration information are finally output, real-time early warning is supported, and the method significantly improves the resolution and accuracy of a detection system in a low-concentration and complex-background gas environment, and has a wide application prospect. The problems that a traditional method is limited by single wavelength, strong interference and low sensitivity are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of natural gas detection, and in particular to a high-sensitivity natural gas leakage laser detection method. Background Art

[0002] Laser detection of natural gas leaks is a high-precision technique that utilizes laser technology and spectral analysis to determine natural gas composition. This method uses a laser source to illuminate a natural gas sample and a multi-channel spectral analysis system to simultaneously monitor absorption signals at different wavelengths. Because natural gas molecules have characteristic absorption spectra at specific wavelengths, analyzing this spectral information allows for highly sensitive detection of changes in natural gas concentration.

[0003] The existing technology has the following shortcomings:

[0004] In natural gas leak accidents, in addition to natural gas (such as methane), other gases such as volatile organic compounds (VOCs) or nitrogen oxides (NOx) may also be present. The absorption spectra of these gases may overlap with the absorption spectra of natural gas, resulting in cross-interference in the detection signals. In such a complex gas mixture, traditional single-wavelength or low-wavelength detection methods may not be able to effectively distinguish the concentration of natural gas. Although multi-channel technology can capture signals of multiple wavelengths, how to accurately distinguish and quantify the concentration of each gas and avoid cross-interference remains a technical challenge. The existence of this problem directly affects the real-time response and accuracy of the gas leak detection system, especially in high-density industrial areas or low-concentration leaks, which may lead to detection failure and even delay the optimal time for leak repair. Summary of the Invention

[0005] The purpose of the present invention is to provide a high-sensitivity natural gas leak laser detection method to solve the shortcomings of the background technology.

[0006] In order to achieve the above objectives, the present invention provides the following technical solutions: a high-sensitivity natural gas leak laser detection method, comprising:

[0007] Set multiple laser emission wavelengths, and respectively correspond to the characteristic absorption spectrum of natural gas and its coexisting interfering gas;

[0008] Utilize multiple tunable lasers to simultaneously emit laser beams at set wavelengths, passing through the gas area to be measured;

[0009] The transmission signal of each wavelength after passing through the gas is received by a multi-channel optical receiving device to obtain corresponding multi-channel absorption spectrum data;

[0010] Preprocess the multi-channel absorption spectrum data, extract the spectral anomaly features in the multi-channel spectrum data and input them into the fusion gas absorption feature prediction model for analysis;

[0011] Distinguish the absorption spectra of target natural gas and interfering gas based on the analysis results and quantitatively calculate the natural gas concentration;

[0012] Output the natural gas identification results and its concentration information, and use the analysis data for real-time early warning.

[0013] Preferably, according to the characteristic absorption bands of natural gas and its coexisting interfering gases, the laser wavelengths are selected respectively: methane: 1657 nm; carbon dioxide: 1572 nm; nitrogen oxides: 2355 nm; volatile organic compounds: 1392 nm and 1450 nm.

[0014] Preferably, the multiple tunable lasers output set wavelengths by adjusting the driving current or operating temperature of the lasers, and the laser beams of multiple wavelengths are combined through a coaxial design or an optical beam splitting device and then pass through the gas area to be measured.

[0015] Preferably, the multi-channel optical receiving device includes a plurality of photodetectors, bandpass filters and collimating optical components, which are used to respectively receive and distinguish the transmitted light signal of each laser wavelength.

[0016] Preferably, the spectral abnormality features include spectral absorption slope jitter value and absorption peak depth abnormality value; wherein, the spectral absorption slope jitter value is obtained by:

[0017] Obtain multi-wavelength absorption spectrum data. For each wavelength λ, calculate the absorption intensity α(λ) at that wavelength. Calculate the spectral slope by taking the difference of the absorption coefficient at adjacent wavelengths. The expression is: α(λ i ) is at wavelength λ i The absorption coefficient at i ) is the slope of the spectrum, i.e., at wavelength λ i The absorption intensity change rate at the absorption peak is extracted; the fluctuation amplitude around the absorption peak is extracted, and the fluctuation of the slope at multiple wavelength points is calculated to obtain the slope jitter value. The expression is: Where D is the slope jitter value, N is the total number of wavelength data points, |S(λ i )-S(λ i-1 )| means calculating the difference between adjacent slopes and taking the absolute value to represent the magnitude of the slope change.

[0018] Preferably, the method for obtaining the absorption peak depth abnormal value is:

[0019] Acquire spectral data, including the value of the absorption coefficient α(λ) varying with wavelength λ, and set the spectral data as: α(λ)={α(λ1),α(λ2),…,α(λ n)}; where λ1,λ2,…,λ n is each wavelength point, α(λ) is the corresponding absorption intensity, and n is the total number of wavelengths; calculate the first-order derivative S1(λ i ), which is the rate at which the absorption coefficient changes with wavelength: Among them, S1(λ i ) is at wavelength λ i The first derivative at λ represents the rate of change of the absorption coefficient with wavelength, and the second derivative S2(λ i ), that is, the rate of change of the first-order derivative with wavelength. The second-order derivative can accurately identify the inflection point of the absorption peak, and the expression is: Among them, S2(λ i ) is the second-order derivative, which indicates the rate at which the first-order derivative changes with wavelength. The point where the second-order derivative reverses is the inflection point of the absorption peak. By detecting the second-order derivative S2(λ i ) inversion point, locate the position of the absorption peak; if S2(λ i )>0 and S2(λ i-1 )<0, then λ i is the top of the absorption peak; if S2(λ i )>0 and S2(λ i+1 )>0, then λ i It is the bottom of the absorption peak; the depth of the absorption peak refers to the difference between the top of the absorption peak and the baseline.

[0020] Preferably, the top position of the absorption peak is set as λpeak, and the baseline absorption value is set as αbaseline. Then, the calculation formula of the absorption peak depth Dpeak is: Dpeak = α(λpeak) - αbaseline; wherein α(λpeak) is the absorption coefficient at the absorption peak, and αbaseline is the baseline absorption value;

[0021] After calculating the depth of each absorption peak, compare it with the preset threshold value, which is set as Dthreshold. If Dpeak is greater than Dthreshold, it is considered that the absorption peak depth is abnormal; if Dpeak is less than or equal to Dthreshold, the absorption peak depth is considered to be within the normal range;

[0022] The depth values ​​of the absorption peaks greater than the threshold Dthreshold within a fixed time period T are collected and a data set is established. The mean of the data set is calculated as the absorption peak depth outlier.

[0023] Preferably, the spectral absorption slope jitter value and the absorption peak depth anomaly value are normalized so that they are both between [0, 1], and the comprehensive spectral analysis anomaly value is calculated based on the normalized spectral absorption slope jitter value and the absorption peak depth anomaly value.

[0024] Preferably, a threshold value is set for each gas: the threshold value of the target natural gas: under a known natural gas concentration, the characteristic value range of the spectrum is calculated to set the threshold value of the target gas; the threshold value of the interfering gas: according to the absorption spectrum line characteristics of the interfering gas, the corresponding threshold value is calculated; the calculated comprehensive spectrum analysis anomaly value Stotal is compared with the preset threshold value: if Stotal is greater than the threshold value of the target gas and less than the threshold value of the interfering gas, it is determined to be the target natural gas; if Stotal is greater than or equal to the threshold value of the interfering gas and less than or equal to the threshold value of the target gas, it is determined to be the interfering gas.

[0025] Preferably, the absorption coefficient is calculated based on the Lambert-Beer law, the gas concentration is obtained by weighted averaging, the type of target gas and the current concentration value are displayed on the interface in real time, and dynamic monitoring is achieved in combination with the concentration change curve; when the detected concentration exceeds the set safety threshold, the alarm system is triggered and the external control device is linked to respond and control.

[0026] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0027] 1. This invention proposes a highly sensitive natural gas leak laser detection method that overcomes the cross-interference and inaccurate detection issues inherent in existing technologies in complex gas mixtures. By setting multiple laser emission wavelengths and precisely matching the characteristic absorption spectra of natural gas and interfering gases, combined with the simultaneous emission of multiple tunable lasers and multi-channel reception and processing of spectral signals, the method significantly improves detection sensitivity and identification capabilities for low-concentration natural gas. In particular, by incorporating a combined analysis model of slope jitter and absorption peak depth anomalies, it accurately extracts spectral anomaly features, effectively distinguishing natural gas from interfering components.

[0028] 2. This invention combines spectral data preprocessing, outlier normalization, comprehensive feature evaluation, and quantitative concentration calculation to accurately identify target gases and conduct real-time concentration assessments. It also features automatic alarm linkage, making it suitable for a variety of applications, including natural gas pipeline networks, petrochemical plants, and urban security. The overall solution offers rapid system response, high accuracy, and strong anti-interference capabilities, making it particularly suitable for continuous monitoring and safety warnings in high-density industrial areas or complex environments with extremely low leakage concentrations. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0030] Figure 1This is a mind map of the method of the present invention. DETAILED DESCRIPTION

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0032] For examples, see Figure 1 As shown, the high-sensitivity natural gas leak laser detection method described in this embodiment includes:

[0033] Set multiple laser emission wavelengths, and respectively correspond to the characteristic absorption spectrum of natural gas and its coexisting interfering gas;

[0034] Utilize multiple tunable lasers to simultaneously emit laser beams at set wavelengths, passing through the gas area to be measured;

[0035] The transmission signal of each wavelength after passing through the gas is received by a multi-channel optical receiving device to obtain corresponding multi-channel absorption spectrum data;

[0036] Preprocess the multi-channel absorption spectrum data, extract the spectral anomaly features in the multi-channel spectrum data and input them into the fusion gas absorption feature prediction model for analysis;

[0037] Distinguish the absorption spectra of target natural gas and interfering gas based on the analysis results and quantitatively calculate the natural gas concentration;

[0038] Output the natural gas identification results and its concentration information, and use the analysis data for real-time early warning.

[0039] This embodiment uses multiple laser wavelengths to simultaneously detect natural gas (such as methane) and its common interfering gases (such as carbon dioxide, nitrogen oxides, and volatile organic compounds). Each gas has a unique absorption spectrum at a specific wavelength, so by selecting laser wavelengths that match the absorption characteristics of these gases, multiple gas components can be effectively distinguished.

[0040] Specifically, this embodiment selects the following laser wavelengths for multi-channel detection:

[0041] Methane: Select a laser with a wavelength of 1657nm, which corresponds to the strong absorption peak of methane and has high sensitivity, making it suitable for the detection of low-concentration methane.

[0042] Carbon dioxide: Select a laser with a wavelength of 1572nm, which is suitable for detecting the absorption spectrum of carbon dioxide and can effectively distinguish between methane and carbon dioxide.

[0043] Nitrogen oxides: Select a laser with a wavelength of 2355nm, which is suitable for the detection of nitrogen oxides and can be distinguished from the absorption spectrum of methane.

[0044] Volatile organic compounds: Select lasers with wavelengths of 1392nm and 1450nm, which correspond to the characteristic absorption spectra of volatile organic compounds, and can effectively identify and quantify the concentration of volatile organic compounds.

[0045] To achieve simultaneous emission of multiple wavelengths, this embodiment uses tunable lasers (e.g., tunable laser diodes (LDs)). These laser sources precisely output corresponding wavelength signals through wavelength selectors and optical beam splitters. The specific configuration is as follows:

[0046] Tunable lasers (LDs): These lasers can adjust their output wavelength by varying the current or temperature. They typically have good wavelength tunability, covering multiple wavelength bands from the near-infrared to the mid-infrared, allowing them to precisely match the absorption characteristics of various gas molecules.

[0047] Wavelength Setting: Different gases have different absorption spectrum characteristics. For example, methane has a strong absorption peak near 1657nm, carbon dioxide has a distinct absorption feature near 1572nm, and nitrogen oxides (NOx) have a characteristic absorption at 2355nm. By selecting the appropriate wavelength, each laser can be used to detect specific gases.

[0048] When multiple tunable lasers are working, they can simultaneously emit laser beams at set wavelengths. These laser beams pass through the gas area to be measured together, ensuring that absorption signals at different wavelengths can be collected simultaneously.

[0049] Each laser emits a laser beam of a specific wavelength through a precise optical system (e.g., focusing lens, optical fiber). Multiple lasers can be combined into a single optical path using appropriate optical beam splitting devices (e.g., optical beam splitters) or coaxial designs, ensuring that they pass through the gas region to be measured almost simultaneously.

[0050] The laser beam is precisely focused and irradiated onto the gas region through specific optical components (such as mirrors, lenses, and fiber couplers). To ensure beam stability and accuracy, the optical path design needs to consider factors such as the size of the gas region, the focusing conditions of the laser beam, and the concentration of the gas sample.

[0051] When a laser beam passes through the gas region being measured, some of the light is absorbed by the gas molecules. The intensity of this absorption is related to the gas concentration and its characteristic absorption spectrum. Each laser wavelength interacts with the gas molecules as it passes through the gas, resulting in varying degrees of absorption.

[0052] A multi-channel optical receiver primarily consists of multiple photodetectors and optical components. Its function is to receive transmitted light signals from lasers of different wavelengths in parallel and convert these signals into electrical signals for subsequent data processing and analysis. As each wavelength of laser light passes through the gas, the intensity of the light is attenuated to varying degrees due to the specific absorption characteristics of different gas molecules for light of different wavelengths. Therefore, by separately receiving the transmitted light at each wavelength, information related to the gas concentration and its absorption characteristics can be obtained.

[0053] In order to achieve multi-channel simultaneous reception, the optical system needs to be precisely designed to ensure that each wavelength of laser beam can be received independently without interference.

[0054] Light beams of different wavelengths emitted by multiple laser sources are typically combined into a single optical path using an optical beam splitter or fiber coupler. The function of the optical beam splitter is to separate the light beams of different wavelengths so that they can pass through the gas sample area independently and avoid mutual interference between the wavelengths.

[0055] Beam splitter: After passing through the beam splitter, each wavelength of laser beam is directed into a different optical path, ensuring that lasers of different wavelengths can pass through the gas area and be detected independently.

[0056] Fiber couplers: In some systems, fiber couplers are used to precisely couple light of different wavelengths. Fibers can direct laser beams into optical receivers, ensuring that signals at each wavelength are effectively collected.

[0057] In order to effectively select and isolate signals of various wavelengths, the system is usually equipped with optical filters. These filters can accurately filter out unwanted wavelength components, ensuring that the detector receives signals of the target wavelength.

[0058] Bandpass filter: For each wavelength, the system uses a bandpass filter to filter out light signals of other irrelevant wavelengths, ensuring that the received transmitted light is only the specific wavelength of interest.

[0059] Lens and collimation system: In order to ensure that the laser beam can be effectively focused on the receiving detector, the optical system uses components such as lenses and collimators to focus and adjust the laser so that the transmitted light can accurately illuminate the detector.

[0060] Photodetectors are the core components of multi-channel optical receivers, converting optical signals into electrical signals and transmitting them to subsequent signal processing systems. After each wavelength of laser beam passes through a gas, a portion of the light is absorbed by the gas, and the remaining transmitted light is received by the corresponding photodetector.

[0061] A photodiode is a common photodetector that converts received light signals into current signals. Each wavelength of transmitted light is detected by the corresponding photodiode:

[0062] For each specific wavelength of light, the system will be configured with a corresponding photodiode to ensure that the signal in each band can be detected independently and accurately.

[0063] Photodiodes generally have a fast response speed and are suitable for real-time gas concentration monitoring.

[0064] Avalanche photodiodes (APDs) are an ideal detector for low-light-intensity, high-sensitivity detection. They offer high gain and improved signal sensitivity in low-light conditions. For low-concentration natural gas detection, APDs provide a higher signal-to-noise ratio and more accurate results.

[0065] Multiple photodetectors synchronously receive transmitted light signals of different wavelengths. To ensure data consistency and synchronization, the data acquisition equipment in the system typically uses high-speed sampling and synchronization control mechanisms to ensure that signals of each wavelength are accurately recorded at the same time.

[0066] The detector output signals must be precisely synchronized to avoid errors caused by signal delays or inconsistent responses. All signals are then transmitted to the data acquisition system for parallel processing.

[0067] The photodetector converts the received transmitted light signals into electrical signals, which are then fed into a signal processing system for analysis. The signal intensity (or attenuation) at each wavelength reflects the absorption of the laser beam at that wavelength in the gas. By analyzing the signals at each wavelength, the system can obtain absorption spectrum data related to the gas concentration.

[0068] The transmission signal of each wavelength reflects the degree of absorption of the gas at that wavelength. Through the Beer-Lambert Law, a quantitative relationship can be established between the intensity of the transmitted light and the gas concentration: I(λ) = I0(λ)·e -α(λ)L Where I(λ) is the transmitted light intensity, I0(λ) is the incident light intensity, α(λ) is the absorption coefficient, and L is the optical path length of the gas sample. By using multi-channel absorption spectrum data, the system can identify the absorption peak corresponding to each wavelength and, based on this data, further calculate gas concentrations and identify gas species.

[0069] First, absorption spectrum data at multiple wavelengths is acquired from a multi-channel optical receiver. The absorption signal at each wavelength represents the absorption intensity of the laser beam by different gases, and these signals contain information about the gas concentration and absorption characteristics.

[0070] Acquire the transmitted light signal and the incident light signal at each wavelength.

[0071] According to the intensity of spectral absorption, the absorption coefficient is calculated using the Beer-Lambert Law: Where L is the optical path length of the gas sample, I(λ) is the transmitted light intensity, I0(λ) is the incident light intensity, and α(λ) is the absorption coefficient at that wavelength. Absorption coefficient data at multiple wavelengths is obtained as the spectral response of each gas component.

[0072] Multi-channel absorption spectrum data preprocessing, the goal of the preprocessing process is to remove noise, correct background signals, and standardize data to make subsequent analysis more accurate.

[0073] Filters are used to suppress signal noise, eliminating high-frequency noise caused by equipment or environmental interference. Common filtering methods include low-pass filtering and median filtering to smooth signals and remove short-term disturbances. Low-pass filtering removes high-frequency noise from the signal, preserving relatively stable signal fluctuations. Median filtering removes outliers caused by sudden changes in the signal, and is particularly effective in detecting impulse noise.

[0074] Remove spectral interferences introduced by background light or equipment. For example, use baseline correction methods to adjust the baseline of the raw signal to zero or a known constant.

[0075] Normalize the data of each channel so that signals of different wavelengths have the same dimension. Normalization methods usually include: The purpose of normalization is to eliminate the magnitude differences between different wavelengths and ensure the uniformity of the data.

[0076] By analyzing multi-channel spectral data, we extract features that reflect spectral anomalies. These anomalies are typically caused by differences in the absorption spectrum between the target natural gas (such as methane) and the interfering gas. Spectral anomaly features include spectral absorption slope jitter and absorption peak depth anomalies.

[0077] The method for obtaining the spectral absorption slope jitter value is as follows:

[0078] Obtain multi-wavelength absorption spectrum data: For each wavelength λ, calculate the absorption intensity α(λ) at the wavelength and calculate the spectral slope. The spectral slope reflects the rate of change of the absorption intensity with wavelength and is calculated by the difference of the absorption coefficient at adjacent wavelengths. The expression is: α(λ i ) is at wavelength λ i The absorption coefficient at i ) is the slope of the spectrum, i.e., at wavelength λ i The rate of change of absorption intensity at

[0079] Based on the calculation of the spectral slope, the fluctuation amplitude around the absorption peak is extracted, and the fluctuation of the slope at multiple wavelengths is calculated to obtain the jitter value. The expression is: Where D is the slope jitter value, N is the total number of wavelength data points (i.e. the length of the data), |S(λ i )-S(λ i-1 )| indicates that the difference between adjacent slopes is calculated and its absolute value is taken to represent the amplitude of the slope change. The jitter value usually reflects sudden or irregular changes in the absorption signal.

[0080] Among them, the method for obtaining the absorption peak depth abnormal value is:

[0081] Obtain spectral data, including the absorption coefficient α(λ) as a function of wavelength λ. Spectral data is usually recorded at each point in the data set with a one-to-one relationship between absorption coefficient and wavelength. Set the spectral data as: α(λ) = {α(λ1), α(λ2), …, α(λ n )}; where λ1,λ2,…,λ n is each wavelength point, α(λ) is the corresponding absorption intensity, and n is the total number of wavelengths.

[0082] To detect absorption peaks and calculate their depths, the second derivative method is used. Absorption peaks usually occur where the second derivative reverses, i.e., changes from positive to negative or vice versa.

[0083] Calculate the first-order derivative S1(λ i ), which is the rate at which the absorption coefficient changes with wavelength: Among them, S1(λ i ) is at wavelength λ i The first derivative at λ represents the rate of change of the absorption coefficient with wavelength. Calculate the second derivative S2(λ i ), which is the rate of change of the first-order derivative with wavelength. The second-order derivative can accurately identify the inflection point of the absorption peak, and the expression is: Among them, S2(λ i ) is the second derivative, representing the rate at which the first derivative changes with wavelength. The point at which the second derivative reverses is the inflection point of the absorption peak. A change from a positive to a negative value indicates the top of the absorption peak, typically the depth of the absorption peak. A change from a negative to a positive value indicates the bottom of the absorption peak.

[0084] By detecting the second-order derivative S2(λ i ) can accurately locate the position of the absorption peak. The absorption peak reversal point corresponds to the zero crossing point of the second-order derivative and is usually determined as follows:

[0085] If S2(λ i )>0 and S2(λ i-1 )<0, then λ i It may be the top of the absorption peak.

[0086] If S2(λ i )>0 and S2(λ i+1 )>0, then λ i It may be the bottom of the absorption peak.

[0087] The depth of an absorption peak is the difference between the peak top and the baseline, which is typically taken as the spectral background value near the absorption peak. In multi-channel spectral data, the baseline can be estimated using smoothing fitting methods or by selecting a standard wavelength range.

[0088] Assuming the top position of the absorption peak to be λpeak and the baseline absorption value to be αbaseline, the calculation formula for the depth of the absorption peak Dpeak is: Dpeak = α(λpeak) - αbaseline; where α(λpeak) is the absorption coefficient at the absorption peak and αbaseline is the baseline absorption value.

[0089] After calculating the depth of each absorption peak, it can be compared with historical data or a preset threshold. If the depth of the absorption peak exceeds the normal range, it is considered to be abnormal. If the threshold is set to Dthreshold, the judgment conditions are as follows:

[0090] If Dpeak is greater than Dthreshold, the absorption peak depth is considered abnormal. If Dpeak is less than or equal to Dthreshold, the absorption peak depth is considered within the normal range.

[0091] The depth values ​​of the absorption peaks greater than the threshold Dthreshold within a fixed time period T are collected and a data set is established. The mean of the data set is calculated as the absorption peak depth outlier.

[0092] The spectral absorption slope jitter value and the absorption peak depth anomaly value are normalized so that they are both between [0, 1]. The comprehensive spectral analysis anomaly value is calculated based on the normalized spectral absorption slope jitter value and absorption peak depth anomaly value.

[0093] For example, the present invention can use the following gas absorption characteristic prediction model formula to calculate the outlier value of the comprehensive spectrum analysis, and the calculation expression is: Where Stotal is the outlier of the comprehensive spectral analysis, D is the spectral absorption slope jitter value, mf is the absorption peak depth outlier value, a1 and a2 are the weight coefficients of the spectral absorption slope jitter value and the absorption peak depth outlier value (which can be optimized based on experimental experience or machine learning), and a1 and a2 are both greater than 0.

[0094] Based on previous experimental data or calibration data, a threshold value is set for each gas. This threshold value is calculated from a large amount of sample data and can distinguish the absorption lines of target gases (such as methane) and interfering gases (such as carbon dioxide, nitrogen oxides, etc.).

[0095] Target natural gas threshold: Usually, the characteristic value range of the spectrum is calculated under known natural gas concentration to set the target gas threshold. Interference gas threshold: The corresponding threshold is calculated based on the absorption spectrum characteristics of the interfering gas.

[0096] The calculated comprehensive spectral analysis anomaly value Stotal is compared with the preset threshold: if Stotal is greater than the threshold of the target gas and less than the threshold of the interfering gas, it is determined to be the target natural gas; if Stotal is greater than or equal to the threshold of the interfering gas and less than or equal to the threshold of the target gas, it is determined to be the interfering gas.

[0097] A multi-channel optical receiver acquires transmission signals at different wavelengths. Each wavelength reflects the absorption of light by gas molecules at that wavelength. A laser is programmed to simultaneously emit laser beams at multiple wavelengths through the natural gas sample, and the receiving optical system acquires the transmission signals at these wavelengths.

[0098] According to the intensity of spectral absorption, the absorption coefficient is calculated using the Beer-Lambert Law: Where L is the optical path length of the gas sample, I(λ) is the transmitted light intensity, I0(λ) is the incident light intensity, and α(λ) is the absorption coefficient at that wavelength. These data are usually organized into absorption coefficient data at multiple wavelengths.

[0099] According to the Lambert-Beer law, there is a linear relationship between gas concentration and absorption coefficient. The specific expression is: α(λ) = ∈(λ)·c·L; where: α(λ) is the absorption coefficient at wavelength λ, ∈(λ) is the absorption cross section (also called absorption coefficient or absorption spectral density) of natural gas at wavelength λ, c is the gas concentration (usually expressed in ppm or %), and L is the optical path length. Therefore, the gas concentration c can be calculated using the formula:

[0100] Since natural gas concentration usually produces different absorption characteristics at different wavelengths, data at multiple wavelengths can be used to estimate gas concentration. To improve the accuracy of the estimation, the following methods are usually used:

[0101] For multiple wavelengths λ1,λ2,…,λ n , we can perform a weighted average based on the absorption coefficient and the corresponding absorption cross section at each wavelength to obtain an estimated value of the total gas concentration. The concentration calculation formula can be written as: Where n is the total number of wavelengths, α(λ i ) is in λ i The absorption coefficient at ,∈(λ i ) is the absorption cross section at that wavelength.

[0102] To improve accuracy, a regression model can be used to fit the relationship between natural gas concentration and absorption coefficient. Common regression models include linear regression, ridge regression, and support vector regression (SVR). These models can produce more accurate concentration predictions using training sample data.

[0103] For example, the formula for the linear regression model is: pred =a1·α(λ1)+a2·α(λ2)+…+a n ·α(λ n )+b; where a1, a2, …, a n is the coefficient of the regression model, b is the intercept term, α(λ i ) is the absorption coefficient at each wavelength.

[0104] By fitting the training data set, we can get the coefficients a1, a2, ..., a of the model n , and then use the new spectral data to calculate the natural gas concentration.

[0105] After obtaining the concentration of natural gas through the above calculation method, the result can be displayed or output in real time. The specific steps are as follows:

[0106] Based on the absorption spectrum data of multiple wavelengths, spectral decomposition algorithms (such as principal component analysis (PCA) and support vector machine (SVM)) can be used to identify the type of gas.

[0107] Principal Component Analysis (PCA): This method reduces the dimensionality of the absorption spectrum data and extracts the main component features. If the identification result points to natural gas (such as methane), the target gas is determined.

[0108] Support Vector Machine (SVM): Train an SVM model to distinguish natural gas from other gases and perform classification when new spectral data is input.

[0109] The calculated results of natural gas concentration can be output through digital display, chart or alarm system. Output can be in the following ways:

[0110] Digital display: Directly displays the current concentration value in ppm or %.

[0111] Real-time monitoring interface: displays real-time changes in concentration through charts (such as a line chart showing concentration changes over time).

[0112] Alarm system: If the concentration exceeds a preset safety threshold, the alarm system automatically triggers, alerting operators or the system to take appropriate emergency measures. For example, when the natural gas concentration reaches a set threshold (such as 500ppm), the system can trigger an alarm. The real-time data is stored in a database for historical analysis, further predicting gas concentration trends, and supporting subsequent report generation or analysis.

[0113] During real-time monitoring, the system will calculate the gas concentration in real time and compare it with the set safety threshold. If the concentration exceeds the threshold, the system will immediately trigger an alarm and take corresponding control measures, such as closing valves and starting ventilation.

[0114] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0115] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0116] It should be understood that the term "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone, where A and B may be singular or plural. In addition, the character " / " herein generally indicates that the objects associated with each other are in an "or" relationship, but it may also indicate an "and / or" relationship, which can be understood by referring to the context. A person of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0117] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A high-sensitivity natural gas leak laser detection method, characterized by: include: Set multiple laser emission wavelengths, and respectively correspond to the characteristic absorption spectrum of natural gas and its coexisting interfering gas; Utilize multiple tunable lasers to simultaneously emit laser beams at set wavelengths, passing through the gas area to be measured; The transmission signal of each wavelength after passing through the gas is received by a multi-channel optical receiving device to obtain corresponding multi-channel absorption spectrum data; Preprocess the multi-channel absorption spectrum data, extract the spectral anomaly features in the multi-channel spectrum data and input them into the fusion gas absorption feature prediction model for analysis; Distinguish the absorption spectra of target natural gas and interfering gas based on the analysis results and quantitatively calculate the natural gas concentration; Output the natural gas identification results and its concentration information, and use the analysis data for real-time early warning.

2. The high-sensitivity natural gas leak laser detection method according to claim 1, characterized in that: According to the characteristic absorption bands of natural gas and its coexisting interfering gases, the laser wavelengths are selected as follows: methane: 1657nm; carbon dioxide: 1572nm; nitrogen oxides: 2355nm; volatile organic compounds: 1392nm and 1450nm.

3. The high-sensitivity natural gas leak laser detection method according to claim 1, characterized in that: The multiple tunable lasers output set wavelengths by adjusting the driving current or operating temperature of the lasers, and the laser beams of multiple wavelengths are combined through a coaxial design or an optical beam splitting device and then pass through the gas area to be measured.

4. The high-sensitivity natural gas leak laser detection method according to claim 3, characterized in that: The multi-channel optical receiving device includes a plurality of photodetectors, bandpass filters and collimating optical components, and is used to respectively receive and distinguish the transmitted light signal of each laser wavelength.

5. The high-sensitivity natural gas leak laser detection method according to claim 1, characterized in that: The spectral abnormality features include spectral absorption slope jitter value and absorption peak depth abnormality value; wherein, the spectral absorption slope jitter value is obtained by: Obtain multi-wavelength absorption spectrum data. For each wavelength λ, calculate the absorption intensity α(λ) at that wavelength. Calculate the spectral slope by taking the difference of the absorption coefficient at adjacent wavelengths. The expression is: α(λ i ) is at wavelength λ i The absorption coefficient at i ) is the slope of the spectrum, i.e., at wavelength λ i The absorption intensity change rate at the absorption peak is extracted; the fluctuation amplitude around the absorption peak is extracted, and the fluctuation of the slope at multiple wavelength points is calculated to obtain the slope jitter value. The expression is: Where D is the slope jitter value, N is the total number of wavelength data points, |S(λ i )-S(λ i-1 )| means calculating the difference between adjacent slopes and taking the absolute value to represent the magnitude of the slope change.

6. The high-sensitivity natural gas leak laser detection method according to claim 5, characterized in that: in, The method for obtaining the absorption peak depth anomaly value is: Acquire spectral data, including the value of the absorption coefficient α(λ) varying with wavelength λ, and set the spectral data as: α(λ)={α(λ1),α(λ2),…,α(λ n )}; where λ1,λ2,…,λ n is each wavelength point, α(λ) is the corresponding absorption intensity, and n is the total number of wavelengths; calculate the first-order derivative S1(λ i ), which is the rate at which the absorption coefficient changes with wavelength: Among them, S1(λ i ) is at wavelength λ i The first derivative at λ represents the rate of change of the absorption coefficient with wavelength, and the second derivative S2(λ i ), that is, the rate of change of the first-order derivative with wavelength. The second-order derivative can accurately identify the inflection point of the absorption peak, and the expression is: Among them, S2(λ i ) is the second-order derivative, which indicates the rate at which the first-order derivative changes with wavelength. The point where the second-order derivative reverses is the inflection point of the absorption peak. By detecting the second-order derivative S2(λ i ) inversion point, locate the position of the absorption peak; if S2(λ i )>0 and S2(λ i-1 )<0, then λ i is the top of the absorption peak; if S2(λ i )>0 and S2(λ i+1 )>0, then λ i It is the bottom of the absorption peak; the depth of the absorption peak refers to the difference between the top of the absorption peak and the baseline.

7. The high-sensitivity natural gas leak laser detection method according to claim 6, characterized in that: Assuming the top position of the absorption peak is λpeak and the baseline absorption value is αbaseline, the calculation formula for the absorption peak depth Dpeak is: Dpeak = α(λpeak) - αbaseline; where α(λpeak) is the absorption coefficient at the absorption peak and αbaseline is the baseline absorption value; After calculating the depth of each absorption peak, compare it with the preset threshold value, which is set as Dthreshold. If Dpeak is greater than Dthreshold, it is considered that the absorption peak depth is abnormal; if Dpeak is less than or equal to Dthreshold, the absorption peak depth is considered to be within the normal range; The depth values ​​of the absorption peaks greater than the threshold Dthreshold within a fixed time period T are collected and a data set is established. The mean of the data set is calculated as the absorption peak depth outlier.

8. The high-sensitivity natural gas leak laser detection method according to claim 7, characterized in that: The spectral absorption slope jitter value and the absorption peak depth anomaly value are normalized so that they are both between [0, 1]. The comprehensive spectral analysis anomaly value is calculated based on the normalized spectral absorption slope jitter value and absorption peak depth anomaly value.

9. The high-sensitivity natural gas leak laser detection method according to claim 8, characterized in that: Set thresholds for each gas: Target natural gas threshold: Calculate the characteristic value range of the spectrum under known natural gas concentration and set the target gas threshold; Interference gas threshold: Calculate the corresponding threshold based on the absorption spectrum characteristics of the interfering gas; Compare the calculated comprehensive spectrum analysis anomaly value Stotal with the preset threshold: If Stotal is greater than the target gas threshold and less than the interfering gas threshold, it is determined to be target natural gas; if Stotal is greater than or equal to the interfering gas threshold and less than or equal to the target gas threshold, it is determined to be an interfering gas.

10. The high-sensitivity natural gas leak laser detection method according to claim 1, characterized in that: The absorption coefficient is calculated based on the Lambert-Beer law, and the gas concentration is obtained by weighted averaging. The type of target gas and the current concentration value are displayed on the interface in real time, and dynamic monitoring is achieved in combination with the concentration change curve. When the detected concentration exceeds the set safety threshold, the alarm system is triggered and the external control device is linked to respond and control.

Images