A system and method for quantitative analysis of overlapping gas components

Abstract

This invention discloses a quantitative analysis system and method for overlapping gas components, belonging to the field of gas analysis and vacuum diagnostics technology. The system comprises: a vacuum chamber to be tested, a first valve connection, and a differential device connected in sequence; the differential device is connected to a mass spectrometry detection unit, a spectral detection unit, and a vacuum pump unit; the mass spectrometry detection unit is used to acquire ion current signals of the analyte gas in multiple mass-to-charge ratio channels; the spectral detection unit is used to acquire characteristic spectral signals of the analyte gas in multiple wavelength channels; a data processing and calculation unit is communicatively connected to the mass spectrometry detection unit and the spectral detection unit, respectively, to obtain and output the partial pressure or concentration of each candidate gas component. Overlapping gas components are those gas components that are difficult to accurately distinguish in mass spectrometry detection due to overlapping mass-to-charge ratio channel responses or overlapping or intersecting characteristic spectral lines in spectral detection. This invention improves the identification capability and quantitative accuracy of overlapping components.

Description

Technical Field

[0001] This invention belongs to the field of gas analysis and vacuum diagnostics technology, specifically relating to a quantitative analysis system and method for overlapping gas components. Background Technology

[0002] In vacuum systems, low-pressure discharge devices, material venting testing devices, and fuel cycle and wall treatment processes of fusion devices, it is often necessary to detect the components of mixed gases in real time or near real time in order to achieve gas component identification, residual gas monitoring, leak diagnosis, operating condition assessment, and process optimization.

[0003] Residual gas analyzers (RGAs) are commonly used vacuum gas analysis instruments. Their core analytical component is typically a mass spectrometer (MS), with quadrupole mass spectrometers (QMS) commonly used in vacuum systems. RGAs ionize the gas using an internal ion source, and then use a quadrupole electric field to spatially separate ions with different mass-to-charge ratios (m / z). The gas composition is identified and analyzed by detecting the ion current signals at different mass-to-charge ratio positions. Because of their high sensitivity, fast response, and suitability for online monitoring, RGAs based on quadrupole mass spectrometry are widely used in vacuum systems and fusion-related experimental devices.

[0004] However, relying solely on RGA for quantitative analysis under multi-component gas mixture conditions often has significant limitations. Limited by the unit mass resolution of a quadrupole mass spectrometer, different gas components with the same or similar mass-to-charge ratios tend to produce overlapping responses in the same mass spectral channel. For example, deuterium (D2), hydrogen tritide (HT), and helium (He) may all produce responses at m / z=4, while nitrogen (N2) and carbon monoxide (CO) may both produce responses at m / z=28. Furthermore, some gases can form fragment peaks at multiple mass-to-charge ratio positions, leading to superposition of responses between different components and increasing analytical ambiguity. Therefore, in the presence of overlapping or fragment peak interference, relying solely on the signals from one or a few mass-to-charge ratio channels measured by RGA is often insufficient to accurately distinguish the actual content of each component.

[0005] To improve the accuracy of gas analysis, existing technologies have incorporated optical emission spectroscopy, absorption spectroscopy, and other optical detection methods. Spectroscopic detection methods can reflect the characteristic spectral line information of specific gas components and provide additional evidence for component identification in certain scenarios. However, when using spectroscopic methods alone, they can be affected by factors such as background radiation, spectral line overlap, signal drift, and calibration instability, making it difficult to obtain stable and reliable absolute quantitative results in complex multi-component environments. Among existing spectroscopic analysis techniques, spectral interference is a common problem, which is currently generally avoided by selecting non-intersecting spectral lines or using spectral deconvolution methods.

[0006] Therefore, how to combine the advantages of mass spectrometry and spectral detection to form a unified quantitative technical solution for resolving overlapping components, especially in vacuum, low-pressure and fusion-related environments, to achieve accurate identification and quantitative analysis of overlapping components, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0008] An overlapping gas component fusion quantitative analysis system includes: a vacuum chamber to be tested, a first vacuum pipeline, a first valve, a vacuum gauge, a differential device, a mass spectrometry detection unit, a spectral detection unit, a data processing and calculation unit, a user interaction and display interface, and a vacuum pumping unit;

[0009] The vacuum chamber to be tested is connected to the first valve through the first vacuum pipe. The output end of the first valve is connected to the differential device. The gas to be tested enters the differential device through the first vacuum pipe and the first valve. The differential device is connected to the mass spectrometry detection unit, the spectral detection unit and the vacuum pumping unit respectively.

[0010] A vacuum gauge is installed on the differential device to monitor the internal pressure of the differential device; a mass spectrometry detection unit is used to collect ion current signals of the gas to be tested in multiple mass-to-charge ratio channels; and a spectral detection unit is used to collect characteristic spectral signals of the gas to be tested in multiple wavelength channels.

[0011] The data processing and calculation unit is connected to the mass spectrometry detection unit and the spectral detection unit respectively, and is used to perform synchronous trigger control, data acquisition, database matching, joint calibration, weighted modeling, constraint inversion solution and result output; the user interaction and display interface is connected to the data processing and calculation unit, and is used to display the partial pressure or concentration results of each candidate gas component and the corresponding trend information.

[0012] A method for quantitative analysis of overlapping gas components, used in the aforementioned quantitative analysis system for quantitative analysis of overlapping gas components, comprising:

[0013] S1, simultaneously acquires mass spectrometry and spectral data of the gas to be tested, and performs data preprocessing and synchronization;

[0014] S2, based on the preprocessed and synchronized mass spectrometry and spectral data, and combined with the fragment abundance information, characteristic spectral line information and relative response information of candidate gas components at multiple mass-to-charge ratio positions pre-stored in the species characteristic database, the candidate gas components are matched and screened to obtain the set of candidate gas components to be solved.

[0015] S3. Based on the candidate gas component set obtained in S2, and combined with the fragment abundance information, characteristic spectral line information and relative response information of the candidate gas components at multiple mass-to-charge ratio positions pre-stored in the species characteristic database, a linear forward model is constructed.

[0016] S4. Based on the calibration results of the standard gas or standard gas mixture, and in conjunction with the calibration parameter database, perform sensitivity calibration and matrix correction on the sensitivity parameters in the linear forward model to obtain the joint response matrix. Mass spectrometry sensitivity parameters and spectral sensitivity parameters ;

[0017] S5, based on mass spectrometry sensitivity parameters and spectral sensitivity parameters Constructing joint observation vectors With joint response matrix ;

[0018] S6 normalizes and weights the mass spectrometry and spectral data, and introduces a weight matrix. ;

[0019] S7, based on joint observation vector Joint response matrix and weight matrix Construct weighted overdetermined equations;

[0020] S8, Solve the weighted overdetermined equations under physical constraints to obtain the partial pressure or concentration of each candidate gas component;

[0021] S9 outputs the quantitative analysis results of the partial pressure or concentration of each candidate gas component, including the partial pressure or concentration of the gas component, and outputs the model fitting residuals, trend curves and diagnostic information.

[0022] The present invention has the following beneficial effects:

[0023] This invention achieves the fusion and utilization of multi-source detection information under a unified framework by simultaneously acquiring mass spectrometry and spectral signals and constructing a joint forward model in conjunction with a gas characteristic database. It solves the problems of insufficient resolution of single mass spectrometry detection under conditions of overlapping mass-to-charge ratio and fragment peak interference, and insufficient quantitative stability of single spectral detection under conditions of spectral line intersection and background drift. It improves the identification ability and quantitative accuracy of overlapping components in complex mixed gases.

[0024] This invention solves the problems of inconsistent response scales and poor comparability of quantitative results between different detection methods by jointly calibrating the mass spectrometry detection link and the spectral detection link, thereby improving the consistency of system response and calibration accuracy.

[0025] This invention addresses the issue of imbalanced contributions from different modes during joint solution by introducing a weighting matrix to uniformly process the differences in dimensions, orders of magnitude, and noise levels between the two types of signals. This improves the stability and reliability of the inversion solution.

[0026] This invention solves the problems of distorted and unstable solution results under conditions of overlapping peaks, fragment peaks, and spectral line intersections by applying non-negative constraints and combining them with regularization constraints or prior information for joint inversion. This improves the physical rationality and engineering applicability of quantitative analysis of complex gas systems, especially overlapping components. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the structure of the overlapping gas component quantitative analysis system based on the fusion of mass spectrometry and spectral information provided in an embodiment of the present invention, wherein 1-vacuum chamber to be tested, 2-first vacuum pipeline, 3-first valve, 4-vacuum gauge, 5-differential device, 6-mass spectrometry detection unit, 7-spectral detection unit, 8-data processing and calculation unit, 9-user interaction and display interface, 10-second valve, 11-second vacuum pipeline, 12-turbomolecular pump, 13-third vacuum pipeline, 14-third valve, 15-backstage pump;

[0028] Figure 2 A flowchart of a quantitative analysis method for overlapping gas components based on the fusion of mass spectrometry and spectral information provided in an embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0030] To address the shortcomings of existing technologies, such as insufficient resolution of single mass spectrometry detection in scenarios with overlapping mass-to-charge ratios and fragment peak interference, and insufficient quantitative stability of single spectral detection under conditions of spectral line intersection and background drift, this invention provides a quantitative analysis system and method for overlapping gas components based on the fusion of mass spectrometry and spectral information, so as to achieve accurate identification and quantitative analysis of overlapping components in complex gas mixtures.

[0031] The present invention provides a quantitative analysis system for overlapping gas components based on the fusion of mass spectrometry and spectral information (hereinafter referred to as the system), comprising a vacuum chamber to be tested 1, a first vacuum pipe 2, a first valve 3, a vacuum gauge 4, a differential device 5, a mass spectrometry detection unit 6, a spectral detection unit 7, a data processing and calculation unit 8, a user interaction and display interface 9, a second valve 10, a second vacuum pipe 11, a turbomolecular pump 12, a third vacuum pipe 13, a third valve 14, and a backing pump 15.

[0032] The vacuum chamber 1 to be tested is connected to the first valve 3 via the first vacuum pipe 2. The output end of the first valve 3 is connected to the differential device 5. The gas to be tested enters the differential device 5 through the first vacuum pipe 2 and the first valve 3, and then enters the subsequent detection area. The differential device 5 is connected to the mass spectrometry detection unit 6, the spectral detection unit 7, and the vacuum pumping unit. The second valve 10, the second vacuum pipe 11, the turbomolecular pump 12, the third vacuum pipe 13, the third valve 14, and the forepump 15, which are connected in sequence, constitute the vacuum pumping unit. The vacuum gauge 4 is installed on the differential device 5 to monitor the internal pressure of the differential device 5. The mass spectrometry detection unit 6 is used to collect the ion current signals of the gas to be tested in multiple mass-to-charge ratio channels. The spectral detection unit 7 is used to collect the characteristic spectral signals of the gas to be tested in multiple wavelength channels.

[0033] The data processing and calculation unit 8 is communicatively connected to the mass spectrometry detection unit 6 and the spectral detection unit 7, respectively, and is used to perform synchronous trigger control, data acquisition, database matching, joint calibration, weighted modeling, constraint inversion solution, and result output; the user interaction and display interface 9 is connected to the data processing and calculation unit 8 and is used to display the partial pressure or concentration results of each candidate gas component and the corresponding trend information.

[0034] This invention further provides a quantitative analysis method for overlapping gas components based on the fusion of mass spectrometry and spectral information, the method comprising the following steps:

[0035] S1, synchronously acquires mass spectrometry and spectral data of the gas to be tested, and performs data preprocessing and synchronization on the acquired raw signals (i.e., mass spectrometry and spectral data of the gas to be tested) to complete denoising, baseline subtraction, peak extraction and time alignment.

[0036] S2, based on the preprocessed and synchronized mass spectrometry and spectral data, and combined with the fragment abundance information, characteristic spectral line information and relative response information of candidate gas components at multiple mass-to-charge ratio positions pre-stored in the species characteristic database, the candidate gas components are matched and screened to obtain the set of candidate gas components to be solved.

[0037] S3. Based on the candidate gas component set obtained in S2, and combined with the fragment abundance information, characteristic spectral line information and relative response information of the candidate gas components at multiple mass-to-charge ratio positions pre-stored in the species characteristic database, a linear forward model is constructed.

[0038] S4. Based on the calibration results of the standard gas or standard gas mixture, and in conjunction with the calibration parameter database, perform sensitivity calibration and matrix correction on the sensitivity parameters in the linear forward model to obtain the joint response matrix. Mass spectrometry sensitivity parameters and spectral sensitivity parameters Matrix correction refers to the overall calibration of the previously constructed linear forward model to make it closer to the actual system response.

[0039] S5, based on the model parameters (mass spectrometry sensitivity parameters) after sensitivity calibration and matrix correction. and spectral sensitivity parameters ), construct joint observation vector With joint response matrix ;

[0040] S6 addresses the differences between mass spectrometry and spectral data in terms of dimensions, orders of magnitude, and noise levels by normalizing and weighting the mass spectrometry and spectral data, and introduces a weight matrix. ;

[0041] S7, based on joint observation vector Joint response matrix and weight matrix Construct weighted overdetermined equations: ;in, This represents the partial pressure or concentration vector of the candidate gas components. This is the joint noise vector;

[0042] S8, Solve the weighted overdetermined equations under physical constraints to obtain the partial pressure or concentration of each candidate gas component;

[0043] S9 outputs the quantitative analysis results of the partial pressure or concentration of each candidate gas component, including the partial pressure or concentration of the gas component, and further outputs the model fitting residuals, trend curves and diagnostic information.

[0044] The relevant parameters and symbols are defined as follows:

[0045] The mass spectrometry detection unit 6 is used to perform mass spectrometry analysis on the gas to be tested. It is preferably a residual gas analyzer (RGA), and its core analytical component is a mass spectrometer (MS). The mass spectrometer is preferably a quadrupole mass spectrometer (QMS).

[0046] The spectral detection unit 7 is preferably an optical emission spectrometer (OES).

[0047] RGA is used to acquire ion current signals of the gas under test in channels with different mass-to-charge ratios, while OES is used to acquire spectral signals of the gas under test in channels with different wavelengths.

[0048] To describe the physical quantities and mathematical relationships in the joint modeling process, the following symbols are further defined:

[0049] m is the ion mass, z is the ion charge number, m / z is the mass-to-charge ratio; λ is the spectral wavelength; Mass spectrometry channel index; For spectral channel indexing; For the first The candidate gas component number is n; n is the total number of candidate gas components. For the first Current measurement noise and background noise corresponding to each mass spectrometry channel; For the first Optical background noise and count noise corresponding to each spectral channel; Counts is the count value.

[0050] The present invention establishes a mass spectrometry observation model and a spectral observation model respectively for the quantitative analysis of overlapping gas components based on the fusion of mass spectrometry and spectral information.

[0051] In S3, the mass spectrometry observation model and the spectral observation model are unified into a linear forward model with gas partial pressure or concentration as unknowns.

[0052] Specifically, for the first Mass spectrometry channels (m / z= ) Measured ion current With the Spectral channels (spectral wavelength λ= Light intensity (unit: nm) They are described as follows:

[0053] ;

[0054] ;

[0055] in, Let be the partial pressure of the j-th gas (to be solved), and be a common state variable; and These are the mass spectrometry sensitivity coefficient and the spectral sensitivity coefficient for the j-th gas, respectively. For this gas at m / z= Fragment abundance coefficient at the location, This indicates that the gas is in the spectral channel. The relative emission intensity factor at that location.

[0056] One of the key aspects of quantitative analysis of overlapping gas components based on the fusion of mass spectrometry and spectral information is sensitivity calibration.

[0057] In S4, the overlapping gas component quantitative analysis system that integrates mass spectrometry and spectral information uses calibration data of known standard gases or standard gas mixtures, combined with a calibration parameter database, to determine the mass spectrometry sensitivity coefficient of the j-th gas. Spectral sensitivity coefficient of the j-th gas Simultaneously, single-component standard gases or multi-component standard mixtures can be used to calibrate and update the mass spectrometry sensitivity coefficient and spectral sensitivity coefficient, respectively, and the updated mass spectrometry sensitivity coefficient and spectral sensitivity coefficient can be stored in the calibration parameter database.

[0058] In S4, the matrix correction determines or updates the sensitivity parameters of each gas component based on the known partial pressure or concentration of each standard gas component and the corresponding mass spectrometry and spectral signals. Based on this, the relevant parameters in the forward model are adjusted to achieve the desired joint response matrix. The calibration improves the accuracy of its characterization of the response of actual detection systems.

[0059] In S5, to unify the representation of mass spectrometry detection information and spectral detection information within the same inversion framework, a mass spectrometry observation vector is constructed. Spectral observation vector Joint observation vector and joint response matrix as follows:

[0060] ;

[0061] ;

[0062] ;

[0063] ;

[0064] in, , , ..., These are the ion current signals measured by mass spectrometry in the 1st to Mth mass spectrometry channels, where the mass-to-charge ratio corresponding to the i-th mass spectrometry channel is m / z=i, i=1, 2, ..., M. , , ..., These represent the light intensity signals measured in the 1st to Kth spectral channels, respectively, where the wavelength corresponding to the kth spectral channel is... , k=1,2,…,K. Where K is the total number of spectral channels, representing the number of wavelength sampling points or characteristic spectral lines acquired by the spectral detection unit; M is the total number of mass spectrometry channels.

[0065] Let be the vector of partial pressures or concentrations of the candidate gas components, where , , ..., The gas partial pressures of the first to nth species are to be determined; This is the mass spectrum response submatrix. For the spectral response submatrix and the mass spectral response submatrix From fragment abundance coefficient With mass spectrometry sensitivity coefficient S j Composition, spectral response submatrix Based on relative emission intensity factor With spectral sensitivity coefficient G j constitute.

[0066] Through the above construction, mass spectrometry observations and spectral observations can be uniformly represented as a joint observation vector with the partial pressure or concentration of candidate gas components as unknowns. With joint response matrix The joint linear model is obtained. This provides a foundation for subsequent normalization and weighted construction, formation of weighted overdetermined equations, and data fusion and inversion solutions.

[0067] In S6, in actual measurement, due to the first Ion current measured by each mass spectrometer channel It is usually measured in amperes, while the first Light intensity of each spectral channel Typically expressed as counts, the two types of data differ significantly in units, orders of magnitude, and noise levels. If directly applied to the joint observation vector... Concatenating and solving these parameters can easily lead to the failure of the least squares method (i.e., large numerical terms dominate the error function). Therefore, this invention addresses this issue by constructing a joint observation vector. With joint response matrix Then, a weight matrix is ​​introduced. Normalization and weighting are performed to construct the weight matrix. Preferably, the weight matrix... It is a diagonal matrix, and its diagonal elements can be determined based on the reciprocal of the noise variance, signal-to-noise ratio, or reciprocal of the maximum signal amplitude of each channel, so as to achieve uniform scaling of mass spectrometry data and spectral data.

[0068] In S7, the weighted overdetermined equations are constructed:

[0069] ;

[0070] in, For joint observation vectors, For the joint response matrix, This represents the partial pressure or concentration vector of the candidate gas components. This is the weight matrix. This represents the joint noise vector. The weighted overdetermined equations constitute an overdetermined linear system when the number of observation equations exceeds the number of candidate gas components to be solved.

[0071] S8 includes solving the weighted overdetermined equations under physical constraints to obtain the partial pressure or concentration results of each candidate gas component.

[0072] The solution can be obtained using ordinary least squares, least squares with non-negativity constraints, regularization methods, or maximum a posteriori estimation based on Bayesian probability. Least squares with non-negativity constraints ensures the solution meets the physical constraint of non-negativity of gas partial pressure or concentration; regularization methods improve the joint response matrix. Solution stability under ill-conditioned conditions; a maximum a posteriori estimation method based on Bayesian probability is used to improve the robustness of the solution results when there is severe overlapping interference or prior information. Preferably, it can be based on the joint response matrix. Choose the appropriate solution method based on the condition number, fitting residuals, and prior information.

[0073] After data fusion and inversion, the quantitative analysis results of the gas are output. The output results include the partial pressure or concentration of each candidate gas component, and may further include fitting residuals, trend curves, uncertainty assessment results, and diagnostic reports to meet the application requirements of real-time monitoring and batch analysis.

[0074] The quantitative analysis method for overlapping gas components based on the fusion of mass spectrometry and spectral information of the present invention is implemented by software within the system. The software can be deployed in the data processing and computing unit 8 to perform steps such as data preprocessing and synchronization, species matching and screening, forward model construction, sensitivity calibration and matrix correction, normalization and weighted construction, data fusion and inversion solution, and result output. This enables automatic processing, joint analysis, and user interaction of mass spectrometry data and spectral data, and outputs quantitative analysis results of the partial pressure or concentration of each candidate gas component.

[0075] This invention relates to a quantitative gas component analysis system and method based on the fusion of mass spectrometry and spectral information. This system is used for multi-component gas analysis in nuclear fusion devices and other high-vacuum industrial environments, and is suitable for monitoring residual gases and process waste gases inside equipment. The method is implemented in software, which can be deployed on a regular personal computer or an embedded industrial control computer, and connects to the mass spectrometry detection unit 6 and the spectral detection unit 7 via a standard communication interface. The system can output partial pressure information obtained from mass spectrometry detection and relative light intensity information obtained from spectral detection, and can output the quantitative partial pressure or concentration results of each candidate gas component in fusion analysis mode. The results interface can further provide real-time data display, historical trend analysis, and diagnostic information, facilitating use by users in real-time monitoring and batch analysis scenarios.

[0076] The technical solution of the present invention will be further described below with reference to embodiments. Figure 1 As shown, this embodiment provides a quantitative analysis system for overlapping gas components based on the fusion of mass spectrometry and spectral information. The hardware structure has been optimized for a wide pressure range. It includes a vacuum chamber to be tested 1, a first vacuum pipe 2, a first valve 3, a vacuum gauge 4, a differential device 5, a mass spectrometry detection unit (such as a residual gas analyzer RGA) 6, a spectral detection unit (such as an optical emission spectrometer OES) 7, a data processing and calculation unit 8, a user interaction and display interface 9, a second valve 10, a second vacuum pipe 11, a turbomolecular pump 12, a third vacuum pipe 13, a third valve 14, and a backing pump 15.

[0077] The vacuum chamber 1 to be tested is connected to the first valve 3 via the first vacuum pipe 2. The output of the first valve 3 is connected to the differential device 5 to control the flow rate of the gas to be analyzed entering the differential device 5. The gas to be analyzed enters the differential device 5 from the vacuum chamber 1 via the first vacuum pipe 2 and the first valve 3, and then enters the subsequent detection area. The vacuum gauge 4 is installed on the differential device 5 to monitor the internal pressure state of the differential device 5 in real time.

[0078] The differential device 5 is connected to both the mass spectrometry detection unit 6 and the spectral detection unit 7. The mass spectrometry detection unit 6 is used for mass spectrometry detection of the analyte gas, and is preferably a quadrupole mass spectrometer (QMS). The spectral detection unit 7 is used for spectral detection of the analyte gas, preferably an optical emission spectrometer (OES). Specifically, the mass spectrometry detection unit 6 acquires ion current signals of the analyte gas in multiple mass-to-charge ratio channels, and the spectral detection unit 7 acquires characteristic spectral signals of the analyte gas in multiple wavelength channels.

[0079] Both the mass spectrometry detection unit 6 and the spectral detection unit 7 are communicatively connected to the data processing and calculation unit 8. The data processing and calculation unit 8 receives the detection data output from the mass spectrometry detection unit 6 and the spectral detection unit 7, and performs operations such as synchronous trigger control, data acquisition, species matching and screening, forward model construction, sensitivity calibration and matrix correction, joint observation vector and joint response matrix construction, normalization and weighted construction, constrained inversion solution, and result output. The user interaction and display interface 9 is connected to the data processing and calculation unit 8, and is used to display the partial pressure or concentration results of each candidate gas component, and provides parameter setting, data viewing, and result interaction functions.

[0080] The lower end of the differential device 5 is connected to the second vacuum pipe 11 via the second valve 10, and the second vacuum pipe 11 is connected to the turbomolecular pump 12. The outlet end of the turbomolecular pump 12 is connected to the third valve 14 via the third vacuum pipe 13, and the third valve 14 is connected to the backing pump 15. The second valve 10, the second vacuum pipe 11, the turbomolecular pump 12, the third vacuum pipe 13, the third valve 14, and the backing pump 15 together constitute a vacuum pumping unit, which is used to continuously pump air from the differential device 5 to maintain a low-pressure or high-vacuum environment suitable for mass spectrometry and spectral detection inside the differential device 5.

[0081] In this embodiment, the gas to be tested enters the differential device 5 after being throttled by the first valve 3. Under the synergistic action of the turbomolecular pump 12 and the forepump 15, the differential device 5 establishes a stable low-pressure detection environment, enabling the mass spectrometry detection unit 6 and the spectral detection unit 7 to detect the gas under suitable operating pressure conditions. Simultaneously, the vacuum gauge 4 monitors the internal pressure of the differential device 5 in real time, and the data processing and calculation unit 8 synchronously acquires and processes the mass spectrometry and spectral data, ultimately outputting the analysis results through the user interaction and display interface 9.

[0082] To reduce the differences in response mechanism and acquisition timing between mass spectrometry detection unit 6 and spectral detection unit 7, data processing and calculation unit 8 can send synchronous trigger signals to mass spectrometry detection unit 6 and spectral detection unit 7 according to a preset cycle, so that the scanning time window of mass spectrometry detection unit 6 and the integration time window of spectral detection unit 7 start at the same trigger time, ensuring time series alignment.

[0083] After each acquisition, the two types of detection units transmit the corresponding mass spectrometry and spectral data to the data processing and calculation unit 8, along with a timestamp, for subsequent fusion modeling and inversion solutions. Finally, the data processing and calculation unit 8 outputs the quantitative analysis results of the partial pressure or concentration of gas components to the user interface 9 for display. Users can interact with the data processing and calculation unit 8 through the user interface 9 to set and adjust data acquisition and algorithm parameters, and select to display mass spectrometry results, spectral results, or fusion analysis results as needed.

[0084] like Figure 2 As shown, the data processing and fusion inversion algorithm executed by data processing and computing unit 8 specifically includes the following steps:

[0085] S1, Data Preprocessing and Synchronization: The data processing and calculation unit 8 synchronously acquires the mass spectrometry data and spectral data of the gas to be tested from the mass spectrometry detection unit 6 and the spectral detection unit 7, respectively, and performs data preprocessing and synchronization on the original signals (i.e., the mass spectrometry data and spectral data of the gas to be tested) to complete denoising, baseline subtraction, peak extraction and time alignment.

[0086] S2, Species Matching and Screening: The system has a built-in species feature database, which pre-stores fragment abundance information, characteristic spectral line information, and relative response information of candidate gas components at multiple mass-to-charge ratio positions, multiple wavelength channels, and other relevant information. Based on the preprocessed mass spectrometry and spectral data, and in conjunction with the species feature database, the system performs species matching and screening on the candidate gas components to obtain a set of candidate gas components to be solved. Common candidate gases may include H2, D2, He, H2O, N2, O2, CO, CO2, CH4, etc.

[0087] S3, Forward Model Construction: Based on the candidate gas component set obtained in S2, and combined with fragment abundance information, characteristic spectral line information, and relative response information from the species characteristic database, a mass spectrometry observation model and a spectral observation model are established respectively, and a linear forward model is constructed. Specifically, the ion current signal measured for the i-th mass spectrometry channel... The spectral signal measured in the k-th spectral channel Linear forward models were established for the candidate gas partial pressure or concentration as defined above, respectively.

[0088] S4, Sensitivity Calibration and Matrix Correction: Based on the calibration results of standard gases or standard gas mixtures, and in conjunction with the calibration parameter database, sensitivity parameters in the linear forward model are calibrated and matrix corrected to obtain the joint response matrix. Mass spectrometry sensitivity parameters and spectral sensitivity parameters The sensitivity calibration and matrix correction are performed based on the known partial pressures or concentrations of each standard gas component and their corresponding mass spectrometry and spectral data. This process calculates and adjusts the relevant parameters in the linear forward model to achieve the desired joint response matrix. Calibration.

[0089] S5, Constructing the joint observation vector With joint response matrix Based on the model parameters after sensitivity calibration and matrix correction, a mass spectrometry observation vector is constructed. Spectral observation vector Joint observation vector and joint response matrix Specifically, the observations from multiple mass spectrometry channels are combined to form a mass spectrometry observation vector. The observations from multiple spectral channels are combined to form a spectral observation vector. Then the mass spectrometry observation vector With spectral observation vector splicing to form a joint observation vector Simultaneously, a mass spectrometry response submatrix is ​​constructed based on the forward model parameters. and spectral response submatrix Then, they are spliced ​​together to form a joint response matrix. .

[0090] S6, Normalization and Weighted Construction: Due to the significant differences between mass spectrometry signals and spectral signals in terms of dimensions, orders of magnitude, and noise levels, data processing and computation unit 8 constructs a joint observation vector. With joint response matrix Then, a weight matrix is ​​introduced. Normalization and weighting are performed. Preferably, the weight matrix... It is a diagonal matrix, and its diagonal elements can be determined based on the inverse of the noise variance, signal-to-noise ratio, or inverse of the maximum signal amplitude of each mass spectrum channel and each spectral channel, so as to achieve unified scaling of the two types of data and reduce the imbalance of different modal observations in joint solution.

[0091] S7, Constructing weighted overdetermined equations: Based on joint observation vectors Joint response matrix and weight matrix Construct weighted overdetermined equations: The diagonal elements of W correspond to the weighting coefficients of each mass spectrum channel and each spectral channel, respectively. : When the number of observation equations exceeds the number of candidate gas components to be solved, the weighted overdetermined equations constitute an overdetermined linear system of equations.

[0092] S8, Data Fusion and Inversion Solution: Solve the weighted overdetermined equations under physical constraints to obtain the partial pressure or concentration results of each candidate gas component. Preferably, the solution can employ ordinary least squares, least squares with non-negativity constraints, regularization methods, or maximum a posteriori estimation based on Bayesian probability. The least squares with non-negativity constraints ensures that the solution results satisfy the physical constraint of non-negativity of gas partial pressure or concentration; the regularization method improves the solution stability under ill-conditioned conditions of the joint response matrix; and the maximum a posteriori estimation based on Bayesian probability improves the robustness of the solution results when there is severe overlap interference or prior information.

[0093] S9, Output Gas Quantitative Analysis Results: The quantitative analysis results of the partial pressure or concentration of each candidate gas component are output to the user interaction and display interface 9 for display. The results include not only the partial pressure or concentration of each candidate gas component, but also the fitting residual, trend curve, uncertainty assessment results, and diagnostic reports. Users can interact with the data processing and calculation unit 8 through the user interaction and display interface 9 to set and adjust data acquisition parameters and algorithm parameters, and select the display mode for mass spectrometry results, spectral results, or fusion analysis results as needed.

[0094] The specific equipment models, brands, or experimental apparatus names mentioned in the specification and accompanying drawings are only used to explain the preferred embodiments of the technical solutions of the present invention and do not constitute a limitation on the scope of protection of the present invention. It should be understood that the mass spectrometry detection unit 6 described in the present invention is not limited to the quadrupole mass spectrometer (QMS) and residual gas analyzer (RGA) mentioned in the embodiments. It can also be a time-of-flight mass spectrometer (TOF-MS), magnetic mass spectrometer, ion trap mass spectrometer, or other detection devices capable of acquiring gas mass-to-charge ratio information. The spectral detection unit 7 is also not limited to an optical emission spectrometer (OES). Any device capable of acquiring gas characteristic spectral signals, such as an absorption spectrometer, Raman spectrometer, laser-induced breakdown spectrometer, or other spectral detection devices suitable for gas composition analysis, can be used as one of the implementation methods of the present invention.

[0095] Furthermore, this invention can be applied not only to the field of vacuum residual gas analysis, but also to gas composition monitoring in low-pressure discharge environments, mixed gas diagnosis in the fuel cycle and wall treatment process of nuclear fusion devices, process gas and tail gas analysis in semiconductor manufacturing processes, outgas monitoring in vacuum coating and material heat treatment processes, online detection of multi-component gases in chemical processes, trace gas identification in environmental monitoring, fuel gas composition analysis in energy systems, and other scenarios that require qualitative or quantitative analysis of complex mixed gases.

[0096] Any technical solution that achieves the identification and quantitative analysis of overlapping gas components through the fusion of mass spectrometry and spectral information, via joint modeling, joint calibration, weighted processing, and constrained inversion, should fall within the protection scope of this invention. Any equivalent substitutions or modifications made by those skilled in the art based on the technical solutions disclosed in this invention should also fall within the protection scope of this invention.

[0097] Those skilled in the art will understand that embodiments of the present invention, based on the fusion of mass spectrometry and spectral information for quantitative analysis of overlapping gas components, can be provided as a method, system, or computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The embodiments of the present invention can be implemented using various computer languages.

[0098] The above description is merely an embodiment of the present invention and does not limit the scope of the invention. Any equivalent structural or procedural transformations made based on the description and drawings of this invention, or direct or indirect applications in other related system fields, are similarly included within the protection scope of this invention. Contents not described in detail in this specification are prior art known to those skilled in the art.

Claims

1. An overlapping gas component fusion quantitative analysis system characterized by, include: The vacuum chamber under test, the first vacuum pipeline, the first valve, the vacuum gauge, the differential device, the mass spectrometry detection unit, the spectral detection unit, the data processing and calculation unit, the user interaction and display interface, and the vacuum pumping unit; The vacuum chamber to be tested is connected to the first valve through the first vacuum pipe. The output end of the first valve is connected to the differential device. The gas to be tested enters the differential device through the first vacuum pipe and the first valve. The differential device is connected to the mass spectrometry detection unit, the spectral detection unit, and the vacuum pumping unit, respectively. A vacuum gauge is installed on the differential device to monitor the internal pressure of the differential device; a mass spectrometry detection unit is used to collect ion current signals of the gas to be tested in multiple mass-to-charge ratio channels; and a spectral detection unit is used to collect characteristic spectral signals of the gas to be tested in multiple wavelength channels. The data processing and calculation unit is connected to the mass spectrometry detection unit and the spectral detection unit respectively, and is used to perform synchronous trigger control, data acquisition, database matching, joint calibration, weighted modeling, constraint inversion solution and result output; the user interaction and display interface is connected to the data processing and calculation unit, and is used to display the partial pressure or concentration results of each candidate gas component and the corresponding trend information.

2. The overlapping gas component fusion quantitative analysis system according to claim 1, characterized by, The vacuum pumping unit includes a second valve, a second vacuum pipe, a turbomolecular pump, a third vacuum pipe, a third valve, and a backing pump connected in sequence.

3. An overlapping gas component fusion quantitative analysis method for the overlapping gas component fusion quantitative analysis system according to claim 1 or 2, characterized by, include: S1, simultaneously acquires mass spectrometry and spectral data of the gas to be tested, and performs data preprocessing and synchronization; S2, based on the preprocessed and synchronized mass spectrometry and spectral data, and combined with the fragment abundance information, characteristic spectral line information and relative response information of candidate gas components at multiple mass-to-charge ratio positions pre-stored in the species characteristic database, the candidate gas components are matched and screened to obtain the set of candidate gas components to be solved. S3. Based on the candidate gas component set obtained in S2, and combined with the fragment abundance information, characteristic spectral line information and relative response information of the candidate gas components at multiple mass-to-charge ratio positions pre-stored in the species characteristic database, a linear forward model is constructed. S4, according to the calibration results of standard gas or standard mixed gas, combining with the calibration parameter database, sensitivity calibration and matrix correction are performed on the sensitivity parameters in the linear forward model to obtain the combined response matrix of mass spectrum sensitivity parameters and spectral sensitivity parameters ; S5, based on the mass spectral sensitivity parameter and the spectral sensitivity parameter constructing a joint observation vector with a joint response matrix ; S6, normalizing and weighting the mass spectrometry data and the spectral data and constructing a weight matrix ; S7, constructing a weighted overdetermined equation based on the joint observation vector , a joint response matrix , and a weight matrix ​ S8, Solve the weighted overdetermined equations under physical constraints to obtain the partial pressure or concentration of each candidate gas component; S9 outputs the quantitative analysis results of the partial pressure or concentration of each candidate gas component, including the partial pressure or concentration of the gas component, and outputs the model fitting residuals, trend curves and diagnostic information.

4. The quantitative analysis method for overlapping gas components according to claim 3, characterized in that, S2 includes: The overlapping gas component fusion quantitative analysis system has a built-in species characteristic database, which pre-stores fragment abundance information of candidate gas components at multiple mass-to-charge ratio positions, characteristic spectral line information at multiple wavelength channels, and relative response information. Based on the preprocessed mass spectrometry and spectral data, and combined with the species characteristic database, species matching and screening of candidate gas components are performed to obtain the set of candidate gas components to be solved.

5. The quantitative analysis method for overlapping gas components according to claim 4, characterized in that, S3 includes: the Ion current measured in each mass spectrometer channel The sum of the partial pressures of each candidate gas component multiplied by its corresponding mass spectrometry sensitivity coefficient and its corresponding fragment abundance coefficient, plus the first... Current measurement noise and background noise corresponding to each mass spectrometry channel; No. Light intensity of each spectral channel It is described as the sum of the partial pressure of each candidate gas component multiplied by the corresponding spectral sensitivity coefficient and the corresponding relative emission intensity factor, plus the optical background noise and counting noise corresponding to the spectral channel.

6. The quantitative analysis method for overlapping gas components according to claim 5, characterized in that, In S4, the matrix correction determines or updates the sensitivity parameters of each gas component based on the known partial pressure or concentration of each standard gas component and the corresponding mass spectrometry and spectral signals. Based on this, the parameters in the forward model are adjusted to achieve the joint response matrix. Calibration.

7. The quantitative analysis method for overlapping gas components according to claim 6, characterized in that, S5 include: Constructing mass spectrometry observation vectors Spectral observation vector Joint observation vector and joint response matrix as follows: Mass spectral observation vector The spectral observation vector is formed by sequentially arranging the ion current signals measured by mass spectrometry in the 1st to Mth mass spectrometry channels; The joint observation vector is composed of the light intensity signals measured in the first to the Kth spectral channels arranged sequentially. From mass spectrometry observation vector and spectral observation vector Composed of top and bottom sections; Joint Response Matrix From the mass spectrum response submatrix and spectral response submatrix Composed of vertically joined components; mass spectrometry response submatrix From fragment abundance coefficient With mass spectrometry sensitivity coefficient S j Composition, spectral response submatrix Based on relative emission intensity factor With spectral sensitivity coefficient G j constitute.

8. The quantitative analysis method for overlapping gas components according to claim 7, characterized in that, In S7, the weight matrix in S6 It is a diagonal matrix, and its diagonal elements are determined based on the reciprocal of the noise variance, signal-to-noise ratio, or reciprocal of the maximum signal amplitude of each channel, so as to achieve uniform scaling of mass spectrometry data and spectral data.

9. The quantitative analysis method for overlapping gas components according to claim 8, characterized in that, In S7, a weight matrix is ​​introduced in S6. Based on this, the joint observation vector Joint response matrix and candidate gas component partial pressure or concentration vector Weighting is performed to form a weighted overdetermined equation for inversion solution.

10. The method for quantitative analysis of overlapping gas components according to claim 1, characterized in that, In S8, the weighted overdetermined equations are solved using ordinary least squares, least squares with non-negative constraints, regularization methods, or maximum a posteriori estimation based on Bayesian probabilities.

Images