Gas concentration inversion method and device

Abstract

The application provides a gas concentration inversion method and device. The method is based on a direct absorption spectrum technology. A driving current of a laser is in a linear function relationship with time, and the driving current of the laser is in a linear relationship with an emission wavelength. The method further comprises the following steps: S1, calibrating and detecting zero gas and standard gas respectively to obtain a fitting calibration coefficient between gas concentration and absorbance, and to obtain a concentration inversion fitting relationship; S2, introducing a to-be-detected gas into a gas chamber, detecting the to-be-detected gas, obtaining time-domain integral absorbance of the to-be-detected gas, substituting the time-domain integral absorbance of the to-be-detected gas into the concentration inversion fitting relationship, and calculating to obtain a concentration value of the to-be-detected gas. The gas concentration inversion method and device provided by the application can simplify the complexity of a device in the related art.

Description

Technical Field

[0001] This application relates to the field of gas detection technology, and in particular to a gas concentration inversion method and apparatus, which can be used for background concentration detection of atmospheric pollutants and monitoring of pollution source emissions, and can also be used for industrial process gas analysis and control. Background Technology

[0002] Tunable Diode Laser Absorption Spectroscopy (TDLAS) is based on molecular absorption spectroscopy theory. When a laser beam passes through the target region, the light is absorbed, resulting in intensity attenuation. For a single absorption of a component, the incident and transmitted light intensities satisfy Beer-Lambert's law. TDLAS technology primarily employs two methods for signal detection: direct absorption and wavelength modulation. Direct absorption spectroscopy involves controlling the laser temperature and drive current to ensure the laser wavelength scan covers the complete absorption peak of the target gas. Concentration information is retrieved by calculating the absorbance. Obtaining the absorbance value requires calculating the integral of the absorption spectral line in the frequency domain. However, the measurement signal obtained using direct absorption spectroscopy is a time-domain dependent function, necessitating a conversion from the time domain to the frequency domain.

[0003] In related technologies, a Fabry-Perot etalon is generally used for conversion. The laser beam is split into two beams by a beam splitter. One beam passes directly through the gas to be measured, while the other is used for the etalon optical path for wavelength calibration. However, this method significantly increases the system complexity. Furthermore, the time-frequency conversion of the etalon introduces errors, and the introduction of the beam splitter increases optical noise in the measurement optical path, affecting measurement accuracy. Additionally, the etalons used for different wavelength lasers are inconsistent, increasing product cost and compatibility challenges. Summary of the Invention

[0004] The purpose of this application is to overcome the problem in related technologies where gas concentration inversion based on direct absorption spectra requires conversion from the time domain to the frequency domain, and the laser beam is split into two beams by a beam splitter, which greatly increases the complexity of the system.

[0005] To address the aforementioned problems, this application provides a gas concentration inversion method. This method is based on TDLAS direct absorption spectroscopy, controlling the laser drive current to ensure the laser wavelength scan covers the complete absorption peak of the gas to be measured. Gas concentration information is inverted by calculating absorbance. The laser drive current has a linear relationship with time and a linear relationship with the emission wavelength.

[0006] The method further includes the following steps:

[0007] S1. Calibrate and detect zero gas and standard gas respectively, obtain the fitting calibration coefficient between gas concentration and absorbance, and obtain the concentration inversion fitting relationship;

[0008] S2. Introduce the gas to be tested into the gas chamber, detect the gas to be tested, obtain the time-domain integrated absorbance of the gas to be tested, substitute the time-domain integrated absorbance of the gas to be tested into the concentration inversion fitting formula, and calculate the concentration value of the gas to be tested.

[0009] Furthermore, the calibration and testing of zero gas and standard gas in step S1 includes the following steps:

[0010] S11. Introduce zero gas into the gas chamber, calculate the time-domain integrated absorbance at this time, and use it as the zero-point coefficient of the system.

[0011] S12. Introduce full-scale standard gas into the gas chamber, calculate the time-domain integrated absorbance at this time, and calculate the system calibration coefficient based on the gas concentration of the full-scale standard gas and the time-domain integrated absorbance at this time, combined with the concentration calculation formula of the direct absorption method.

[0012] S13. Based on the system zero-point coefficient and the system calibration coefficient, obtain the corresponding fitting calibration coefficients between the gas concentration and absorbance, and obtain the concentration inversion fitting relationship.

[0013] Furthermore, the concentration inversion fitting relationship is expressed as follows:

[0014]

[0015] In the formula, A t (λ) represents the time-domain integrated absorbance of the gas to be measured, Pm represents the gas pressure during measurement, Lm represents the absorption optical path length during measurement, b(λ) represents the zero-point coefficient of the system; λ1 and λ2 represent the start and end wavelengths of the laser scan; α(T, P) is a temperature and pressure related correction coefficient; β(λ) is a nonlinear response function describing the correlation of optical elements in the optical path with wavelength; S(T, λ) is a temperature and wavelength related spectral line intensity function; X Calib This indicates the concentration of the standard gas.

[0016] Furthermore, the zero-point coefficients of the system are expressed as follows:

[0017]

[0018] In the formula, A t Zero (λ) represents the time-domain integrated absorbance when zero gas is introduced into the gas chamber.

[0019] Furthermore, the system calibration coefficients are expressed as follows:

[0020]

[0021] In the formula, Pc represents the gas pressure during calibration, Lc represents the absorption optical path length during calibration, and A tCalib (λ) represents the absorbance when full-scale standard gas is introduced into the gas chamber; b(λ) represents the zero-point coefficient of the system; λ1 and λ2 represent the start and end wavelengths of the laser scan; α(T, P) is a temperature and pressure-dependent correction coefficient; β(λ) is a nonlinear response function describing the wavelength dependence of optical components in the optical path; S(T, λ) is a temperature and wavelength-dependent spectral line intensity function; X Calib This indicates the concentration of the standard gas.

[0022] Furthermore, step S2 also includes:

[0023] S21. Set up at least two correction environments, measure the time-domain integrated absorbance of a standard gas with a known concentration in each correction environment, and calculate the theoretical time-domain integrated absorbance based on the gas concentration in each correction environment, the gas concentration of the full-scale standard gas during calibration, and the corresponding time-domain integrated absorbance during calibration.

[0024] S22. Perform polynomial fitting on the time-domain integrated absorbance corresponding to the gas concentration value under each modified environment and the theoretical time-domain integrated absorbance to obtain a fitting model between the time-domain integrated absorbance corresponding to the gas concentration value under the modified environment and the theoretical time-domain integrated absorbance.

[0025] S23. Substitute the time-domain integrated absorbance of the gas to be tested into the fitting model to obtain the corrected time-domain integrated absorbance of the gas to be tested, and substitute the corrected time-domain integrated absorbance of the gas to be tested into the concentration inversion fitting formula to calculate the concentration value of the gas to be tested.

[0026] Furthermore, the theoretical time-domain integrated absorbance is expressed by the following formula:

[0027]

[0028] In the formula, A ln X represents the theoretical time-domain integral absorbance corresponding to the nth corrected environment. fixn X represents the concentration of the standard gas under the nth correction environment. Calib A represents the gas concentration of the full-scale standard gas during calibration. tCalib The time-domain integral absorbance represents the absorbance at calibration. f(T,P,H) is an environmental correction function customized based on experimental data to adjust for the effects of temperature, pressure, and humidity.

[0029] Furthermore, the fitting model between the time-domain integrated absorbance corresponding to the gas concentration value under the modified environment and the theoretical time-domain integrated absorbance is expressed as follows:

[0030]

[0031] In the formula, A ln A represents the theoretical time-domain integral absorbance. rn This represents the time-domain integrated absorbance corresponding to the gas concentration values ​​under the corrected environment, a0, a1, a2…a n The fitting coefficient, ∈(T,P,H), is an error term used to simulate the influence of environmental conditions on the fitting accuracy.

[0032] Furthermore, step S23 also includes:

[0033] S231. Based on the absorbance curve obtained from detecting the gas to be tested, select the non-absorption region data at both ends of the absorbance curve of the gas to be tested and perform first-order polynomial fitting to obtain the baseline signal.

[0034] S232. Calculate the mean of the baseline signal to obtain the baseline mean. Divide the absorbance curve of the gas to be tested and the baseline signal by the baseline mean to obtain the normalized absorption waveform and the normalized baseline of the gas to be tested.

[0035] S233. Perform differential processing on the normalized absorption waveform and the normalized baseline of the gas to be tested to obtain the direct absorption spectrum of the gas to be tested.

[0036] S234. Integrate the direct absorption spectrum of the gas to be tested to obtain the time-domain integrated absorbance of the gas to be tested.

[0037] This application also provides a gas concentration inversion device, the device comprising:

[0038] A gas chamber, used to contain the gas to be measured;

[0039] A laser emitting unit is used to emit a tunable laser. The emission wavelength of the laser emitting unit scans and covers the complete absorption peak of the gas to be measured. The driving current of the laser emitting unit is a linear function of time, and the driving current of the laser emitting unit is linearly related to the emission wavelength.

[0040] The detection unit receives laser light passing through the gas to be tested and obtains the absorbance curve of the gas to be tested.

[0041] The signal processing unit is used to calculate the time-domain integrated absorbance of the gas to be tested based on the absorbance curve of the gas to be tested. Furthermore, the signal processing unit processes the time-domain integrated absorbance of the gas to be tested according to the gas concentration inversion method described above, and calculates the concentration of the gas to be tested.

[0042] Based on the above description, this application provides a gas concentration inversion method and apparatus. Based on TDLAS direct absorption spectroscopy, the driving current of the laser is controlled, and the wavelength of the laser emitted by the laser is linearly related to the driving current. Furthermore, the driving current of the laser is a linear function of time. A calibration fitting formula is introduced to express the relationship between the concentration of the gas to be measured and the time-domain integral absorbance of the gas to be measured in the time domain. This replaces the process of converting the time domain to the frequency domain in related technologies, thereby simplifying the components such as the spectrometer and etalon in related technologies and reducing the complexity of the apparatus. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of a gas concentration inversion device based on direct absorption spectroscopy in related technologies;

[0044] Figure 2 This is a schematic diagram of the gas concentration inversion device provided in the embodiments of this application;

[0045] Figure 3 This is a flowchart of the gas concentration inversion method provided in the embodiments of this application;

[0046] Figure 4 This is a schematic diagram of the original absorbance curve of the gas to be tested provided in the embodiments of this application;

[0047] Figure 5 This is a schematic diagram of the normalized absorption waveform of the gas to be tested provided in an embodiment of this application.

[0048] Figure 6 A schematic diagram of the direct absorption spectrum of the gas to be tested obtained after differential processing, as provided in an embodiment of this application;

[0049] Figure 7 This is a schematic diagram of the absorption peak signal and corresponding slope signal provided in the embodiments of this application. Detailed Implementation

[0050] The present application will be described in detail below with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present application. Any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are included within the protection scope of the present application.

[0051] Direct absorption spectroscopy involves controlling the laser temperature and drive current to ensure the laser wavelength scan covers the complete absorption peak of the gas being measured. Concentration information is retrieved by calculating absorbance. Obtaining the absorbance value requires calculating the integral of the absorption spectral line in the frequency domain. However, the measurement signal obtained using direct absorption spectroscopy is a time-domain dependent function; therefore, it is necessary to convert the time domain to the frequency domain. For example... Figure 1 As shown, in related technologies, the Fabry-Perot etalon is generally used for conversion. The laser beam is split into two beams by a beam splitter. One beam passes directly through the gas to be measured, and the other beam is used for the etalon optical path for wavelength calibration.

[0052] To overcome the problem in related technologies where gas concentration inversion based on direct absorption spectra requires converting the time domain to the frequency domain and splitting the laser beam into two beams via a beam splitter, which greatly increases the complexity of the system.

[0053] This application provides a gas concentration inversion device 100. For example... Figure 2 As shown, the gas concentration inversion device 100 provided in this application embodiment includes: a gas chamber 12, a laser emission unit 11, a detection unit 13, and a signal processing unit 14.

[0054] The laser emitting unit 11 can be a laser, for example, a semiconductor laser, including but not limited to distributed feedback lasers (DFB), vertical-cavity surface-emitting lasers (VCSELs), and interband cascaded lasers (ICLs). The laser emitting unit 11 emits tunable laser light. The emission wavelength of the laser emitting unit 11 scans to cover the complete absorption peak of the gas under test. The driving current of the laser emitting unit 11 is a linear function of time, and the emission wavelength of the laser emitting unit 11 is linearly related to the driving current.

[0055] The gas chamber 12 is used to contain the gas to be measured. As an optional implementation, the gas chamber 12 can be provided with multiple reflective units, which can increase the absorption optical path through multiple reflections and reduce the volume of the gas chamber 12.

[0056] The detection unit 13 is used to receive the laser light passing through the gas to be tested and obtain the absorbance curve of the gas to be tested.

[0057] The signal processing unit 14 is used to calculate the time-domain integrated absorbance of the gas to be tested based on the absorbance curve of the gas to be tested, and to obtain the gas concentration information of the gas to be tested by inversion based on the time-domain integrated absorbance of the gas to be tested.

[0058] Based on the above description, the gas concentration inversion device 100 provided in this application embodiment does not require spectral processing of the laser emitted by the laser emission unit, which greatly simplifies the structure of the device and makes it easier to detect the gas to be tested.

[0059] Based on the gas concentration inversion device 100 provided in this application embodiment, this application embodiment also provides a gas concentration inversion method. The gas to be tested is detected by the gas concentration inversion device 100 provided in this application embodiment to obtain the absorbance curve of the gas to be tested. The gas concentration inversion method provided in this application embodiment is then used to process the absorbance curve of the gas to be tested, and the concentration of the gas to be tested can be calculated.

[0060] like Figure 3 As shown, the gas concentration inversion method provided in this application embodiment further includes the following steps:

[0061] S1. Calibrate and detect zero gas and standard gas respectively, obtain the fitting calibration coefficient between gas concentration and absorbance, and obtain the concentration inversion fitting relationship;

[0062] S2. Introduce the gas to be tested into the gas chamber 12, detect the gas to be tested, obtain the time-domain integrated absorbance of the gas to be tested, substitute the time-domain integrated absorbance of the gas to be tested into the concentration inversion fitting formula, and calculate the concentration value of the gas to be tested.

[0063] As an optional implementation, the laser's driving current adopts a sawtooth wave scanning method so that the laser's driving current has a linear relationship with time.

[0064] As an optional implementation, the gas concentration inversion method provided in this application embodiment includes the following steps in step S1 for calibrating and detecting zero gas and standard gas:

[0065] S11. Introduce zero gas into gas chamber 12, calculate the time-domain integrated absorbance at this time, and use it as the zero-point coefficient of the system.

[0066] For example, N2 can be introduced into gas chamber 12 as zero gas, the time-domain integrated absorbance can be calculated, and this time-domain integrated absorbance can be used as the system zero-point coefficient. The system zero-point coefficient can then be expressed as:

[0067]

[0068] In the formula, A t Zero (λ) represents the absorbance when zero gas is introduced into gas chamber 12.

[0069] S12. Introduce full-scale standard gas into gas chamber 12 and calculate the time-domain integrated absorbance at this time. Based on the gas concentration of the full-scale standard gas and the time-domain integrated absorbance at this time, and combined with the concentration calculation formula of the direct absorption method, calculate the system calibration coefficient.

[0070] For example, if a certain type of gas is used as the analyte, a standard gas mixture containing a known concentration needs to be introduced. Calculating the time-domain integrated absorbance at this point yields the system calibration coefficients. The system calibration coefficients can be expressed as:

[0071]

[0072] In the formula, A tCalib (λ) represents the absorbance when full-scale standard gas is introduced into gas chamber 12; b(λ) represents the zero-point coefficient of the system, which may vary with wavelength; λ1 and λ2 represent the start and end wavelengths of the laser scan; α(T, P) is a temperature and pressure-dependent correction coefficient used to correct for changes in spectral line intensity under different conditions, obtained through tests conducted using the hitran database and laboratory simulations of temperature and pressure conditions; β(λ) is a nonlinear response function describing the correlation of optical components such as lenses and mirrors in the optical path with wavelength; S(T, λ) is a temperature and wavelength-dependent spectral line intensity function, which can be found in the hitran database; X Calib Pc represents the gas concentration of the standard gas, Pc represents the gas pressure during calibration, and Lc represents the absorption optical path length during calibration.

[0073] Spectral data acquisition: During the calibration process, the full-scale standard gas is spectrally scanned from λ1 to λ2 by precisely controlling the temperature and driving current of the laser;

[0074] Data preprocessing: Wavelength-related baseline correction is performed on the collected absorbance data, and b(λ) is used to correct the data at each wavelength point;

[0075] Integration processing: Calculate the integral of the corrected data over the entire wavelength range to obtain A. tCalib (λ);

[0076] Physical parameter correction: The integral results are corrected by applying α(T, P) and β(λ) to account for the effects of temperature, pressure and optical response.

[0077] S13. Based on the system zero-point coefficient and system calibration coefficient, obtain the corresponding fitting calibration coefficients between gas concentration and absorbance, and obtain the concentration inversion fitting relationship.

[0078] As an optional implementation, the system zero-point coefficients can be used as the intercept of the concentration inversion fitting equation. The system calibration coefficients can be used as the slope of the concentration inversion fitting equation. Therefore, the concentration inversion fitting equation can be expressed as:

[0079]

[0080] In the formula, A t(λ) represents the time-domain integrated absorbance of the gas to be measured, Pc and Pm represent the gas pressure during calibration and measurement, respectively, and Lc and Lm represent the absorption optical path length during calibration and measurement, respectively.

[0081] After obtaining the time-domain integrated absorbance of the gas to be tested, the concentration of the gas to be tested can be calculated using the concentration inversion fitting formula provided in the embodiments of this application. There is no need to convert from the time domain to the frequency domain. This simplifies the complexity of the device while meeting the detection accuracy requirements and makes it easier to detect the gas to be tested.

[0082] As an optional implementation, in the gas concentration inversion method provided in this application embodiment, step S2 further includes the following steps:

[0083] S21. Establishment of a Multi-Environment Correction Model

[0084] Multiple environment settings: Define at least two sets of modified environments, each with different temperature, pressure and humidity conditions to simulate various working conditions in real-world applications.

[0085] Dynamic Measurement: Under each correction environment, the time-domain integrated absorbance of a standard gas containing a known concentration was measured using a tuned laser diode (TDLAS) system, while simultaneously recording environmental condition data such as temperature T, pressure P, and humidity H.

[0086] Theoretical absorbance calculation: Based on the physical absorption model, the theoretical absorbance is calculated for each environment. Considering the influence of environmental conditions, an enhanced absorption model formula is used.

[0087]

[0088] In the formula, A ln X represents the theoretical time-domain integral absorbance corresponding to the nth corrected environment. fixn X represents the concentration of the standard gas under the nth correction environment. Calib A represents the gas concentration of the full-scale standard gas during calibration. tCalib This represents the time-domain integral absorbance at calibration. f(T,P,H) is an environmental correction function customized based on experimental data to adjust for the effects of temperature, pressure, and humidity. This function may include nonlinear and interaction terms, such as:

[0089]

[0090] θPH, where the coefficients of each term Optimization based on experimental data

[0091] S22, Statistical Analysis

[0092] Data preprocessing: All measurement data are preprocessed, including noise reduction, baseline correction, and normalization.

[0093] Polynomial fitting: High-order polynomial fitting techniques are used, combined with machine learning algorithms such as random forests or gradient boosting machines, to fit the actual absorbance and theoretical absorbance under each correction environment. The fitting model can be expressed as:

[0094]

[0095] In the formula, A ln A represents the theoretical time-domain integral absorbance. rn This represents the time-domain integrated absorbance corresponding to the gas concentration values ​​under the corrected environment, a0, a1, a2…a n The fitting coefficient, ∈(T,P,H), is an error term used to simulate the influence of environmental conditions on the fitting accuracy.

[0096] S23. Concentration Inversion and Error Correction

[0097] Concentration inversion: Substitute the time-domain integrated absorbance of the gas to be measured into the fitting model to calculate the corrected time-domain integrated absorbance.

[0098] Error Correction: A data-driven error correction model is applied to further optimize the concentration calculation of the analyte gas. This model uses historical data and current environmental parameters to predict and correct for potential errors, using the formula:

[0099] X corrected =X raw .(1+δ(T,P,H))

[0100] Among them, X corrected X is the corrected concentration. raw For theoretical concentration calculations, δ(T,P,H) is an error correction factor learned from historical data. The error correction model will continuously learn from new measurement data and update the correction factor in real time.

[0101] This adaptive learning mechanism enables the system to cope with problems such as sensor aging and environmental changes during long-term operation.

[0102] As an optional implementation, in the gas concentration inversion method provided in this application embodiment, step S23 further includes:

[0103] S231. Based on the absorbance curve obtained from the gas to be tested, select the data of the non-absorption regions at both ends of the absorbance curve of the gas to be tested and perform first-order polynomial fitting to obtain the baseline signal.

[0104] S232. Calculate the mean of the baseline signal to obtain the baseline mean. Divide the absorbance curve of the gas to be tested and the baseline signal by the baseline mean to obtain the normalized absorption waveform and the normalized baseline of the gas to be tested.

[0105] S233. Perform differential processing on the normalized absorption waveform and the normalized baseline of the gas to be measured to obtain the direct absorption spectrum of the gas to be measured.

[0106] S234. Integrate the direct absorption spectrum of the gas to be tested to obtain the time-domain integrated absorbance of the gas to be tested.

[0107] It should be noted that for steps S232 to S233, the conventional technical solution is: direct absorption spectrum = -log(absorbance curve / fitted baseline). In the non-absorption region, the background interference signal will cause the fitted baseline to have negative values. Since negative numbers do not have logarithmic values, calculation errors will occur. The signal processing method of the present invention can effectively avoid this problem.

[0108] The integration region can be set in the software. Generally, a wider integration region is less sensitive to collision broadening under different background gases, has weaker requirements for gas composition consistency during calibration and measurement, and is suitable for applications where gas collision broadening has a significant impact. A narrower integration region results in better drift performance and more stable measurements. For multi-component measurements, integration regions for different components can be set separately to achieve simultaneous detection of multiple components.

[0109] Further, in step S231, data points are selected from the non-absorption regions at both ends of the absorbance curve of the gas to be tested. Since SG filtering results in a fixed value window before and after the absorbance curve, points with the first 20 and last 20 points on the x-axis are generally discarded. Additionally, points between 100 and 200, where the absorption region is located, should also be avoided. Generally, ten points are selected for fitting, such as 30 / 40 / 50 / 60 / 70 / 230 / 240 / 250 / 260 / 270. Figure 4 As shown, the number of fitting points can be increased when the signal is not stable. The baseline signal fitting equation is a linear function, and the fitting correlation coefficient is required to be greater than 0.99.

[0110] like Figure 5 As shown, it illustrates a schematic diagram of the normalized absorption waveform of the gas to be tested provided in an embodiment of this application.

[0111] like Figure 6 As shown, it illustrates a schematic diagram of the direct absorption spectrum of the gas to be tested obtained after differential processing, according to an embodiment of this application. Figure 6 By integrating the direct absorption spectrum of the gas to be tested, the time-domain integrated absorbance of the gas can be obtained.

[0112] Calculating the time-domain integrated absorbance of the gas to be measured requires obtaining the area enclosed by the direct absorption spectrum in the absorption range. The key step lies in the identification of the spectral peaks.

[0113] As an optional implementation, the gas concentration inversion method provided in this application also includes the following peak identification step:

[0114] S31. Calculate the slope of each point on the absorption peak of the direct absorption spectrum.

[0115] Among them, the point where the slope gradually increases from close to 0 is the starting point of the absorption peak, the point where the slope changes from positive to negative is the peak point of the absorption peak, and the point where the slope gradually approaches 0 from negative is the ending point of the absorption peak.

[0116] The slope can be calculated using six consecutive data points, X. i The slope at point can be expressed as:

[0117] K i =(5(X) i -X i-1 )+3(X i+1 -X i-2 )+(X i+2 -X i-3 )) / 9

[0118] S32. Set the width of the data window. The data window contains n sequentially sampled direct absorption spectrum data. Slide the window and calculate the slope corresponding to each data point in the window.

[0119] Specifically, the n data points in the data window are denoted as X0, X1, ..., X... n-1 And calculate the slopes K0, K1, ..., K at each point respectively. n-1 After each processing step, slide the data window one position to the left, that is, remove the leftmost data point from the window and move in a data point from the rightmost position, and calculate the slope of each data point.

[0120] It is worth noting that when the data window width is set to n (where n is an even number), the slope values ​​of the first n / 2 and last n / 2 numbers of the normalized absorption waveform are 0 by default, and the slope values ​​are calculated starting from the n / 2+1th number.

[0121] S33, Peak feature point determination. For example... Figure 7As shown, taking the width n = 4 of the data window as an example, the slopes of each data unit are compared in sequence according to the sampling point order. When the slopes of at least two consecutive data units exceed the positive slope threshold PT (which can be set), and the data satisfies 0 < K0 < PT < K1 < K2 < K3, it is considered that a peak is detected, and the value X0 on the left side of the data window is taken as the starting point A of this peak; continue to search forward for the vertex. When the slopes of at least two consecutive data units in the window are lower than the positive slope threshold PT, and the data in the window satisfies K0 > PT > K l > K2 > K3, it is considered that the original signal starts to enter the peak vertex region; there should be at least FW / 2 units (minimum half-peak width) between A and B, otherwise it is considered a spike interference and should be discarded (after discarding, start searching for the starting point of the peak again). Pushing forward from point B, when the slopes of at least two consecutive data units in the data window are lower than the negative slope threshold -PT, and the data in the window satisfies 0 > K0 > -PT > K l > K2 > K3, such as Figure 7 the point C in, then take the highest point in the middle of BC as the vertex. Pushing forward from point B, if the point C of the spectrogram is not found, then discard the current starting point A and the inflection point B, and search for the starting point A of the peak again. When the slopes of at least two consecutive data units in the window are greater than the negative slope threshold -PT, and satisfy K0 < -PT < K1 < K2 < K3, it means that the end point or valley point E of the peak is found.

[0122] The present application also provides a gas concentration inversion device, and the device includes:

[0123] A gas chamber for containing the gas to be measured;

[0124] A laser emission unit for emitting tunable laser, the emission wavelength of the laser emission unit scans and covers the complete absorption peak of the gas to be measured, the drive current of the laser emission unit has a linear function relationship with time, and moreover, the drive current of the laser emission unit has a linear relationship with the emission wavelength;

[0125] A detection unit for receiving the laser passing through the gas to be measured and obtaining the absorbance curve of the gas to be measured;

[0126] A signal processing unit for calculating the time-domain integral absorbance of the gas to be measured according to the absorbance curve of the gas to be measured, and moreover, the signal processing unit also processes the time-domain integral absorbance of the gas to be measured according to the above gas concentration inversion method to calculate and obtain the concentration of the gas to be measured.

[0127] Based on the above description, this application provides a gas concentration inversion method and apparatus. Based on TDLAS direct absorption spectroscopy, the driving current of a laser is controlled, and the wavelength of the emitted laser is linearly related to the driving current. Furthermore, the driving current of the laser is a linear function of time. A concentration inversion fitting formula is introduced to express the relationship between the concentration of the gas to be measured and the time-domain integrated absorbance of the gas in the time domain. This replaces the time-domain to frequency-domain conversion process in related technologies, thereby simplifying components such as spectrometers and etalons in related technologies and reducing the complexity of the apparatus. Simultaneously, the signal processing method of this method improves measurement accuracy and stability compared to related technologies.

[0128] The above-disclosed embodiments are merely preferred embodiments of this application, but are not intended to limit the scope of this application. Those skilled in the art will understand that any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and scope of this application and the appended claims are equivalent substitutions and still fall within the scope of this application.

Claims

1. A method for inverting gas concentration, characterized in that, The method is based on TDLAS direct absorption spectroscopy, controlling the laser drive current to ensure the laser wavelength scan covers the complete absorption peak of the gas to be measured. Gas concentration information is retrieved by calculating absorbance. The laser drive current has a linear relationship with time and a linear relationship with the emission wavelength. The method further includes the following steps: S1. Calibrate and detect zero gas and standard gas respectively, obtain the fitting calibration coefficient between gas concentration and absorbance, and obtain the concentration inversion fitting relationship; S2. Introduce the gas to be tested into the gas chamber, detect the gas to be tested, obtain the time-domain integrated absorbance of the gas to be tested, substitute the time-domain integrated absorbance of the gas to be tested into the concentration inversion fitting formula, and calculate the concentration value of the gas to be tested. The calibration and testing of zero gas and standard gas in step S1 includes the following steps: S11. Introduce zero gas into the gas chamber, calculate the time-domain integrated absorbance at this time, and use it as the zero-point coefficient of the system. S12. Introduce full-scale standard gas into the gas chamber, calculate the time-domain integrated absorbance at this time, and calculate the system calibration coefficient based on the gas concentration of the full-scale standard gas and the time-domain integrated absorbance at this time, combined with the concentration calculation formula of the direct absorption method. S13. Based on the system zero-point coefficient and the system calibration coefficient, obtain the corresponding fitting calibration coefficient between gas concentration and absorbance, and obtain the concentration inversion fitting relationship. The concentration inversion fitting relationship is expressed as follows: ； In the formula, Pm represents the time-domain integrated absorbance of the gas being measured, Pm represents the gas pressure during measurement, and Lm represents the absorption optical path length during measurement. b Represents the zero-point coefficients of the system; and The start and end wavelengths of the laser scan are represented; α(T, P) is a temperature and pressure-related correction coefficient; β(λ) is a nonlinear response function describing the correlation of optical elements in the optical path with wavelength. It is a spectral line intensity function that is temperature- and wavelength-dependent; K Indicates the system calibration coefficients; X This represents the concentration of the gas to be measured obtained from the concentration inversion fitting.

2. The gas concentration inversion method according to claim 1, characterized in that, The zero-point coefficients of the system are expressed as follows: b ； In the formula, A t Zero This represents the time-domain integral absorbance when zero gas is introduced into the gas chamber.

3. The gas concentration inversion method according to claim 2, characterized in that, The system calibration coefficients are expressed as follows: ； In the formula, Pc represents the gas pressure during calibration, and Lc represents the absorption optical path length during calibration. A t Calib This indicates the absorbance when a full-scale standard gas is introduced into the gas chamber. b Represents the zero-point coefficients of the system; and The start and end wavelengths of the laser scan are represented; α(T, P) is a temperature and pressure-related correction coefficient; β(λ) is a nonlinear response function describing the correlation of optical elements in the optical path with wavelength. It is a spectral line intensity function that is temperature- and wavelength-dependent; X Calib This indicates the concentration of the standard gas.

4. The gas concentration inversion method according to claim 1, characterized in that, Step S2 also includes: S21. Set up at least two correction environments, measure the time-domain integrated absorbance of a standard gas with a known concentration in each correction environment, and calculate the theoretical time-domain integrated absorbance based on the gas concentration in each correction environment, the gas concentration of the full-scale standard gas during calibration, and the corresponding time-domain integrated absorbance during calibration. S22. Perform polynomial fitting on the time-domain integrated absorbance corresponding to the gas concentration value under each modified environment and the theoretical time-domain integrated absorbance to obtain a fitting model between the time-domain integrated absorbance corresponding to the gas concentration value under the modified environment and the theoretical time-domain integrated absorbance. S23. Substitute the time-domain integrated absorbance of the gas to be tested into the fitting model to obtain the corrected time-domain integrated absorbance of the gas to be tested, and substitute the corrected time-domain integrated absorbance of the gas to be tested into the concentration inversion fitting formula to calculate the concentration value of the gas to be tested.

5. The gas concentration inversion method according to claim 4, characterized in that, The theoretical time-domain integral absorbance is expressed by the following formula: ； In the formula, A ln This represents the theoretical time-domain integral absorbance corresponding to the nth modified environment. X fixn This represents the concentration of the standard gas under the nth correction environment. X Calib This indicates the gas concentration of the full-scale standard gas during calibration. A t Calib This represents the time-domain integral absorbance at calibration. It is an environmental correction function customized based on experimental data, used to adjust the effects of temperature, pressure and humidity.

6. The gas concentration inversion method according to claim 4, characterized in that, The fitting model between the time-domain integrated absorbance corresponding to the gas concentration value under the modified environment and the theoretical time-domain integrated absorbance is expressed as follows: ； In the formula, A ln Represents the theoretical time-domain integral absorbance. A rn This represents the time-domain integrated absorbance corresponding to the gas concentration value under the corrected environment. , , … Represents the fitting coefficient. It is an error term used to simulate the effect of environmental conditions on the fitting accuracy.

7. The gas concentration inversion method according to claim 6, characterized in that, Step S23 further includes: S231. Based on the absorbance curve obtained from detecting the gas to be tested, select the non-absorption region data at both ends of the absorbance curve of the gas to be tested and perform first-order polynomial fitting to obtain the baseline signal. S232. Calculate the mean of the baseline signal to obtain the baseline mean. Divide the absorbance curve of the gas to be tested and the baseline signal by the baseline mean to obtain the normalized absorption waveform and the normalized baseline of the gas to be tested. S233. Perform differential processing on the normalized absorption waveform and the normalized baseline of the gas to be tested to obtain the direct absorption spectrum of the gas to be tested. S234. Integrate the direct absorption spectrum of the gas to be tested to obtain the time-domain integrated absorbance of the gas to be tested.

8. A gas concentration inversion device, characterized in that, The device includes: A gas chamber, used to contain the gas to be measured; A laser emitting unit is used to emit a tunable laser. The emission wavelength of the laser emitting unit scans and covers the complete absorption peak of the gas to be measured. The driving current of the laser emitting unit is a linear function of time, and the driving current of the laser emitting unit is linearly related to the emission wavelength. The detection unit receives laser light passing through the gas to be tested and obtains the absorbance curve of the gas to be tested. The signal processing unit further processes the time-domain integrated absorbance of the gas to be tested according to the gas concentration inversion method according to any one of claims 1 to 7, and calculates the concentration of the gas to be tested.

Images