A broadband cavity enhanced absorption spectroscopy system detection optimization method, medium and device

Abstract

The application discloses a broadband cavity enhanced absorption spectrum system detection optimization method, medium and equipment, the method first calibrates the broadband cavity enhanced absorption spectrum detection system, obtains the effective optical path curve and the absorption cross section of the to-be-detected gas of the system, and then constructs a plurality of spectral fitting band parameters according to the intensity of the absorption peak in the effective optical path curve and the absorption cross section of the to-be-detected gas. Actual observation is carried out, and an atmospheric actual measurement spectrum diagram is obtained. The actual measurement spectrum diagram is first subjected to spectral fitting by using low-range spectral fitting parameters, and the fitting result is judged. If the fitting result is within the measurement range of the spectral fitting parameters, the result is output. If the range of the fitting result exceeds the measurement range, higher-range spectral fitting parameters are replaced, spectral fitting is carried out again, and until the spectral fitting result meets the measurement range, the result is output.

Description

Technical Field

[0001] This invention relates to the field of atmospheric trace gas detection technology, specifically to a method, medium, and equipment for optimizing the detection of a broadband cavity enhanced absorption spectroscopy system. Background Technology

[0002] With the rapid development of my country's economy, air pollution has become a widely recognized problem. To meet various needs such as the types of gases measured, high measurement accuracy, and different application environments, various technologies for monitoring atmospheric pollution have emerged. Broadband cavity enhanced absorption spectroscopy is one of the most advanced trace gas detection technologies internationally. This technology achieves qualitative and quantitative measurement by measuring the "fingerprint" characteristic absorption of light radiation by trace gas components. Its core lies in the design of its optical cavity. Based on the principle of a resonant cavity, the incident light undergoes multiple reflections within the optical cavity, significantly increasing the optical path and thus greatly improving the detection sensitivity of trace gases, achieving a resolution of 10⁻⁶ ppt(10⁻⁶). -12 It boasts advantages such as high sensitivity, high precision, high stability, high temporal resolution, and real-time online detection. Furthermore, by avoiding chemical interference, the measurement results are more accurate and reliable. Since its inception over two decades ago, broadband cavity enhanced absorption spectroscopy has been successfully used for highly sensitive measurements of various atmospheric trace gases, such as NO2, NO3, SO2, I2, IO, HONO, and CHOCHO. The measurement range is one of the important indicators for evaluating the performance of measurement and analysis instruments; it determines the instrument's applicable range and thus affects its performance. While broadband cavity enhanced absorption spectroscopy is a highly sensitive detection technique, primarily used for measuring trace gases, it may exceed the instrument's detection limit in situations involving pollution source emissions or moderate / heavy pollution, affecting the accuracy of the measurement results. Summary of the Invention

[0003] The present invention proposes a detection optimization method for a broadband cavity enhanced absorption spectroscopy system, which can at least solve one of the technical problems in the background art.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A method for optimizing the detection of a broadband cavity-enhanced absorption spectroscopy system, characterized by comprising the following steps:

[0006] S1. Calibrate the broadband cavity enhanced absorption spectroscopy detection system and obtain the effective optical path curve and the absorption cross section of the gas to be measured.

[0007] S2. Based on the effective optical path curve and the intensity of the absorption peak in the cross section of the gas to be measured, construct multiple spectral fitting band parameters;

[0008] S3. Conduct actual observations to obtain measured atmospheric spectra, and use spectral fitting parameters to perform spectral fitting on the measured spectra;

[0009] S4. Based on the measurement range, determine whether the spectral fitting result meets the requirements. If not, change the spectral fitting parameters and repeat step S3 until the spectral system measurement optimization is completed.

[0010] Furthermore, the method for obtaining the effective optical path curve of the spectral detection system and the absorption cross section of the gas to be measured in step S1 of the present invention includes:

[0011] The optical cavity is filled with high-purity helium and nitrogen gas respectively. The reflectivity R(λ) of the cavity mirror is calculated based on the difference in Rayleigh scattering between helium and nitrogen gas, as shown in the following formula:

[0012]

[0013] Among them, I He (λ) and These are the spectral intensities after the cavity is filled with helium and nitrogen, respectively. and denoted as Rayleigh scattering coefficients for helium and nitrogen, respectively, and d is the effective cavity length.

[0014] Calculate the effective optical path L from the cavity mirror reflectivity and cavity length. eff (λ), the formula is as follows:

[0015]

[0016] The absorption cross section σ(λ) of the gas to be measured is obtained by convolving internationally accepted standard data with instrument functions.

[0017] Furthermore, the spectral fitting band in step S2 of the present invention includes:

[0018] Low-range spectral fitting parameters: The fitting band covers the region with long effective optical path and large absorption peak;

[0019] High-range spectral fitting parameters: The fitted band covers the region with short effective optical path and small absorption peak.

[0020] Furthermore, the specific method for constructing multiple spectral fitting band parameters in this invention includes:

[0021] S210. Determine the fitted band:

[0022] Select a band with obvious absorption characteristics: Based on the absorption spectrum of the gas to be measured, select a band with obvious absorption peaks that are not disturbed by other gases.

[0023] Divide the spectrum into multiple sub-bands: Divide the entire absorption spectrum into multiple sub-bands, each of which contains one or more absorption peaks;

[0024] Avoid interference bands: exclude bands that may cause interference from other gases and environmental factors;

[0025] S220. Construct a spectral fitting model;

[0026] According to the Beer-Lambert law, the absorption coefficient α is [value missing] within each sub-band. i (λ) is related to gas concentration C and absorption cross section σ. i (λ) and effective optical path L eff，i The relationship of (λ) is:

[0027]

[0028] For multi-component gases, the model can be extended to:

[0029]

[0030] Where m is the number of gas components; I0(λ) is the reference spectrum when the optical cavity is filled with high-purity nitrogen and no gas is being measured; I(λ) is the measurement spectrum when the optical cavity is filled with the gas being measured.

[0031] By jointly fitting the absorption coefficient data of multiple sub-bands, a global optimization model is constructed:

[0032]

[0033] Where n is the number of sub-bands, α i (λ) is the absorption coefficient of the i-th sub-band;

[0034] S230, Optimize the fitting parameters;

[0035] Based on the calibration results, the absorption cross section σ for each sub-band is... i (λ) and effective optical path L eff，i (λ) Sets the initial value;

[0036] Optimizing fitting parameters using genetic algorithms minimizes the residuals between the model and experimental data. Regularization methods improve model applicability and prevent overfitting. Specific fitting parameter optimization methods include:

[0037] S231. Define the optimization problem;

[0038] Define the objective function:

[0039]

[0040] Where θ is the parameter to be optimized. For the measured absorbance coefficient, Predict the absorption coefficient for the model.

[0041] Adding a regularization term, we get

[0042] Where, θ j It is the j-th parameter in the parameter vector to be optimized.

[0043] The objective function is used to optimize the gas concentration C, the background term B(λ), and the absorption cross-section correction coefficient K.

[0044] A set of initial parameter values ​​is randomly generated, forming an initial parameter group. Each individual represents a possible set of parameter combinations. For each individual, the objective function f is calculated. reg (θ) is used as the fitness; by randomly selecting several individuals and retaining the individual with the best fitness, each parameter is swapped with a certain probability, and crossover is performed on the selected individuals to generate new individuals. New values ​​are randomly generated within the parameter range to increase the diversity of the parameter group; the selection, crossover, and mutation operations are repeated until the termination condition is met;

[0045] The crossover probability is 0.7-0.9, the mutation probability is 0.01-0.1, and the parameter cluster size is 50-200.

[0046] S240. Verify the fitting results and determine the fitting parameters;

[0047] The absorption cross section σ obtained by fitting i (λ) and effective optical path L eff，i (λ) is used for actual measurements, and the gas concentration C′ is retrieved through multi-band joint inversion:

[0048]

[0049] Furthermore, the spectral system measurement optimization method in step S4 of the present invention includes:

[0050] The relative error is obtained by comparing the concentration C0 of the gas to be measured with the gas concentration C. When the relative error is less than 0.5%, the fitting result meets the requirements and the optimization ends; otherwise, the spectral fitting parameters are changed and step S3 is repeated.

[0051] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0052] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0053] As can be seen from the above technical solution, the broadband cavity enhanced absorption spectroscopy system detection optimization method of the present invention constructs two or more spectral fitting band parameters based on the effective optical path curve and the intensity of the absorption peak in the cross-section of the gas to be measured. Different range spectral fitting bands are used sequentially from low to high to participate in the fitting, which can improve the measurement range without changing the hardware of the original broadband cavity enhanced absorption spectroscopy system. Especially in field experiments encountering periods of high pollution or pollution sources, it can still accurately measure the concentration of pollutants. Attached Figure Description

[0054] Figure 1 This is a flowchart of the method in an embodiment of the present invention;

[0055] Figure 2 This is the standard spectrum of SO2;

[0056] Figure 3 This is the measured spectrum of SO2. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0058] like Figure 1 As shown in this embodiment, a detection optimization method for a broadband cavity enhanced absorption spectroscopy system includes the following steps:

[0059] S1. Calibrate the broadband cavity enhanced absorption spectroscopy detection system to obtain the effective optical path curve and the absorption cross section of the gas to be measured.

[0060] S2. Based on the effective optical path curve and the intensity of the absorption peak in the cross section of the gas to be measured, construct multiple spectral fitting band parameters;

[0061] S3. Conduct actual observations to obtain measured atmospheric spectra, and use spectral fitting parameters to perform spectral fitting on the measured spectra;

[0062] S4. Based on the measurement range, determine whether the spectral fitting result meets the requirements. If not, change the spectral fitting parameters and repeat step S3 until the spectral system measurement optimization is completed.

[0063] Taking the measurement of SO2 gas using a broadband cavity enhanced absorption spectroscopy system with a center wavelength of 365 nm as an example, the steps are explained in detail:

[0064] S1. Calibrate the broadband cavity enhanced absorption spectroscopy detection system to obtain the effective optical path curve of the spectral detection system and the absorption cross section of the gas to be measured.

[0065] The optical cavity is filled with high-purity helium and nitrogen gas respectively. The reflectivity R(λ) of the cavity mirror is calculated based on the difference in Rayleigh scattering between helium and nitrogen gas, as shown in the following formula:

[0066]

[0067] Among them, I He (λ) and These are the spectral intensities after the cavity is filled with helium and nitrogen, respectively. and denoted as Rayleigh scattering coefficients for helium and nitrogen, respectively, and d is the effective cavity length.

[0068] Calculate the effective optical path L from the cavity mirror reflectivity and cavity length. eff (λ), the formula is as follows:

[0069]

[0070] The absorption cross section σ(λ) of the gas to be measured is obtained by convolving internationally accepted standard data with instrument functions.

[0071] S2. Based on the effective optical path curve and the intensity of the absorption peak in the cross section of the gas to be measured, construct multiple spectral fitting band parameters;

[0072] The fitting bands mainly cover the regions with long effective optical path lengths and large absorption peaks, which are used as low-range spectral fitting parameters. The fitting bands mainly cover the regions with short effective optical path lengths and small absorption peaks, which are used as high-range spectral fitting parameters. The measurement ranges of different range spectral fitting parameters are set.

[0073] In trace gas detection, constructing parameters for multiple spectral fitting bands is a crucial step in improving measurement accuracy and reliability. By selecting multiple bands for fitting, noise interference can be reduced, the signal-to-noise ratio improved, and gas concentration more accurately retrieved. Specific methods for constructing parameters for multiple spectral fitting bands include:

[0074] S210. Determine the fitted band:

[0075] Select a band with obvious absorption characteristics: Based on the absorption spectrum of the gas to be measured (e.g., SO2 in the ultraviolet band, CO2 in the infrared band), select a band with obvious absorption peaks that are not disturbed by other gases.

[0076] Divide the spectrum into multiple sub-bands: Divide the entire absorption spectrum into multiple sub-bands, each of which contains one or more absorption peaks;

[0077] Avoid interfering bands: Eliminate bands that may cause interference from other gases or environmental factors (such as water vapor, CO2).

[0078] S220. Construct a spectral fitting model;

[0079] According to the Beer-Lambert law, the absorption coefficient α is [value missing] within each sub-band. i (λ) is related to gas concentration C and absorption cross section σ. i (λ) and effective optical path L eff，i The relationship of (λ) is:

[0080]

[0081] For multi-component gases, the model can be extended to:

[0082]

[0083] Where m is the number of gas components; I0(λ) is the reference spectrum when the optical cavity is filled with high-purity nitrogen and no gas is being measured; I(λ) is the measurement spectrum when the optical cavity is filled with the gas being measured.

[0084] By jointly fitting the absorption coefficient data of multiple sub-bands, a global optimization model is constructed:

[0085]

[0086] Where n is the number of sub-bands, α i (λ) is the absorption coefficient of the i-th sub-band, and θ is the parameter to be optimized (such as gas concentration, background term, etc.). For the measured absorbance coefficient, Predict the absorption coefficient for the model;

[0087] S230, Optimize the fitting parameters;

[0088] Based on the calibration results, the absorption cross section σ for each sub-band is... i (λ) and effective optical path L eff，i (λ) Sets the initial value;

[0089] Optimizing fitting parameters using genetic algorithms minimizes the residuals between the model and experimental data. Regularization methods improve model applicability and prevent overfitting. Specific fitting parameter optimization methods include:

[0090] S231. Define the optimization problem;

[0091] Define the objective function:

[0092]

[0093] Where θ is the parameter to be optimized (such as gas concentration, background term, etc.). For the measured absorbance coefficient, Predict the absorption coefficient for the model.

[0094] Adding a regularization term, we get

[0095] Where, θ j It is the j-th parameter in the parameter vector to be optimized.

[0096] The objective function is used to optimize the gas concentration C, the background term B(λ), and the absorption cross-section correction coefficient K.

[0097] A set of initial parameter values ​​is randomly generated, forming an initial parameter group. Each individual represents a possible set of parameter combinations. For each individual, the objective function f is calculated. reg (θ) is used as the fitness; by randomly selecting several individuals and retaining the individual with the best fitness, each parameter is swapped with a certain probability, and crossover is performed on the selected individuals to generate new individuals. New values ​​are randomly generated within the parameter range to increase the diversity of the parameter group; the selection, crossover, and mutation operations are repeated until the termination condition is met;

[0098] The crossover probability is 0.7-0.9, the mutation probability is 0.01-0.1, and the parameter cluster size is 50-200.

[0099] S240. Verify the fitting results and determine the fitting parameters;

[0100] The absorption cross section σ obtained by fitting i (λ) and effective optical path L eff，i (λ) is used for actual measurements, and the gas concentration C′ is retrieved through multi-band joint inversion:

[0101]

[0102] The spatiotemporal distribution characteristics of gas concentration were analyzed, and the results were corrected based on environmental parameters.

[0103] S4. Based on the measurement range, determine whether the spectral fitting result meets the requirements. If not, change the spectral fitting parameters and repeat step S3 until the spectral system measurement optimization is completed.

[0104] The relative error is obtained by comparing the concentration C0 of the gas to be measured with the gas concentration C. When the relative error is less than 0.5%, the fitting result meets the requirements and the optimization ends; otherwise, the spectral fitting parameters are changed and step S3 is repeated.

[0105] The measurement of SO2 gas using a broadband cavity enhanced absorption spectroscopy system with a center wavelength of 365 nm will be explained in detail as an example.

[0106] The fundamental principle of cavity-enhanced absorption spectroscopy is the Lambert-Beer law. While the Lambert-Beer law is linear within its applicable range, it deviates from linearity when the concentration is above or below a threshold, leading to significant measurement errors. As shown in the formula above, gas concentration is negatively correlated with both lens reflectivity and the reference cross-section. Therefore, reducing lens reflectivity alone, using a weak absorption reference cross-section alone, or reducing both simultaneously can raise the measurement upper limit of a known system. This invention, based on the last approach, proposes that selecting a band with both low reflectivity and a weak absorption cross-section during fitting can improve the measurement upper limit of the cavity-enhanced spectroscopy system. The advantage of this invention lies in this: it achieves an increased upper limit without changing the instrument setup.

[0107] like Figure 1 The method shown is for improving the measurement range of cavity-enhanced absorption spectroscopy. The feasibility of this method is verified by the following experiments, the specific steps of which are as follows:

[0108] 1. First, calibrate the system to obtain the effective optical path curve and the absorption cross section of the gas to be measured.

[0109] 2. Construct parameters for multiple spectral fitting bands based on the effective optical path curve and absorption peak intensity;

[0110] like Figure 2 The image shows the standard spectrum of sulfur dioxide (SO2). The black line represents the spectrum of the absorption peaks of the SO2 absorption cross section. The red line represents the effective optical path curve of the calibrated broadband cavity enhanced absorption system. The blue and green boxes represent multiple spectral fitting parameters constructed based on the effective optical path curve and absorption peak intensities. The green box represents the 367.7-379.0 nm band, which is the low-range spectral fitting parameter, while the blue box represents the 356.2-367.1 nm fitting band, which is the high-range spectral fitting parameter.

[0111] 3. High-purity nitrogen gas is introduced to first obtain the background spectrum I0(λ), and then measurements are carried out.

[0112] 4. Introduce standard SO2 gas of different concentrations, decreasing the concentration from 600 ppm to 60 ppm in sequence, decreasing by 30 ppm each time, and obtain the absorption spectrum I(λ) of SO2 at different concentrations.

[0113] 5. First, use the low-range spectral fitting parameters to perform spectral fitting on the absorption spectrum to obtain the results of spectral fitting using the low-range spectral fitting parameters.

[0114] like Figure 3 The figures show the fitting results using low-range and high-range spectral fitting parameters, respectively. The black line represents the standard SO2 concentration, the blue line represents the spectral fitting result using the low-range spectral fitting parameters, and the red line represents the spectral fitting result using the high-range spectral fitting parameters.

[0115] 6. For the same measured spectrum, comparing the results obtained by fitting different range spectral fitting parameters, it was found that in the low concentration range (such as SO2 standard concentration less than 200 ppm), the results of fitting using low range spectral fitting parameters are basically consistent with the standard concentration.

[0116] As shown in Table 1, as the standard SO2 concentration increases, using the same spectral fitting parameters for spectral fitting leads to a gradual increase in the difference between the fitted spectrum and the standard concentration. However, reusing higher-range spectral fitting parameters results in a spectral fit that is essentially consistent with the standard concentration.

[0117] Table 1 Comparison of inversion results for different measurement ranges

[0118]

[0119] The procedure for optimizing the fitting results by changing the fitting parameters is as follows:

[0120] import numpy as np from deap import base,creator,tools,algorithms

[0121] # Define the regularization objective function

[0122] def regularized_objective_function(individual):C, B, k=individual#Package parameters

[0123] #Computational Model Spectrum

[0124] A_model=k*sigma(lambda)*c*L_eff(lambda)+B

[0125] #Calculate the sum of squared residuals

[0126] residual=np.sum((A_meas-A_model)**2)

[0127] #Add L2 regularization term

[0128] regularization=lambda_value*(C**2+B**2+k**2)

[0129] return residual+regularization,

[0130] #Creating the Houchuan Algorithm Framework

[0131] creator.create("FitnessMin" base.Fitness,weights=(-1.0,))creator.create("Indivi dual",list,fitness=creator.FitnessMin

[0132] to0lbox = base.Toolbox()

[0133] toolbox.register("attr_c"np.random.uniform,0,100) # Gas concentration range

[0134] toolbox.register("attr_B", np.random.uniform.-0.1, 0.1) # Background item range

[0135] toolbox.register("attr_k",np.random.uniform,0.9,1.1) # Range of correction coefficients

[0136] toolbox.register("individual",tools.initCycle,creator.Individual(toolbox.attr_C, toolbox.attr_Btoolbox.attr_k), n=1) toolbox.register("population",tools.initRepeat,list,toolbox.individual)

[0137] toolbox.register("evaluate",regularized_objective_functiontoolbox.register("ma te",tools.cxBlend,alpha=0.5)#Cross replacement

[0138] `toolbox.register("mutate"too1s.mutGaussian, mu=0, sigma=0.1, indpb=0.1)` # Mutation operation. `toolbox.register("select", tools.selTournament, tournsize=3)` # Selection operation.

[0139] #Run the communication algorithm

[0140] population=toolbox.population(n=50)

[0141] algorithms.easimple(population,toolbox,cxpb=0.7,mutpb=0.1,ngen=100,

[0142] verbose-True)

[0143] # Output the optimal solution

[0144] best_individual=tools.selBest(population,k=1)[0].

[0145] 7. When conducting actual measurements on the broadband cavity enhanced absorption spectroscopy measurement system being tested, first use low-range spectral fitting parameters (367.7-379.0 nm) for spectral fitting. If the SO2 fitting result exceeds 200 ppm, using fixed spectral fitting parameters will affect the measurement accuracy of the system, i.e., it will exceed the upper limit of the system's measurement.

[0146] At this point, the high-range spectral fitting parameters (356.2-367.1 nm) should be used for spectral fitting. The results are shown in Table 1, which ensures the accuracy of the measurement results and improves the upper limit of the system's measurement.

[0147] Therefore, the above verification experiments demonstrate that by constructing two or more spectral fitting band parameters based on the effective optical path curve and the intensity of the absorption peak in the cross-section of the gas to be measured, and using spectral fitting bands of different ranges sequentially from low to high, the measurement range can be improved without changing the hardware of the original broadband cavity enhanced absorption spectroscopy system. Especially in field experiments encountering periods of high pollution or pollution sources, the concentration of pollutants can still be accurately measured.

[0148] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0149] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0150] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the broadband cavity enhanced absorption spectroscopy system detection optimization methods described in the above embodiments.

[0151] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0152] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. 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. 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 (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0153] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0154] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0155] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A detection optimization method for a broadband cavity-enhanced absorption spectroscopy system, characterized in that, Includes the following steps: S1. Calibrate the broadband cavity enhanced absorption spectroscopy detection system and obtain the effective optical path curve and the absorption cross section of the gas to be measured. S2. Based on the effective optical path curve and the intensity of the absorption peak in the cross section of the gas to be measured, construct multiple spectral fitting band parameters; S3. Conduct actual observations to obtain measured atmospheric spectra, and use spectral fitting parameters to perform spectral fitting on the measured spectra; S4. Based on the measurement range, determine whether the spectral fitting result meets the requirements. If not, change the spectral fitting parameters and repeat step S3 until the spectral system measurement optimization is completed. The spectral fitting bands in step S2 include: Low-range spectral fitting parameters: The fitting band mainly covers the region with long effective optical path and large absorption peak; High-range spectral fitting parameters: The fitting band mainly covers the region with short effective optical path and small absorption peak; Specific methods for constructing parameters for multiple spectral fitting bands include: S210. Determine the fitted band: Select a band with obvious absorption characteristics: Based on the absorption spectrum of the gas to be measured, select a band with obvious absorption peaks that are not disturbed by other gases. Divide the spectrum into multiple sub-bands: Divide the entire absorption spectrum into multiple sub-bands, each of which contains one or more absorption peaks; Avoid interference bands: exclude bands that may cause interference from other gases and environmental factors; S220. Construct a spectral fitting model; According to Beer-Lambert law, the absorption coefficient varies within each sub-band. With respect to gas concentration C and absorption cross section and effective optical path The relationship is: For multi-component gases, the model is extended as follows: Where m is the number of gas components; It involves filling the optical cavity with high-purity nitrogen gas to measure the reference spectrum when there is no gas to be measured. It is the measurement spectrum of the gas to be measured filled into the optical cavity; By jointly fitting the absorption coefficient data of multiple sub-bands, a global optimization model is constructed: Where n is the number of sub-bands, It is the absorption coefficient of the i-th sub-band; S230, Optimize the fitting parameters; Based on the calibration results, the absorption cross section for each sub-band is... and effective optical path Set initial values; The fitting parameters are optimized using a genetic algorithm to minimize the residuals between the model and the experimental data. Regularization methods are used to improve the model's applicability and prevent overfitting. Specific fitting parameter optimization methods include: S231. Define the optimization problem; Define the objective function: Where θ is the parameter to be optimized. For the measured absorbance coefficient, Predict the absorption coefficient for the model; Adding a regularization term, we get Where θj is the j-th parameter in the parameter vector to be optimized; Optimize the gas concentration C using the objective function, background term. , the absorption cross section correction factor K; A set of initial parameter values ​​is randomly generated to form an initial parameter group, and each individual represents a possible combination of parameters; for each individual, the objective function is calculated. The fitness is used as the basis for selection; several individuals are randomly selected and the individual with the best fitness is retained. Each parameter is swapped with a certain probability, and crossover is performed on the selected individuals to generate new individuals; new values ​​are randomly generated within the parameter range to increase the diversity of the parameter group; selection, crossover and mutation operations are repeated until the termination condition is met. The crossover probability is 0.7-0.9, the mutation probability is 0.01-0.1, and the parameter cluster size is 50-200. S240. Verify the fitting results and determine the fitting parameters; The fitted absorption cross section and effective optical path For practical measurement, gas concentration is retrieved through multi-band joint inversion. : 。 2. The method for optimizing the detection of a broadband cavity enhanced absorption spectroscopy system according to claim 1, characterized in that, The methods for obtaining the effective optical path curve of the spectral detection system and the absorption cross section of the gas to be measured in step S1 include: The optical cavity was filled with high-purity helium and nitrogen gas respectively, and the reflectivity of the cavity mirror was calculated based on the difference in Rayleigh scattering between helium and nitrogen gas. The formula is as follows: in, and These are the spectral intensities after the cavity is filled with helium and nitrogen, respectively. and These are the Rayleigh scattering coefficients for helium and nitrogen, respectively. d Effective cavity length; Calculate the effective optical path using the cavity mirror reflectivity and cavity length. The formula is as follows: Gas absorption cross section It is obtained by convolving internationally recognized standard data with instrument functions.

3. The method for optimizing the detection of a broadband cavity enhanced absorption spectroscopy system according to claim 1, characterized in that, The method for optimizing the spectral system measurement in step S4 includes: The relative error is obtained by comparing the concentration C0 of the gas to be measured with the gas concentration C. When the relative error is less than 0.5%, the fitting result meets the requirements and the optimization ends; otherwise, the spectral fitting parameters are changed and step S3 is repeated.

4. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the processor performs the method as described in any one of claims 1 to 3.

5. A computer device, comprising a memory and a processor, characterized in that, The memory stores a computer program that, when executed by the processor, causes the processor to perform the method as described in any one of claims 1 to 3.

Images