A wide-range concentration detection system and method for a gas analyzer

Abstract

The present invention relates to a wide-range concentration detection system and method for a gas analyzer. The system includes a light source, a collimator, a measuring cell, a spectrometer, a temperature controller, a signal converter, a data acquisition module, and a data analysis module. The method collects a reference zero-point spectrum and concentration sequence spectra of n gases as reference original spectrum sequences, performs differential, normalization, and polynomial fitting on the reference zero-point spectrum and obtains n fitting coefficient matrices for the n gases. The matrix is ​​then combined with the real-time differential spectrum of the gas to be measured to form a nonlinear equation system. The least squares method is used to solve the nonlinear equation system to obtain the initial normalized concentration solution of each gas. The initial normalized concentration solution is further optimized and denormalized to obtain the final concentration value of each gas. The present invention solves the problem of gas concentration detection when the absorption cross section and concentration of the gas to be measured have a nonlinear relationship, and has high detection efficiency and detection accuracy, and a wide concentration detection range.

Description

Technical Field

[0001] The present invention relates to a gas analyzer, in particular to a wide-range concentration detection system and method for a gas analyzer, belonging to the technical field of environmental detection. Background Art

[0002] The most commonly used technology for gas analyzers to detect gas concentration is differential absorption spectroscopy. Its principle is to use the absorption cross-section of the gas to be measured to identify the gas, and to invert the gas concentration based on the absorption intensity. By pre-storing the differential absorption spectrum of the gas to be measured and the fitting coefficients between each concentration in the analyzer, during online real-time detection, the fitting coefficients stored offline and the differential absorption spectrum of the measured gas are used to establish a linear equation group, and then the least squares method is used to invert the concentration of the gas to be measured during real-time measurement.

[0003] This detection method is based on the linear relationship between the differential absorption spectrum and concentration of the gas to be detected. However, with the increasing demand in the gas analysis industry, the concentration detection range of gas analyzers has gradually widened. In actual applications, it is found that when the concentration of the gas to be detected is high, a nonlinear change will occur between the differential absorption spectrum and the concentration of the gas to be detected. Therefore, the previous method of inverse calculation of concentration cannot accurately calculate the concentration value of the gas to be detected, which has a great impact on the calibration of gas concentration monitoring equipment.

[0004] To address the issue of gas concentration detection when the relationship between the differential absorption spectrum and concentration of the gas being detected is nonlinear, existing gas analyzers often use iterative methods to solve nonlinear equations for online concentration inversion. However, this iterative approach often encounters issues such as initial value estimation, parameter tuning, and difficulty in estimating the total number of calculations, resulting in increased computational costs and a poor user experience. Summary of the Invention

[0005] To overcome the shortcomings of the existing technology, the present invention provides a wide-range concentration detection system and method for a gas analyzer. When the differential absorption spectrum of the gas to be measured and its concentration present a nonlinear relationship, the least squares method can be used to simplify the solution steps of the nonlinear equation group, thereby improving detection efficiency. The concentration detection range is wide and the detection accuracy is high, ensuring user experience.

[0006] A method for detecting wide-range concentration of a gas analyzer comprises the following steps:

[0007] S1, collect reference zero-point spectrum and concentration sequence spectra of n gases as reference original spectrum sequence;

[0008] S2, respectively calculating the differential absorption spectrum between the reference original spectrum and the reference zero-point spectrum of each concentration of each gas to obtain n differential reference spectrum sequences;

[0009] S3, respectively selecting the values ​​of all differential reference spectra at each pixel point within the absorption band of the mixed gas to be measured in each differential reference spectrum sequence to form a new differential reference spectrum, thereby obtaining n new differential reference spectrum sequences;

[0010] S4, normalizing the concentration sequences of n gases to obtain n normalized concentration sequences, and fitting them with n new differential reference spectrum sequences to obtain n fitting coefficient matrices;

[0011] S5, introducing the mixed gas to be tested into the gas analyzer, collecting the spectrum of the mixed gas to be tested online in real time, and calculating and obtaining the real-time differential spectrum of each spectrum of the mixed gas to be tested and the reference zero-point spectrum within the absorption band;

[0012] S6, n fitting coefficient matrices are arranged in columns to obtain A, the real-time differential spectrum is b, and the nonlinear equation system Ax = b is established, and the initial normalized concentration vector x = (A T A) -1 b;

[0013] S7, optimizing the initial normalized concentration vector to obtain the optimal normalized concentration solution of each gas component of the mixed gas to be measured, and performing denormalization processing on the optimal normalized concentration solution to obtain the final concentration solution of each gas component of the mixed gas to be measured.

[0014] Specifically, in step S4, the calculation steps of normalization and fitting are as follows:

[0015] S41, taking the maximum value of each gas concentration among n gases, normalizing each gas concentration sequence to obtain n normalized concentration sequences;

[0016] S42, taking the normalized concentration sequence of each gas in the n gases as X, taking each differential reference spectrum in the new differential reference spectrum sequence of each gas in the n gases as Y, fitting X and Y with a p-1 order polynomial to obtain l p-dimensional fitting coefficient vectors;

[0017] Where p is the number of concentration types in each gas concentration sequence; l is the number of pixels in the absorption band of the mixed gas to be measured;

[0018] S43, arranging the lp-dimensional fitting coefficient vectors corresponding to each gas into an l×p-dimensional fitting coefficient matrix by row, obtaining n fitting coefficient matrices and storing them in the gas analyzer.

[0019] The initial normalized concentration vector in S7 includes p-dimensional concentration vectors of n gas components to be measured, and each p-dimensional concentration vector includes one concentration solution variable C and p-1 other concentration variables based on C.

[0020] Specifically, in S7, the initial normalized concentration vector is optimized to obtain the optimal normalized concentration solution, and the optimal normalized concentration solution is denormalized. The steps are as follows:

[0021] S71, in the initial normalized concentration vector, respectively optimizing the concentration solution variables of each gas component to be measured to obtain n sequences of better concentration (deleted solutions) vectors;

[0022] S72, performing structured processing on the initial normalized concentration vector to obtain an initial structured concentration vector, and performing structured processing on each of the better concentration vector sequences of the gas components to be measured to obtain a structured concentration vector sequence of each gas component of the mixed gas to be measured;

[0023] S73, sequentially substituting each vector in the structured concentration vector sequence of each gas component in the mixed gas to be measured and the initial structured concentration vector into the nonlinear equation system (Ax=b) to calculate the residual, thereby obtaining a residual sequence corresponding to each gas component in the mixed gas to be measured;

[0024] S74, determining the minimum residual in the residual sequence of each gas component in the mixed gas to be measured, selecting the concentration solution variable in the structured concentration vector corresponding to the minimum residual as the optimal normalized concentration solution, and performing denormalization processing on the optimal normalized concentration solution to obtain the final concentration solution of the gas component.

[0025] Furthermore, the method for optimizing the concentration solution variables in S71 to obtain a better concentration vector sequence is as follows:

[0026] For a certain gas component in the mixed gas to be measured, in the initial normalized concentration vector, the concentration solution variable of the gas component is optimized in proportion a%. Value 2m times, that is, The values ​​are processed as follows: At the same time, other concentration variables in the initial normalized concentration vector remain unchanged, and a better concentration vector sequence of the gas component is obtained; according to the above method, n gas components of the mixed gas to be measured are processed to obtain a total of n better concentration vector sequences.

[0027] Furthermore, the method for structured processing of the concentration vector in S72 is as follows:

[0028] For each p-dimensional concentration vector of the gas component to be measured in the concentration vector to be structured, keep the concentration solution variable of the gas component to be measured Keep the other concentration variables constant and satisfy the theoretical structure

[0029] A wide-range concentration detection system for a gas analyzer includes a light source, a collimator, a measuring cell, a spectrometer, a temperature controller, a signal converter, a data acquisition module, and a data analysis module. Broad-spectrum light emitted by the light source is collimated and then emitted into the measuring cell. The broad-spectrum light passes through the gas in the measuring cell and is received by the spectrometer. The temperature controller is used to control the temperature of the light source. The spectrometer converts the collected optical signal into a digital signal and sends it to the data acquisition module. The data acquisition module is connected to the data analysis module. The data analysis module includes the method steps described in the wide-range concentration detection method for a gas analyzer and is used to calculate the concentration of the gas to be measured.

[0030] The system and method adopted by the present invention fully utilize the special structure of the problem to establish an ingenious nonlinear equation group when the differential absorption spectrum and concentration of the gas to be measured are in a nonlinear relationship, so that the least squares method can be used to quickly obtain the optimal solution of the nonlinear equation group, avoiding the problems of initial value guessing, parameter tuning, and difficulty in estimating the total number of calculations encountered in the iterative method for solving the nonlinear equation group, significantly saving computing costs and improving user experience; at the same time, after obtaining the optimal solution, the present invention can obtain a high-precision solution to the nonlinear equation group through optimization and structured processing, effectively controlling the error range of the detection value; in addition, the present invention can simultaneously detect the concentration values ​​of multiple gas components in the mixed gas, and greatly broaden the scope of gas concentration detection, so that the gas concentration monitoring equipment has a wider range of use and is applicable to more occasions. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Flow chart of the method of the present invention;

[0032] Figure 2 is the SO2 fitting coefficient matrix in the specific embodiment;

[0033] Figure 3 is the NO fitting coefficient matrix in a specific embodiment;

[0034] Figure 4 It is the real-time differential spectrum after band selection;

[0035] Figure 5 Schematic diagram of the system principle of the present invention. DETAILED DESCRIPTION

[0036] To illustrate the present invention more clearly, the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0037] like Figure 1 As shown, a wide-range concentration detection method for a gas analyzer includes the following steps:

[0038] S1, collect reference zero-point spectrum and concentration sequence spectra of n gases as reference original spectrum sequence;

[0039] In this embodiment, a mixed gas containing two gas components, SO2 and NO, is introduced into the measuring cell in sequence. The background gas, the concentration of which is 8000, 6000, 4000, 2000, 1000 (unit: mg / m 3 ) of SO2 and concentrations of 100, 70, 50, 25, 10 (unit: mg / m 3 ) of NO, respectively, collect the reference zero point spectrum, the reference original spectrum sequence of SO2, and the reference original spectrum sequence of NO and store them in advance in the gas analyzer, wherein the reference original spectrum sequence of SO2 and the reference original spectrum sequence of NO both contain 5 reference original spectra. The concentration sequence of the gas is a sequence of gas concentration values, such as concentrations of 8000, 6000, 4000, 2000, 1000 (unit: mg / m 3 )’s SO2 concentration sequence is {8000, 6000, 4000, 2000, 1000}, and the concentrations are 100, 70, 50, 25, 10 (unit: mg / m 3 )’s NO concentration sequence is {100, 70, 50, 25, 10}.

[0040] S2, respectively calculating the differential absorption spectrum between the reference original spectrum and the reference zero-point spectrum of each concentration of each gas to obtain n differential reference spectrum sequences;

[0041] In this embodiment, a second-order difference is used with a difference interval of 6 to sequentially calculate the differential absorption spectra of the reference original spectra of the five concentrations in the SO2 reference original spectrum sequence and the reference zero-point spectrum to obtain a SO2 differential reference spectrum sequence. The differential absorption spectra of the reference original spectra of the five concentrations in the NO reference original spectrum sequence and the reference zero-point spectrum to obtain a NO differential reference spectrum sequence.

[0042] S3, respectively selecting the values ​​of all differential reference spectra at each pixel point within the absorption band of the mixed gas to be measured in each differential reference spectrum sequence to form a new differential reference spectrum, thereby obtaining n new differential reference spectrum sequences;

[0043] In this embodiment, the absorption band of the SO2 and NO mixed gas corresponds to a pixel interval of 234-301, which is 68 pixels long. The values ​​of the 1st, 2nd, ..., 68th pixels in the pixel interval 234-301 of the five differential reference spectra in the SO2 differential reference spectrum sequence are sequentially taken to form the 1st, 2nd, ..., 68th new differential reference spectra to obtain a new SO2 differential reference spectrum sequence. The values ​​of the 1st, 2nd, ..., 68th pixels in the pixel interval 234-301 of the five differential reference spectra in the NO differential reference spectrum sequence are sequentially taken to form the 1st, 2nd, ..., 68th new differential reference spectra to obtain a new NO differential reference spectrum sequence.

[0044] S4, normalize the concentration sequences of n gases to obtain n normalized concentration sequences, and fit them with n new differential reference spectrum sequences to obtain n fitting coefficient matrices. The calculation steps are as follows:

[0045] S41, taking the maximum value of each gas concentration among n gases, normalizing each gas concentration sequence to obtain n normalized concentration sequences;

[0046] In this embodiment, the maximum concentration of SO2 in the mixed gas to be measured is 8000, and the SO2 concentration sequence {8000, 6000, 4000, 2000, 1000} (unit: mg / m 3 ) is normalized to obtain the normalized SO2 concentration sequence {1, 0.75, 0.5, 0.25, 0.125}. The maximum concentration of NO in the mixed gas to be tested is 100. The NO concentration sequence {100, 70, 50, 25, 10} (unit: mg / m 3 ) was normalized to obtain the NO normalized concentration sequence {1, 0.7, 0.5, 0.25, 0.1}.

[0047] S42, taking the normalized concentration sequence of each gas in the n gases as X, taking each differential reference spectrum in the new differential reference spectrum sequence of each gas in the n gases as Y, fitting X and Y with a p-1 order polynomial to obtain l p-dimensional fitting coefficient vectors;

[0048] Where p is the number of concentration types in each gas concentration sequence; l is the number of pixels in the absorption band of the mixed gas to be measured;

[0049] In this embodiment, the SO2 normalized concentration sequence {1, 0.75, 0.5, 0.25, 0.125} is taken as X, and each differential reference spectrum in the new differential reference spectrum sequence of SO2 is taken as Y in turn. The coefficients of the independent variables 0, 1, ..., and the fourth power are obtained by fitting a 4th-order polynomial and arranged in order into a 5-dimensional fitting coefficient vector, resulting in 68 5-dimensional fitting coefficient vectors. Take the NO normalized concentration sequence {1, 0.7, 0.5, 0.25, 0.1} as X, take each differential reference spectrum in the new NO differential reference spectrum sequence as Y, fit the 4th-order polynomial to obtain the coefficients of the independent variables 0, 1, ..., 4th power and arrange them in order into a 5-dimensional fitting coefficient vector, and obtain 68 5-dimensional fitting coefficient vectors

[0050] S43, arranging the lp-dimensional fitting coefficient vectors corresponding to each gas into an l×p-dimensional fitting coefficient matrix by row, obtaining n fitting coefficient matrices and storing them in the gas analyzer.

[0051] In this embodiment, Figure 2As shown, there are 68 5-dimensional fitting coefficient vectors Arrange the rows to form a fitting coefficient matrix like Figure 3 As shown, there are 68 5-dimensional fitting coefficient vectors Arrange the rows to form a fitting coefficient matrix

[0052] S5, introducing the mixed gas to be tested into the gas analyzer, collecting the spectrum of the mixed gas to be tested online in real time, and calculating and obtaining the real-time differential spectrum of each spectrum of the mixed gas to be tested and the reference zero-point spectrum within the absorption band;

[0053] To prove the accuracy and effectiveness of this calculation method, a mixed gas consisting of a SO2 concentration of 5000 and a NO concentration of 30 (unit: mg / m 3 ) is tested, and the spectrum of the mixed gas to be tested is collected online in real time. The second-order difference is used and the difference interval is 6. The real-time difference spectrum of the mixed gas spectrum to be tested and the reference zero point spectrum in the pixel interval 234-301 is calculated, and the real-time difference spectrum after the selected band is obtained as follows Figure 4 As shown;

[0054] S6, n fitting coefficient matrices are arranged in columns to obtain A, the real-time differential spectrum is b, and the nonlinear equation system Ax = b is established, and the initial normalized concentration vector x = (A T A) -1 b;

[0055] In this embodiment, b l×1 For the real-time differential spectrum after band selection, the least squares method is used to solve Ax=b to obtain the initial normalized concentration vector as follows: [2.0286, 0.6040, 0.3553, 0.2135, 0.0637, -0.7237, 0.2552, 0.4890, 0.5718, 0.6121] T From step S42, it can be seen that the initial normalized concentration vector contains the 5-dimensional concentration vectors of the SO2 and NO2 gas components to be measured [2.0286, 0.6040, 0.3553, 0.2135, 0.0637] T 、[-0.7237,0.2552,0.4890,0.5718,0.6121] T , where the 5-dimensional concentration vector of each gas component to be measured should satisfy the structure The elements of represent the 0, 1, ..., p-1 power values ​​of the normalized concentration of the gas component to be measured, that is, is the concentration solution variable, and the other values ​​are The concentration variable is based on the equation; however, in engineering practice, due to interference factors such as polynomial fitting errors, the normalized concentration vector of the gas component to be measured usually does not satisfy the theoretical structure. Therefore, the initial normalized concentration vector is not optimal and needs to be further optimized and structured to find the final concentration solution. The calculation steps are as follows:

[0056] S7, optimizing the initial normalized concentration vector to obtain the optimal normalized concentration solution of each gas component of the mixed gas to be measured, and performing denormalization processing on the optimal normalized concentration solution to obtain the final concentration solution of each gas component of the mixed gas to be measured.

[0057] S71, in the initial normalized concentration vector, optimizing the concentration solution variables of each gas component to be measured to obtain n better concentration vector sequences;

[0058] Specifically, the method for optimizing the concentration solution variables to obtain a better concentration vector sequence is as follows:

[0059] For a certain gas component in the mixed gas to be measured, in the initial normalized concentration vector, the concentration solution variable of the gas component is optimized in proportion a%. Value 2m times, that is, The values ​​are processed as follows: At the same time, other concentration variables in the initial normalized concentration vector remain unchanged, and a better concentration vector sequence of the gas component is obtained; according to the above method, n gas components of the mixed gas to be measured are processed to obtain a total of n better concentration vector sequences.

[0060] In this embodiment, a=1, m=5, that is, the initial normalized concentration vector [2.0286, 0.6040, 0.3553, 0.2135, 0.0637, -0.7237, 0.2552, 0.4890, 0.5718, 0.6121] T In the above example, the concentration solution variable 0.6040 of SO2 is processed in sequence according to the formulas 0.6040(1-1%), 0.6040(1-2%), ..., 0.6040(1-5%), 0.6040(1+1%), 0.6040(1+2%), ..., 0.6040(1+5%), while all other concentration variables in the initial normalized concentration vector remain unchanged, thus obtaining a better concentration vector sequence of SO2. In the normalized concentration vector, the NO concentration solution variable 0.2552 is processed in sequence according to the formulas 0.2552(1-1%), 0.2552(1-2%), ..., 0.2552(1-5%), 0.2552(1+1%), 0.2552(1+2%), ..., 0.2552(1+5%), while all other concentration variables in the initial normalized concentration vector remain unchanged, thus obtaining a more optimal concentration vector sequence for NO;

[0061] S72, performing structured processing on the initial normalized concentration vector to obtain an initial structured concentration vector, and performing structured processing on each of the better concentration vector sequences of the gas components to be measured to obtain a structured concentration vector sequence of each gas component of the mixed gas to be measured;

[0062] The method of structured processing is as follows: for each p-dimensional concentration vector of the gas component to be measured in the concentration vector to be structured, the concentration solution variable of the gas component to be measured is maintained. unchanged, and make Other concentration variables based on the theoretical structure

[0063] In this embodiment, each concentration vector in the SO2 optimal concentration vector sequence, each concentration vector in the NOx optimal concentration vector sequence, and the initial normalized concentration vector are structured to obtain a SO2 structured concentration vector sequence, a NOx structured concentration vector sequence, and an initial structured concentration vector. The initial structured concentration vector is as follows: [1, 0.6040, 0.3648, 0.2203, 0.1331, 1, 0.2552, 0.0651, 0.0166, 0.0042] T ;

[0064] S73, sequentially substituting each vector in the structured concentration vector sequence of each gas component in the mixed gas to be measured and the initial structured concentration vector into the nonlinear equation group Ax=b to calculate the residual, thereby obtaining a residual sequence corresponding to each gas component in the mixed gas to be measured;

[0065] In this embodiment, each vector in the structured concentration vector sequence of SO2 and the initial structured concentration vector are sequentially substituted into the nonlinear equation group Ax=b to calculate the residual, thereby obtaining the residual sequence of SO2; each vector in the structured concentration vector sequence of NO and the initial structured concentration vector are sequentially substituted into the nonlinear equation group Ax=b to calculate the residual, thereby obtaining the residual sequence of NO;

[0066] S74, determining the minimum residual in the residual sequence of each gas component in the mixed gas to be measured, selecting the concentration solution variable in the structured concentration vector corresponding to the minimum residual as the optimal normalized concentration solution, and performing denormalization processing on the optimal normalized concentration solution to obtain the final concentration solution of the gas component

[0067] In this example, the minimum residual in the residual sequence of SO2 is 4.406e-4. The concentration solution variable of SO2 in the structured concentration vector corresponding to the residual 4.406e-4 is 0.6040. Using 8000 mg / m 3 The final concentration of SO2 obtained by denormalization is 4832 mg / m 3 , its standard concentration is 5000mg / m3 , the measurement error is 1.36%; the minimum residual in the residual sequence of NO is 4.349e-4, and the concentration solution variable of NO in the structured concentration vector corresponding to the residual 4.349e-4 is 0.2654, using 100mg / m 3 The final concentration of NO obtained by reverse normalization is 26.54 mg / m 3 , its standard concentration is 30mg / m 3 , the measurement error is 11.53%.

[0068] like Figure 5 As shown, a wide-range concentration detection system for a gas analyzer includes a light source, a collimator, a measuring cell, a spectrometer, a temperature controller, a signal converter, a data acquisition module, and a data analysis module. The wide-spectrum light emitted by the light source is collimated and then emitted into the measuring cell. After passing through the gas in the measuring cell, it is received by the spectrometer. The temperature controller is used to control the temperature of the light source. The spectrometer converts the collected optical signal into a digital signal and sends it to the data acquisition module. The data acquisition module is connected to the data analysis module. The data analysis module includes the method steps described in the wide-range concentration detection method for the gas analyzer and is used to calculate the concentration of the gas to be measured.

[0069] A non-transitory computer-readable storage medium includes instructions for executing the wide-range concentration detection method for a gas analyzer described in any one of the above embodiments.

[0070] An electronic device includes a non-transitory computer-readable storage medium; and one or more processors capable of executing the instructions of the non-transitory computer-readable storage medium.

[0071] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0072] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0075] Specific embodiments are used in the present invention to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A method for detecting wide-range concentration of a gas analyzer, characterized in that: The following steps are involved: S1, collect reference zero-point spectrum and concentration sequence spectra of n gases as reference original spectrum sequence; S2, respectively calculating the differential absorption spectrum between the reference original spectrum and the reference zero-point spectrum of each concentration of each gas to obtain n differential reference spectrum sequences; S3, respectively selecting the values ​​of all differential reference spectra at each pixel point within the absorption band of the mixed gas to be measured in each differential reference spectrum sequence to form a new differential reference spectrum, thereby obtaining n new differential reference spectrum sequences; S4, normalizing the concentration sequences of n gases to obtain n normalized concentration sequences, and fitting them with n new differential reference spectrum sequences to obtain n fitting coefficient matrices; S5, introducing the mixed gas to be tested into the gas analyzer, collecting the spectrum of the mixed gas to be tested online in real time, and calculating and obtaining the real-time differential spectrum of each spectrum of the mixed gas to be tested and the reference zero-point spectrum within the absorption band; S6, n fitting coefficient matrices are arranged in columns to obtain A, the real-time differential spectrum is b, and the nonlinear equation system Ax = b is established, and the initial normalized concentration vector x = (A T A) -1 b; S7, optimizing the initial normalized concentration vector to obtain the optimal normalized concentration solution of each gas component of the mixed gas to be measured, and performing denormalization processing on the optimal normalized concentration solution to obtain the final concentration solution of each gas component of the mixed gas to be measured.

2. The wide-range concentration detection method of a gas analyzer according to claim 1, characterized in that: In step S4, the calculation steps of normalization and fitting are as follows: S41, taking the maximum value of each gas concentration among n gases, normalizing each gas concentration sequence to obtain n normalized concentration sequences; S42, taking the normalized concentration sequence of each gas in the n gases as X, taking each differential reference spectrum in the new differential reference spectrum sequence of each gas in the n gases as Y, fitting X and Y with a p-1 order polynomial to obtain l p-dimensional fitting coefficient vectors; Where p is the number of concentration types in each gas concentration sequence; l is the number of pixels in the absorption band of the mixed gas to be measured; S43, arranging the lp-dimensional fitting coefficient vectors corresponding to each gas into an l×p-dimensional fitting coefficient matrix by row, obtaining n fitting coefficient matrices and storing them in the gas analyzer.

3. The wide-range concentration detection method of a gas analyzer according to claim 2, characterized in that: The initial normalized concentration vector in S7 includes p-dimensional concentration vectors of n gas components to be measured, and each p-dimensional concentration vector includes 1 concentration solution variable and p-1 Other concentration variables based on .

4. The wide-range concentration detection method of a gas analyzer according to claim 3, characterized in that: In S7, the initial normalized concentration vector is optimized to obtain the optimal normalized concentration solution, and the optimal normalized concentration solution is denormalized. The steps are as follows: S71, in the initial normalized concentration vector, optimizing the concentration solution variables of each gas component to be measured to obtain n better concentration vector sequences; S72, performing structured processing on the initial normalized concentration vector to obtain an initial structured concentration vector, and performing structured processing on each of the better concentration vector sequences of the gas components to be measured to obtain a structured concentration vector sequence of each gas component of the mixed gas to be measured; S73, sequentially substituting each vector in the structured concentration vector sequence of each gas component in the mixed gas to be measured and the initial structured concentration vector into the nonlinear equation system to calculate the residual, thereby obtaining a residual sequence corresponding to each gas component in the mixed gas to be measured; S74, determining the minimum residual in the residual sequence of each gas component in the mixed gas to be measured, selecting the concentration solution variable in the structured concentration vector corresponding to the minimum residual as the optimal normalized concentration solution, and performing denormalization processing on the optimal normalized concentration solution to obtain the final concentration solution of the gas component.

5. The wide-range concentration detection method of a gas analyzer according to claim 4, characterized in that: The method for optimizing concentration solution variables to obtain a better concentration vector sequence is as follows: For a certain gas component in the mixed gas to be measured, in the initial normalized concentration vector, the concentration solution variable of the gas component is optimized in proportion a%. Value 2m times, that is, The values ​​are processed as follows: At the same time, other concentration variables in the initial normalized concentration vector remain unchanged, and a better concentration vector sequence of the gas component is obtained; according to the above method, n gas components of the mixed gas to be measured are processed to obtain a total of n better concentration solution vector sequences.

6. The wide-range concentration detection method of a gas analyzer according to claim 5, characterized in that: The structured processing method in S72 is as follows: For each p-dimensional concentration vector of the gas component to be measured in the concentration vector to be structured, keep the concentration solution variable of the gas component to be measured Keep the other concentration variables constant and satisfy the theoretical structure 7. A wide-range concentration detection system for a gas analyzer, using the wide-range concentration detection method for a gas analyzer according to any one of claims 1 to 6 for detection, characterized in that: The system includes a light source, a collimator, a measuring cell, a spectrometer, a temperature controller, a signal converter, a data acquisition module, and a data analysis module. The broad-spectrum light emitted by the light source is collimated and then emitted into the measuring cell. The broad-spectrum light passes through the gas in the measuring cell and is received by the spectrometer. The temperature controller is used to control the temperature of the light source. The spectrometer converts the collected optical signal into a digital signal and sends it to the data acquisition module. The data acquisition module is connected to the data analysis module. The data analysis module includes the method steps described in the wide-range concentration detection method of the gas analyzer and is used to calculate the concentration of the gas to be measured.

Images