Boric acid concentration detector for pressurized water reactor neutron-free radiation source

Abstract

The present application discloses a boric acid concentration detector for a pressurized water reactor neutronless radiation source, comprising: a sampling tube component; an optical path system, the optical path system comprising a Raman probe assembly and a signal sampling unit; a signal processing system, for acquiring a data signal characterizing a Raman spectrum sent by the signal sampling unit, and processing the data signal to determine the boric acid concentration in a solution to be tested; wherein a calibration object is provided in the optical path system; the signal sampling unit acquires a Raman spectrum of a solution at a specific concentration containing a Raman characteristic peak of the calibration object as a standard spectrum, and the signal sampling unit is also used to detect the solution to be tested to obtain an original Raman measurement spectrum; the signal processing system calibrates the original Raman measurement spectrum based on the standard spectrum to obtain a Raman calibration spectrum, and the signal processing system also obtains the concentration of the solution to be tested based on the Raman calibration spectrum. The detector provided by the present application can be used for online, high-precision quantitative detection of boric acid concentration in a pressurized water reactor neutronless radiation source.

Description

Technical Field

[0001] The present application relates to the technical field of concentration measurement, and in particular to a boric acid concentration detector for a pressurized water reactor neutron-free radiation source. Background Art

[0002] Pressurized water reactor nuclear power plants usually use boron addition or dilution in the primary cooling water to achieve compensatory control of reactivity, flatten the core power distribution, and increase the burnup depth; in addition, during shutdown, refueling or overhaul, nuclear power plants must also add boron to keep the active area in a subcritical state for a long time. In order to prevent the accidental dilution of boron, a complete and independent online boron concentration measurement system is required to monitor the boron concentration in the reactor and the primary circuit system coolant for a long time to ensure the safety of reactor operation.

[0003] Based on this, the present application provides a solution concentration monitoring system based on Raman spectrometer, which is applied to the concentration detection of pressurized water reactor boric acid solution. The system can realize online measurement of pressurized water reactor boron concentration with the help of Raman spectrometer. However, the quantitative concentration detection of Raman spectrometer in the related art has the following difficulties:

[0004] 1. During the spectrum acquisition process, there are signal fluctuations in the laser light source, light detection device and data processing circuit. This fluctuation causes the Raman spectrum of the same concentration of boric acid solution to be unstable. In addition, the intensity of Raman scattering light is about 10 times that of the scattered light intensity dominated by Rayleigh scattering. -6 ~10 -10 , which is a very weak light signal. The above fluctuations even exceed the changes in the intensity of the Raman characteristic peak caused by changes in concentration, which leads to huge errors in quantitative detection. The traditional calibration using the water Raman peak as the internal standard will also cause changes in peak intensity due to solute changes. Therefore, for the application of Raman spectroscopy for high-precision quantitative detection, high-precision calibration is a major problem. Especially at low boric acid concentrations (<100 mg / L), the Raman signal is extremely weak, and high-precision detection is more dependent on high-precision calibration.

[0005] 2. In aqueous solution, the intensity of the characteristic peak of boric acid is affected by the fluorescence signal and environmental impurities, which makes it difficult to achieve high-precision detection based on the model established based on the linear relationship between the characteristic peak intensity and the boric acid concentration. Therefore, establishing a high-precision quantitative model for the concentration of boric acid solution is the key to using Raman spectroscopy to carry out boric acid concentration detection.

[0006] 3. In general, quantitative concentration detection algorithms are only applicable to the equipment that collects data when the algorithm is established. When the established quantitative detection algorithm is migrated to other equipment, the algorithm is not applicable due to deviations between instrument components and optical structures. This is another important issue that hinders the application of Raman spectroscopy for quantitative detection and its promotion.

[0007] According to the above description, the quantitative concentration detection of Raman spectrometer in the related technology has the problems of great difficulty in online acquisition, low quantitative detection accuracy (especially at low concentrations, such as <100 mg / L), and large quantitative detection error, which restricts the application of Raman spectrometer in the long-term online measurement of boron concentration in pressurized water reactors. Summary of the invention

[0008] In order to solve the deficiencies of the prior art, the present application provides a boric acid concentration detector for a pressurized water reactor neutron-free radiation source that can collect data online and detect with high precision.

[0009] To achieve the above objectives, this application adopts the following technical solutions:

[0010] A boric acid concentration detector for a pressurized water reactor neutron-free radiation source, comprising:

[0011] A sampling tube component, used for holding the solution to be tested;

[0012] An optical system is connected to the sampling tube component, and the optical system includes a Raman probe assembly and a signal sampling unit. The Raman probe assembly is used to emit a laser beam to the solution to be tested and receive a Raman scattering signal generated when the laser beam acts on the solution to be tested. The signal sampling unit receives the Raman scattering signal through the Raman probe assembly and forms a Raman spectrum.

[0013] A signal processing system is connected to the optical path system, and acquires a data signal representing the Raman spectrum sent by the signal sampling unit, and processes the data signal to determine the boric acid concentration in the solution to be tested;

[0014] Wherein, a calibration object is arranged in the optical path system;

[0015] The signal sampling unit obtains the Raman spectrum of the solution at a specific concentration containing the Raman characteristic peak of the calibration substance as the standard spectrum. The signal sampling unit is also used to detect the solution to be tested to obtain the original Raman measurement spectrum;

[0016] The signal processing system obtains the Raman scattered light intensity within a preset range around the characteristic peak position of the calibration object based on the standard spectrum, and calibrates the original Raman measurement spectrum based on the Raman scattered light intensity to obtain the Raman calibration spectrum; the signal processing system also obtains the concentration of the solution to be tested based on the Raman calibration spectrum.

[0017] Furthermore, the signal processing system is used to set a pixel interval, and the characteristic peak of the calibration object is located in the pixel interval; the signal processing system is also used to obtain a standard value and a reference value, wherein the standard value is the Raman scattering light intensity in the pixel interval calculated at any solution concentration, and the reference value is the Raman scattering light intensity in the pixel interval calculated at other solution concentrations;

[0018] The signal processing system calculates the ratio between the standard value and the reference value to obtain calibration data, and calibrates the original Raman measurement spectrum at the corresponding solution concentration based on the calibration data.

[0019] Furthermore, both the standard spectrum and the original Raman measurement spectrum are spectra without baseline correction.

[0020] Furthermore, the signal processing system comprises:

[0021] A sample extraction unit, used to obtain a first Raman spectrum formed by the interaction between a boric acid solution of a first preset concentration and a laser beam;

[0022] A feature extraction unit, used to extract a first interval covering a characteristic peak characterizing boric acid in the first Raman spectrum, and use the pixel points included in the first interval and the light intensity values ​​corresponding to the pixel points as first spectrum data;

[0023] A model building unit, wherein the model building unit builds a regression model based on the first spectral data and the first preset concentration and based on a Lasso regression algorithm;

[0024] The concentration detection unit is used to detect the actual concentration of the boric acid solution of unknown concentration according to the regression model.

[0025] Furthermore, the sample extraction unit is further used to obtain a second Raman spectrum formed by the interaction between the boric acid solution of the second preset concentration and the laser beam; the feature extraction unit is further used to extract a second interval covering a characteristic peak of boric acid in the second Raman spectrum, and use the pixel points contained in the second interval and the light intensity values ​​corresponding to the pixel points as second spectrum data;

[0026] The signal processing system also includes a result verification unit, which is used to obtain second spectral data and input the second spectral data into the regression model. The result verification unit compares the calculated data generated by the regression model with the second spectral data to obtain a fit index of the regression model and determines whether the fit index is greater than a fit threshold. The model building unit outputs the regression model when the fit index is greater than the fit threshold, and adjusts the penalty coefficient in the Lasso regression algorithm when the fit index is less than or equal to the fit threshold to generate an updated regression model.

[0027] Furthermore, the sample extraction unit is used to extract the number of pixels in the first spectral data or the second spectral data, and the sample extraction unit is also used to store the first preset concentration and the second preset concentration, and extract the number of concentrations of the boric acid solution with different preset concentrations;

[0028] The model building unit generates an updated regression model according to the number of pixels, the number of concentrations, and the updated penalty coefficient. The objective function of the updated regression model satisfies the following relationship:

[0029] ;

[0030] Wherein, X represents the first spectral data and the second spectral data, y represents the first preset concentration and the second preset concentration, α represents the penalty coefficient, β represents the regression coefficient, N represents the number of concentrations, and P represents the number of pixels.

[0031] Furthermore, both the first Raman spectrum and the second Raman spectrum are Raman spectra that have not been subjected to background subtraction data processing.

[0032] Furthermore, the signal processing system comprises:

[0033] A migration model construction unit, the migration model construction unit is used to obtain a first model and a second model trained in a first pressurized water reactor neutronless radiation source boric acid concentration detector, the first model being a functional relationship between the Raman characteristic spectrum intensity of a target substance at a preset concentration under standard conditions and the focal position of the first pressurized water reactor neutronless radiation source boric acid concentration detector, and the second model being a functional relationship between the Raman characteristic spectrum intensity of a calibration object under standard conditions and the focal position of the first pressurized water reactor neutronless radiation source boric acid concentration detector;

[0034] A focus position inversion unit, which inverts the focus position of the second pressurized water reactor neutronless radiation source boric acid concentration detector based on the second model and the Raman characteristic spectrum intensity of the calibration object collected from the second pressurized water reactor neutronless radiation source boric acid concentration detector under standard conditions;

[0035] The concentration measurement model construction unit is used to obtain a Raman characteristic spectrum intensity correction coefficient according to the focal position of the second pressurized water reactor neutronless radiation source boric acid concentration detector in combination with the first model, and combine the Raman characteristic spectrum intensity correction coefficient with the first concentration measurement model to obtain a second concentration measurement model, wherein the first concentration measurement model is obtained by training the first pressurized water reactor neutronless radiation source boric acid concentration detector under standard conditions, and the second pressurized water reactor neutronless radiation source boric acid concentration detector calculates the concentration of the target substance based on the second concentration measurement model.

[0036] Furthermore, the first model is expressed by the following formula:

[0037] ;

[0038] In the formula, I A0 It represents the intensity of the Raman characteristic spectrum detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector for the target substance of preset concentration under standard conditions and when the focus of the first pressurized water reactor neutronless radiation source boric acid concentration detector is located at the first focus position, I AnIt means that under standard conditions, the focus of the boric acid concentration detector of the first pressurized water reactor neutron-free radiation source is at the n In the case of the first pressurized water reactor neutronless radiation source boric acid concentration detector detecting the target substance of the preset concentration, the Raman characteristic spectrum intensity is f n Indicates that the focus of the boric acid concentration detector of the first pressurized water reactor neutron-free radiation source is at n At each location, F 1(*) indicates the first functional relationship;

[0039] The second model is expressed by the following formula:

[0040] ;

[0041] In the formula, I B0 It means that under standard conditions, and the focus of the first pressurized water reactor neutronless radiation source boric acid concentration detector is located at the first focus position, the intensity of the Raman characteristic spectrum detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector on the calibration object is, I Bn It means that under standard conditions, the focus of the boric acid concentration detector of the first pressurized water reactor neutron-free radiation source is at the n In the case of the position, the intensity of the Raman characteristic spectrum detected by the boric acid concentration detector of the first pressurized water reactor neutronless radiation source on the calibration object is f n Indicates that the focus of the boric acid concentration detector of the first pressurized water reactor neutron-free radiation source is at n At each location, F 2(*) indicates the second functional relationship.

[0042] Furthermore, the Raman characteristic spectrum intensity correction coefficient is expressed by the following formula:

[0043] ;

[0044] In the formula, F 1(*) indicates the first functional relationship, f’ Indicates the focal position of the boric acid concentration detector of the second pressurized water reactor neutron-free radiation source;

[0045] The first concentration measurement model is expressed by the following formula:

[0046] ;

[0047] In the formula, Q A represents the regression model under standard conditions in the boric acid concentration detector of the first pressurized water reactor neutronless radiation source, I AIt represents the intensity of the Raman characteristic spectrum detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector on the target substance under standard conditions and when the focus of the first pressurized water reactor neutronless radiation source boric acid concentration detector is located at the first focus position;

[0048] The second concentration measurement model is expressed by the following formula:

[0049] ;

[0050] In the formula, Q A represents the regression model under standard conditions in the boric acid concentration detector of the first pressurized water reactor neutronless radiation source, M represents the correction coefficient of the Raman characteristic spectrum intensity, I’ A It indicates the intensity of Raman characteristic spectrum detected by the boric acid concentration detector of the second pressurized water reactor neutronless radiation source on the target substance under standard conditions.

[0051] Furthermore, the optical path system includes an outer shell, and the Raman probe assembly and the signal sampling unit are both arranged in the outer shell;

[0052] The pressurized water reactor neutronless radiation source boric acid concentration detector also includes a flange window and a planar superlens installed on the flange window. The flange window is arranged between an outer shell and a sampling tube component. The outer shell is connected to the sampling tube component through the flange window. The planar superlens is arranged between the Raman probe assembly and the sampling tube component. The planar superlens acts on the laser beam so that the laser beam is focused and then acts on the solution to be tested to form a Raman scattering signal. The pressure resistance of the planar superlens is not less than 3MPa.

[0053] Furthermore, the surface of the planar superlens has a micro-nanohole structure so that the phase of the laser beam is changed sequentially from the center point of the planar superlens to its edge.

[0054] According to the above description, the pressurized water reactor neutronless radiation source boric acid concentration detector provided in the present application can be used for online, high-precision pressurized water reactor neutronless radiation source boric acid concentration quantitative detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of the structure of a boric acid concentration detector for a pressurized water reactor neutron-free radiation source provided in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of an uncalibrated raw Raman measurement spectrum provided by an embodiment of the present application;

[0057] Figure 3 for Figure 2 Schematic diagram of the calibrated spectrum;

[0058] Figure 4A schematic diagram of a spectrum before calibration of a repeatability test spectrum provided in an embodiment of the present application;

[0059] Figure 5 for Figure 4 Schematic diagram of the spectrum after calibration;

[0060] Figure 6 A structural block diagram of a signal processing system provided by an embodiment of the present invention;

[0061] Figure 7 A schematic diagram of a Raman spectrum of the solubility of a boric acid solution provided in an embodiment of the present invention;

[0062] Figure 8 An experimental data diagram of the relationship between the penalty coefficient and the fit index provided by the embodiment of the present invention;

[0063] Fig. 9 An error measurement diagram of a regression model established based on a Lasso regression algorithm provided in an embodiment of the present invention;

[0064] Fig.10 A schematic diagram of the model error obtained after training with spectral samples without baseline deduction and after baseline deduction provided in an embodiment of the present application;

[0065] Fig.11 A schematic diagram of a detector for a concentration measurement model migration method for quantitative detection based on Raman spectroscopy provided in an embodiment of the present application;

[0066] Fig.12 A schematic diagram of a Raman system provided in an embodiment of the present application monitoring an optical structure to obtain a Raman spectrum;

[0067] Fig.13 A schematic diagram of the migration of the concentration measurement model provided in an embodiment of the present application;

[0068] Fig.14 A flowchart for obtaining a first model provided in an embodiment of the present application;

[0069] Fig.15 A schematic diagram of training a first model in a first Raman system provided in an embodiment of the present application;

[0070] Fig.16 A flowchart for obtaining a second model provided in an embodiment of the present application;

[0071] Fig.17 A schematic diagram of training a second model in a first Raman system provided in an embodiment of the present application;

[0072] Fig.18 A schematic diagram of a metasurface lens provided in an embodiment of the present application;

[0073] Fig.19 A comparison chart of the spectrum collected by the Raman system with integrated superlens and flange window provided in an embodiment of the present application and the spectrum collected by the traditional Raman system. DETAILED DESCRIPTION

[0074] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the specific implementation manner of the present application will be clearly and completely described below in conjunction with the drawings in the implementation manner of the present application.

[0075] The quantitative concentration detection by Raman spectrometer in related technologies has the problems of great difficulty in online acquisition, low quantitative detection accuracy (especially at low concentrations, such as <100 mg / L), and large quantitative detection error, which restricts the application of Raman spectrometer in long-term online measurement of boron concentration in pressurized water reactors.

[0076] In this regard, the present application makes improvements in three aspects, namely, calibration of Raman spectroscopy, measurement model training, and measurement model migration, to achieve high test accuracy, high repeatability, and high reliability in monitoring the concentration of pressurized water reactor boric acid solution based on Raman spectrometer.

[0077] First,

[0078] The present application makes improvements to the boric acid concentration detector for pressurized water reactor neutron-free radiation sources, such as Figure 1 As shown, the embodiment of the present application provides a pressurized water reactor neutronless radiation source boric acid concentration detector (hereinafter referred to as the detector), which can perform quantitative concentration detection on the solution to be tested, including:

[0079] The sampling tube component 100 is used to contain the solution to be tested.

[0080] The optical path system 200 is connected to the sampling tube component 100. The optical path system 200 includes a Raman probe assembly 21 and a signal sampling unit 22. The Raman probe assembly 21 is used to emit a laser beam to the solution to be tested and receive a Raman scattering signal generated when the laser beam acts on the solution to be tested. The signal sampling unit 22 receives the Raman scattering signal through the Raman probe assembly 21 and forms a Raman spectrum.

[0081] The signal processing system 300 is connected to the optical path system 200, and obtains the data signal representing the Raman spectrum sent by the signal sampling unit 22, and processes the data signal to determine the boric acid concentration in the solution to be tested;

[0082] Wherein, a calibration object is provided in the optical path system 200;

[0083] The signal sampling unit 22 obtains the Raman spectrum of the solution at a specific concentration containing the Raman characteristic peak of the calibration object as the standard spectrum. The calibration object is set in the optical path of the Raman spectroscopy system, and the calibration object is not in the solution to be tested;

[0084] The signal sampling unit 22 is also used to detect the solution to be tested to obtain the original Raman measurement spectrum;

[0085] The signal processing system 300 calibrates the original Raman measurement spectrum based on the standard spectrum to obtain a Raman calibration spectrum. The signal processing system 300 also obtains the concentration of the solution to be measured based on the Raman calibration spectrum.

[0086] According to the above description, in the detector provided by the embodiment of the present application, by selecting an object other than the solution to be tested as a calibration object, the influence of the introduction of the calibration object on the solution to be tested is avoided, and the calibration object is not affected by factors such as the concentration and flow rate of the solution to be tested. Therefore, the Raman characteristic peak of the calibration object is only affected by environmental factors, laser power fluctuations and system errors. The detector provided by the embodiment of the present application uses the characteristic Raman spectrum of the calibration object as the standard spectrum and uses the standard spectrum to calibrate the original Raman measurement spectrum, which can eliminate the influence of environmental factors, laser power fluctuations and system errors, making the quantitative concentration detection of the solution to be tested using the Raman spectroscopy system more accurate.

[0087] As an optional implementation method, the detector provided in the embodiment of the present application can select the device capable of stimulating Raman light in the optical path system 200 as a calibration object, or, a device capable of stimulating Raman light can be inserted between the sampling tube component 100 and the optical path system 200 and the device can be used as a calibration object.

[0088] Specifically, in the optical path system 200, the light source emits a light beam, which is focused to the solution to be tested after passing through the Raman probe assembly 21, and Raman scattered light is excited at the solution to be tested, and the Raman scattered light is finally received by the signal sampling unit 22 in the optical path system 200, thereby obtaining a Raman spectrum. From the light source to the solution to be tested, there may be some devices that can also excite Raman light. In the related art, these Raman lights excited by the devices are generally regarded as stray light, and will not be considered for application during the test process. Even in the design and construction of the detector, optical devices with Raman activity will be avoided as much as possible to avoid interference with the Raman signal of the substance to be detected. However, in the present application, the original optical device with Raman activity in the system when exciting the Raman spectrum of the solution is specially selected as a calibration object. The spectral signal of the calibration object is usually regarded as a stray light signal, but it will not be affected by factors such as solution concentration and flow rate, so that the spectrum calibration can be more accurate.

[0089] In addition, the original optical device in the optical path system 200 is used as a calibration object, and the structural parameters and optical parameters of the calibration object are fixed, which is more convenient for spectral calibration. In addition, the method does not require the introduction of additional optical devices, so there is no need to consider the installation of the device when introducing additional calibration objects. The solution is low in cost and easy to operate.

[0090] It is easy to understand that in addition to selecting the original optical device in the optical system 200 as a calibration object, a substance with Raman activity can be inserted between the optical system 200 and the solution to be tested as a calibration object. In this way, the accurate calibration of the Raman spectrum can also be achieved.

[0091] As an optional implementation, in the embodiment of the present application, the calibration object is placed close to the solution to be tested, and is located near the focus of the light beam emitted by the optical path system 200.

[0092] As an optional implementation, in the embodiment of the present application, the material for the observation window is selected as the calibration object. It should be noted that the observation window is arranged on the pipeline through which the solution to be tested circulates, and the light beam emitted by the optical path system 200 is focused by the optical device and then irradiated to the solution to be tested through the observation window. It is easy to understand that the focus of the light beam emitted by the optical path system 200 is located at the solution to be tested. Therefore, in the overall light path of the optical path system 200, the position of the observation window is closest to the focus of the light beam, the observation window can obtain the maximum excitation energy, and the Raman light excited by the observation window is more stable. Therefore, in the embodiment of the present application, the observation window is specially selected as the calibration object, which is conducive to the calibration of the spectrum.

[0093] As an optional implementation, the wave number difference between the Raman characteristic peak of the calibration object and the Raman characteristic peak of the solution to be tested is greater than or equal to 10 cm -1 In this way, it can be ensured that there is a significant difference between the Raman characteristic peak of the optical device used as the calibration object and the Raman characteristic peak of the substance to be detected, thereby avoiding interference with the Raman signal of the substance to be detected.

[0094] As an optional implementation method, the wave number difference between the Raman characteristic peak of the calibration object and the Raman characteristic peak of the solution to be tested should be less than 1000 cm -1 In this way, it is possible to avoid a large difference between the Raman characteristic peaks of the calibration object and the Raman characteristic peaks of the substance to be detected.

[0095] According to the above description, the wave number difference between the Raman characteristic peak of the calibration object and the Raman characteristic peak of the test solution should be greater than or equal to 10 cm -1 , and, less than 1000 cm -1In this way, it can be ensured that there is an appropriate deviation between the optical device used for calibration and the Raman characteristic peak of the substance to be detected, thereby avoiding error problems when other photophysics are introduced (such as fluorescence effect). Taking the fluorescence effect as an example, if the two Raman characteristic peaks are not much different, then the changes of the two Raman characteristic peaks are linear. If the difference is large, there is a nonlinear factor between the changes of the two Raman characteristic peaks. Therefore, through the above settings, the optical device selected for calibration should have an appropriate deviation from the Raman characteristic peak of the substance to be detected, which can avoid error problems when other photophysics are introduced.

[0096] As an optional implementation, in the detector provided in the embodiment of the present application, the signal processing system 300 obtains the Raman scattering light intensity at the characteristic peak position of the calibration object based on the standard spectrum, and calibrates the original Raman measurement spectrum based on the Raman scattering light intensity.

[0097] Specifically, the signal processing system 300 is used to set a pixel interval, and the characteristic peak of the calibration object is located within the pixel interval; the signal processing system 300 is also used to obtain a standard value and a reference value, wherein the standard value is the Raman scattering light intensity within the pixel interval calculated at any solution concentration, and the reference value is the Raman scattering light intensity within the pixel interval calculated at other solution concentrations; the signal processing system 300 calculates the ratio between the standard value and the reference value to obtain calibration data, and calibrates the original Raman measurement spectrum at the corresponding solution concentration based on the calibration data.

[0098] Specifically, Figure 2 As shown, it shows a schematic diagram of the uncalibrated original Raman measurement spectrum. In the embodiment of the present application, a laser with the same excitation power is selected to excite the Raman spectrum of solutions containing different substance concentrations, and the same integration time and number of integration times are selected during the excitation process. The lens or window material on the light path generated by the excitation of the solution Raman spectrum is used as a calibration object, and the spectral integral area of ​​the Raman characteristic peak of the calibration object in the preset pixel interval is calculated, and the characteristic peak integral value of the calibration object is used as the standard value. According to the standard value, the integral value of the preset pixel interval in the Raman spectrum under other solution concentrations is compared, and the signal intensity at other positions of the spectrum is calibrated using the ratio as the calibration data, and finally the calibrated Raman spectrum is obtained. The calibrated spectral curve is shown in Figure 3 As shown. Figure 3 , the present application can achieve a high-sensitivity solution concentration detection of 35ppm.

[0099] It should be noted that the Raman scattered light intensity is expressed as the integrated area of ​​the spectrum within a pixel interval in the Raman spectrum. In the self-calibration and measurement method provided in the embodiment of the present application, the Raman scattered light intensity at the characteristic peak position of the calibration object is used as a reference for calibration. This calibration method is different from the method of using the signal intensity of a single pixel point on the detector as a reference for calibration in the traditional calibration method. It can avoid the problem of calibration error caused by signal fluctuations caused by noise in the signal intensity under a single pixel, which is conducive to deducting the deviation caused by system performance fluctuations, making the calibration more accurate, and improving the spectral stability and repeatability of the calibration.

[0100] In order to further illustrate the detector provided in the embodiment of the present application, the present application also conducts multiple tests on the substances in the stable solution, that is, multiple tests are conducted on the solution with the same concentration at different times. Figure 4 and Figure 5 It can be seen that the Raman spectrum calibrated by the detector can show very good consistency at low solution concentration (<100 mg / L), which proves that the stability and repeatability of the Raman test spectrum calibrated by this method have been greatly improved.

[0101] It should be noted that in the self-calibration measurement method provided in the embodiment of the present application, the standard spectrum and the original Raman measurement spectrum can be smoothed in advance, but neither the standard spectrum nor the original Raman measurement spectrum is baseline corrected, nor is a differential calculation performed, and only the integral area of ​​the spectrum within the pixel interval is calibrated. In this way, the influence of the processing during spectrum calibration on the information contained in the original Raman spectrum can be avoided, ensuring the detection accuracy of the subsequent concentration prediction model.

[0102] Second,

[0103] The embodiment of the present application provides a detector that can calculate the concentration of the boric acid solution based on a trained concentration measurement model combined with the spectral sampling results of a Raman spectrometer, so as to improve the detection accuracy of the boric acid solution concentration and reduce the detection error, which is helpful for carrying out quantitative detection of the boric acid solution concentration in practical applications.

[0104] like Figure 6 As shown, the detector provided by the embodiment of the present invention includes:

[0105] The sample extraction unit 31 is used to obtain a first Raman spectrum formed by the interaction between a boric acid solution of a first preset concentration and the laser beam.

[0106] The excitation light generated by the light source will be scattered when it is irradiated on the boric acid solution. In addition to the elastic component (Rayleigh scattering) with the same frequency as the excitation light, the scattered light also has components with lower and higher frequencies than the excitation light. The latter phenomenon is collectively called the Raman effect. The inelastic scattering caused by the interaction between the molecular vibration, optical phonons in the solid and the excitation light is called Raman scattering. Generally, the spectrum formed by combining Rayleigh scattering and Raman scattering is called Raman spectrum.

[0107] In the examples of this application, boric acid solutions with concentrations of 0 mg / L, 35 mg / L, 60 mg / L, 113 mg / L, 282 mg / L, 517 mg / L, 800 mg / L, 1656 mg / L, 2660 mg / L, and 3990 mg / L are used as training samples to interact with the excitation light to generate Raman spectra of different concentrations. It should be noted that the boric acid solutions with the above concentrations are only examples, and do not mean that only boric acid solutions with the above concentration ranges can be selected as training samples.

[0108] In the first modeling process, any of the above-mentioned boric acid solutions of concentration is selected as a training sample, the Raman spectrum generated by the interaction between the training sample and the excitation light is used as the first Raman spectrum, and the first Raman spectrum is used in the subsequent modeling process.

[0109] Exemplarily, the embodiments of the present invention are mainly for the detection of the concentration of boric acid solution, which is specifically applied in the field of nuclear power, because boric acid controls the speed of uranium nuclear fission in nuclear power plants and affects the power generation of nuclear power plants. As a soluble poison for regulating the reactivity of nuclear power plants, boric acid is widely used in pressurized water reactor coolants, and the changes in long-term reactivity are controlled by adjusting its concentration. It can be seen that for the above-mentioned application scenarios, the concentration control requirements of the boric acid solution are relatively high, especially for the detection accuracy of the boric acid solution concentration. The existing concentration detection method has a low accuracy, which has a great impact on the production capacity of nuclear power plants.

[0110] The feature extraction unit 32 is used to extract a first interval covering a characteristic peak characterizing boric acid from the first Raman spectrum, and use the pixels included in the first interval and the light intensity values ​​corresponding to the pixels as first spectrum data.

[0111] The first Raman spectrum includes a certain number of Raman peaks, each of which represents the wavelength position and intensity of the corresponding Raman scattered light, and each of which corresponds to a specific molecular bond vibration.

[0112] The embodiment of the present invention selects an interval characterizing a characteristic peak of boric acid, and the interval contains at least one Raman peak. In the prior art, a Raman peak of a single wavelength is generally selected for subsequent modeling. Compared with the prior art, by selecting an interval characterizing a characteristic peak of boric acid for subsequent modeling, it is possible to avoid the influence of the power of the excitation light and other environmental factors that may cause errors in the constructed model.

[0113] Specifically, due to the influence of the fluctuation of the excitation light power and the presence of some impurities in the solution containing specific substances, the model established based on the Raman peak of a single wavelength has a large error in concentration detection. In addition, environmental interference, such as temperature, humidity, electromagnetic field and other factors, may lead to deviations in concentration detection, and due to the above uncertain factors, the model established based on the Raman peak of a single wavelength has poor reproducibility.

[0114] In the example of the present application, the interval characterizing the characteristic peak of boric acid includes a number of pixels, and the number of pixels N is greater than or equal to 450 and less than or equal to 650. This avoids the influence of uncertainty of environmental factors on the accuracy of the model due to the characteristic peak interval being too small, and avoids the existence of too many invalid features due to the characteristic peak interval being too large.

[0115] like Figure 7 The schematic diagram of the Raman spectrum shown in FIG. 1 shows that the horizontal axis represents the pixel point, and the vertical axis represents the light intensity value corresponding to the pixel point. The portion between the two dashed lines is the selected characteristic peak interval, specifically the pixel points with the horizontal axis between 300 and 850.

[0116] It should be noted that the above-mentioned use of the pixel points with a horizontal coordinate between 300 and 850 as the interval for characterizing the characteristic peak of boric acid is only a preferred implementation mode, and in practical applications, the selected interval can be appropriately offset in the horizontal direction.

[0117] The model building unit 33 builds a regression model according to the first spectral data and the first preset concentration and based on the Lasso regression algorithm.

[0118] The Lasso regression algorithm obtains a more refined model by constructing a penalty function, which compresses some coefficients and sets some coefficients to zero.

[0119] In the specific calculation process, the first preset concentration is used as the independent variable, the first spectral data is used as the dependent variable, the penalty coefficient in the Lasso regression algorithm is adjusted to obtain the regression coefficient, and a regression model is established.

[0120] Since the concentration of the boric acid solution is a known value in each modeling process, the regression model calculated by the above method can be verified by a boric acid solution of known concentration to obtain a final accurate regression model.

[0121] Specifically, the first spectral data includes N pixels, each pixel corresponds to its own light intensity value, and an N*2 data matrix is ​​constructed, the data matrix includes 1 row of N pixels and 1 row of N pixels corresponding to the light intensity values, and the data matrix is ​​used as the first spectral data.

[0122] Exemplarily, when the first preset concentration is 35 mg / L, a 35 mg / L boric acid solution forms a Raman spectrum after being irradiated with excitation light, and pixels between 300 and 800 are extracted therefrom to obtain the light intensity value corresponding to each pixel, and a data matrix as shown in Table 1 is established as the first spectral data.

[0123] Table 1:

[0124]

[0125] The concentration detection unit 34 is used to detect the unknown concentration of the boric acid solution to be tested according to the regression model to determine the actual concentration of the boric acid solution to be tested.

[0126] Through the above settings, the regression model of the functional relationship between the concentration of the boric acid solution and the light intensity value of the interval representing the characteristic peak of boric acid in the spectrum diagram established by the Lasso regression algorithm is used to improve the detection accuracy of the concentration of the boric acid solution. In particular, it is applied to the field of nuclear power and can more accurately control the reaction process. In an embodiment of the present invention, the regression model established by the Lasso regression algorithm has a significant improvement in the detection accuracy of the concentration of the boric acid solution. In a boric acid solution of 1000 mg / L, the detection error is greater than or equal to 12 ppm and less than or equal to 18 ppm, thereby improving the production capacity of nuclear power equipment and extending its service life.

[0127] In some embodiments, the sample extraction unit 31 is further used to obtain a second Raman spectrum formed by the interaction between the boric acid solution of a second preset concentration and the laser beam.

[0128] It should be noted that the boric acid solution of the first preset concentration is used to establish a preliminary regression model, and the boric acid solution of the second preset concentration is used to verify the preliminary regression model, and the first preset concentration and the second preset concentration are different.

[0129] Since the principle of generating Raman spectra of training samples with different concentrations under the irradiation of excitation light is the same, it will not be described here in detail.

[0130] The feature extraction unit 32 is further used to extract a second interval covering a characteristic peak characterizing boric acid from the second Raman spectrum, and use the pixel points included in the second interval and the light intensity values ​​corresponding to the pixel points as second spectrum data.

[0131] The second interval selected from the second Raman spectrum has the same number of pixels as the first interval selected from the first Raman spectrum, and the ranges of the characteristic peaks of the two are consistent. From the above introduction to the characteristic peak interval selected from the first Raman spectrum, it can be seen that the pixels with abscissas between 300 and 850 are used as the interval of the second Raman spectrum. In this way, the variables in the modeling process are controlled to avoid the poor reproducibility of the regression model caused by environmental factors or the power of the excitation light.

[0132] It can be understood that the second spectrum data includes N pixels, each pixel corresponds to its own light intensity value, and an N*2 data matrix is ​​constructed, and the data matrix is ​​used as the second spectrum data.

[0133] Exemplarily, when the first preset concentration is 60 mg / L, a 60 mg / L boric acid solution forms a Raman spectrum after being irradiated with excitation light, and pixels between 300 and 800 are extracted therefrom to obtain the light intensity value corresponding to each pixel, and a data matrix as shown in Table 2 is established as the second spectral data.

[0134] Table 2:

[0135]

[0136] The signal processing system 300 further includes a result verification unit, which is used to input the second preset concentration as a variable into the regression model, and compare the calculated data generated by the regression model with the second spectral data to obtain a fit index of the regression model.

[0137] The second preset concentration is a known quantity. The second preset concentration is input into the regression model as an independent variable. The regression model can be used to calculate the calculation data of the training sample of the second preset concentration, wherein the calculation data includes the pixels included in the second interval and the light intensity values ​​corresponding to the pixels.

[0138] Furthermore, the calculated data is compared with the second spectrum data to obtain a goodness of fit index of the regression model, and the goodness of fit index is used to measure the accuracy of the regression model.

[0139] The result verification unit determines whether the fit index is greater than the fit threshold. If the judgment result is yes, the regression model is output. If the judgment result is no, the penalty coefficient in the Lasso regression algorithm is adjusted and the regression model is updated.

[0140] It should be noted that in the process of building a regression model, if the fit index of the regression model is less than the fit threshold, the regression model is updated by adjusting the penalty coefficient so that the fit index of the updated regression model is closer to or exceeds the fit threshold.

[0141] In the example of the present application, the fitness threshold is greater than or equal to 0.99.

[0142] Through the above settings, the detection accuracy of the boric acid solution concentration can be improved, and the regression model established by the Lasso regression algorithm has smaller error and higher reproducibility.

[0143] On the basis of the above-mentioned updated regression model, a boric acid solution of a third preset concentration can also be selected as a training sample, and interacts with the excitation light to form a third Raman spectrum, a third interval covering the characteristic peak of boric acid is selected in the third Raman spectrum, and the pixels contained in the third interval and the light intensity values ​​corresponding to the pixels are used as third spectral data, and the third preset concentration is input as a variable into the updated regression model, and the calculated data generated by the updated regression model is compared with the third spectral data to obtain the fit index of the updated regression model, and to determine whether the fit index of the updated regression model is greater than the fit threshold, if the judgment result is yes, then the updated regression model is output as the final regression model, and if the judgment result is no, then the above steps S301 to S306 are repeated. In repeating the above steps S301 to S306, boric acid solutions of different concentrations can be selected as training samples to improve the accuracy of the final regression model, so that the final regression model has high reproducibility.

[0144] like Figure 8 As shown, the present application also provides an experimental data diagram of obtaining a higher fit index by adjusting the penalty coefficient during the establishment and update process of the regression model. It can be seen that in the process of obtaining a more accurate model, since there is a nonlinear relationship between the penalty coefficient and the fit index, the accuracy of the regression model obtained by the above method is higher.

[0145] The detector provided in an embodiment of the present invention utilizes a Lasso regression algorithm and establishes a regression model based on first spectral data and a first preset concentration. The first spectral data is obtained as the first input data, the first preset concentration is obtained as the second input data, and a penalty coefficient within a set range is selected as the third input data. The first input data, the second input data, and the third input data are processed based on the Lasso regression algorithm, and the regression coefficient corresponding to when the objective function of the regression model reaches the minimum value is obtained through training.

[0146] In some embodiments, the sample extraction unit 31 is used to extract the number of pixels in the first spectral data or the second spectral data. The sample extraction unit 31 is also used to store the first preset concentration and the second preset concentration, and extract the number of concentrations of boric acid solutions with different preset concentrations.

[0147] The model building unit 33 generates an updated regression model according to the number of pixels, the number of concentrations and the updated penalty coefficient. The objective function of the updated regression model satisfies the following relationship:

[0148] ;

[0149] Wherein, X represents the first input data, y represents the second input data, α represents the penalty coefficient, β represents the regression coefficient, N represents the number of concentrations of training samples of different concentrations in the second input data, and P represents the number of pixels contained in the characteristic peak interval.

[0150] Or, the new objective function satisfies the following relationship:

[0151]

[0152] Wherein, X represents the first input data, y represents the second input data, α represents the penalty coefficient, and β represents the regression coefficient.

[0153] In the example of the present application, the penalty coefficient is greater than or equal to 0 and less than or equal to 300. Any value in the range of 0 to 300 is selected as the penalty coefficient α of the Lasso regression algorithm, and the regression coefficient β of the regression model that meets the fitting threshold is obtained by the Lasso regression algorithm according to the training samples of different concentrations, and the penalty coefficient α is saved, wherein the penalty coefficient is an integer.

[0154] In other examples, any value in the range of 0 to 300 is selected as the penalty coefficient α of the Lasso regression algorithm, and the regression coefficient β of the regression model that meets the fitting threshold is obtained according to different test conditions through the Lasso regression algorithm, and the penalty coefficient α is saved, wherein the different test conditions include at least one of the following test conditions:

[0155] The power of the excitation light, temperature, humidity, electromagnetic field, etc.

[0156] Optionally, different concentrations and different test conditions can be taken as a variable and incorporated into the calculation process of the Lasso regression algorithm to establish a more accurate regression model and retain the penalty coefficient α of the regression model.

[0157] In the example of the present application, the penalty coefficient selected for the training sample of the first preset concentration is different from the penalty coefficient selected for the training sample of the second preset concentration. When the regression model is updated by training samples of different preset concentrations, the penalty coefficient is changed to obtain the regression coefficient corresponding to when the objective function reaches the minimum value.

[0158] like Fig. 9As shown, an embodiment of the present invention further provides an error comparison diagram of a regression model calculated based on a Lasso regression algorithm and an elastic network (ElaticNet), wherein the horizontal axis represents a boric acid solution of known concentration, and the regression models obtained by the above two regression algorithms respectively input the boric acid solution of known concentration as an independent variable into their respective regression models, and their respective calculation data are obtained by calculating their respective regression models, and their respective calculations include a number of pixel points characterizing boric acid and light intensity values ​​corresponding to the pixel points, and the light intensity values ​​obtained by the above calculations are compared with the light intensity values ​​of the corresponding pixel points in the spectrum diagram to obtain the prediction error represented by the vertical axis.

[0159] When the prediction error value is closer to 0, it means that the accuracy of the regression model is higher. Fig. 9 It can be seen that the prediction error of the regression model based on the Lasso regression algorithm varies less, and is closer to 0 than the prediction error of the regression model based on the elastic network.

[0160] In addition, compared with other regression algorithms, such as ridge regression algorithm, linear regression, polynomial regression, etc., the Lasso regression algorithm applied to the detection of boric acid solution concentration has higher detection accuracy, and its regression model has high reproducibility.

[0161] Since the Lasso regression algorithm is generally used to establish classification models, it has better performance than other regression algorithms in the field of concentration detection. Especially in the special scenario of boric acid solution concentration detection in the nuclear power field, a regression algorithm with higher accuracy and better reproducibility is needed.

[0162] As an implementation manner, both the first Raman spectrum and the second Raman spectrum are Raman spectra that have not been subjected to background subtraction processing.

[0163] Optionally, the first Raman spectrum and the second Raman spectrum may be processed to eliminate errors caused by environmental interference and noise, wherein the data processing includes at least one of the following processing methods:

[0164] Smoothing and noise reduction.

[0165] It should be noted that since the Raman spectrum signal itself belongs to the detection of weak signals, it is affected by factors such as the test method, optical system, detection instrument, and detection environment. Since the information contained in the Raman spectrum is relatively complex, especially the original Raman spectrum contains the most abundant information, simple mathematical processing is very likely to lose some physical information, especially during the test process, the slightest change in the environment will cause errors in the regression model. Therefore, the Raman spectrum that has not been processed by background subtraction is used as a training object, and the interval representing the characteristic peak of boric acid is selected from it, which can retain the original sample data in the machine learning process as much as possible. Thereby improving the detection accuracy of the obtained regression model.

[0166] Specific as Fig.10 As shown in the figure, it shows the model error diagram obtained after training with spectral samples without baseline subtraction and after baseline subtraction, where the horizontal axis represents the predicted sample concentration, and the vertical axis represents the mean value and error range of the sample concentration predicted by the training model. The center point on the line represents the predicted mean concentration, the upper and lower error bars represent the error range of the predicted concentration, and the spacing size represents the size of the error. Fig.10 It can be seen that under low concentration conditions (the concentration of the boric acid solution to be tested is less than 700 mg / L), the predicted concentration error of the model obtained after training with the spectral samples after baseline subtraction is significantly greater than that without baseline subtraction. In the embodiment of the present application, during the process of spectral self-calibration, the spectral data of the spectral training samples and the test samples are not subjected to baseline subtraction processing, thereby reducing the predicted concentration error of the regression model under low concentration conditions and improving the detection accuracy of the regression model.

[0167] In summary, the embodiments of the present invention provide a regression model established based on a method for detecting the solubility of boric acid solution by Raman spectroscopy, which is used to detect the concentration of boric acid solution, improve the detection accuracy of the concentration of boric acid solution, reduce the detection error, and the regression model has high reproducibility when using boric acid solutions of different concentrations as calculation conditions.

[0168] like Fig.11 As shown, an embodiment of the present application provides a detector, which is based on a Raman spectroscopy detection system. The detector applies a concentration measurement model migration method for quantitative detection based on Raman spectroscopy (hereinafter referred to as the concentration measurement model migration method) provided in an embodiment of the present application, so that the concentration measurement model established based on any detector can be migrated and used between different detectors, avoiding the influence of Raman instrument module performance, optical structure and other factors causing quantitative detection errors. Among them, the signal processing system 300 of the detector includes a migration model construction unit 35, a focus position inversion unit 36 ​​and a concentration measurement model construction unit 37.

[0169] The migration model construction unit 35 is used to obtain the first model and the second model trained in the first pressurized water reactor neutronless radiation source boric acid concentration detector, the first model is the functional relationship between the Raman characteristic spectrum intensity of the target substance at a preset concentration under standard conditions and the focal position of the first pressurized water reactor neutronless radiation source boric acid concentration detector, and the second model is the functional relationship between the Raman characteristic spectrum intensity of the calibration object under standard conditions and the focal position of the first pressurized water reactor neutronless radiation source boric acid concentration detector;

[0170] The focus position inversion unit 36 ​​inverts and obtains the focus position of the second pressurized water reactor neutronless radiation source boric acid concentration detector based on the second model and the Raman characteristic spectrum intensity of the calibration object collected in the second pressurized water reactor neutronless radiation source boric acid concentration detector under standard conditions;

[0171] The focus position inversion unit 36 ​​is used to obtain the Raman characteristic spectrum intensity correction coefficient according to the focus position of the second pressurized water reactor neutronless radiation source boric acid concentration detector in combination with the first model, and combine the Raman characteristic spectrum intensity correction coefficient with the first concentration measurement model to obtain the second concentration measurement model, wherein the first concentration measurement model is obtained by training the first pressurized water reactor neutronless radiation source boric acid concentration detector under standard conditions, and the second pressurized water reactor neutronless radiation source boric acid concentration detector calculates the concentration of the target substance based on the second concentration measurement model.

[0172] Specifically, the concentration measurement model migration method includes the following steps:

[0173] Step S11, training a first concentration measurement model under standard conditions in a first pressurized water reactor neutronless radiation source boric acid concentration detector (hereinafter referred to as the first detector);

[0174] Step S12, training a first model and a second model in the first detector, wherein the first model is a functional relationship between the Raman characteristic spectrum intensity of the target substance at a preset concentration under standard conditions and the focal position of the first detector, and the second model is a functional relationship between the Raman characteristic spectrum intensity of the calibration object under standard conditions and the focal position of the first detector;

[0175] Step S13, under standard conditions, collecting the Raman characteristic spectrum intensity of the target substance of preset concentration and the Raman characteristic spectrum intensity of the calibration object in a second pressurized water reactor neutronless radiation source boric acid concentration detector (hereinafter referred to as the second detector);

[0176] Step S14, based on the Raman characteristic spectrum intensity of the calibration object in the second detector and the second model, invert and obtain the focal position of the second detector;

[0177] Step S15, obtaining a Raman characteristic spectrum intensity correction coefficient according to the focus position of the second detector in combination with the first model;

[0178] Step S16: Combining the Raman characteristic spectrum intensity correction coefficient with the first concentration measurement model to obtain a second concentration measurement model, and the second detector calculates the concentration of the target substance based on the second concentration measurement model.

[0179] According to the above description, the concentration measurement model migration method provided in the embodiment of the present application improves the accuracy and universality of the concentration measurement model trained in the first detector by selecting a calibration object associated with a target substance of a preset concentration for calibration under standard conditions. Then, the first concentration measurement model, the first model, and the second model obtained by training in the first detector are migrated and applied to the second detector to obtain the second concentration measurement model, thereby calculating the concentration of the target substance, so that the concentration measurement model established based on the first detector can be migrated and used between different Raman systems, thereby improving the effectiveness and consistency of the concentration measurement model applied on different Raman systems. The concentration measurement model migration method of the embodiment of the present application enables the concentration measurement model established in a certain Raman system to be migrated and used between different Raman systems, effectively reducing the impact of differences between Raman systems on concentration measurement results, and improving the measurement accuracy of Raman spectroscopy in quantitative analysis.

[0180] It should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0181] It is easy to understand that the first detector and the second detector of the embodiment of the present application refer to two detection systems based on Raman spectrometers. There are inevitably systematic errors (such as focal position) between the two Raman spectrometer systems. The concentration measurement model on one Raman spectrometer system may not be compatible with the other Raman spectrometer system. Therefore, it is necessary to provide a migration method to migrate the concentration measurement model trained on one Raman spectrometer system to other Raman spectrometer systems. The embodiment of the present application is explained by taking the migration of the first concentration measurement model trained in the first detector to the second detector as an example.

[0182] It should be noted that the standard conditions in the embodiments of the present application mean that the quantitative detection of Raman spectroscopy needs to be carried out under pre-set standard conditions, such as pre-setting the temperature to T degrees Celsius, the pipeline pressure to G kilopascals, the flow rate to F, etc. as standard conditions for quantitative detection of Raman spectroscopy. In actual applications, the standard conditions can be set according to specific needs.

[0183] In order to more clearly illustrate the application of the concentration measurement model migration method of the embodiment of the present application, Fig.12The figure shows a schematic diagram of the focal position of the detector in an application scenario of the present application. Specifically, under standard conditions, when the test solution in the sample pipe is monitored by the detector to obtain the Raman spectrum, a flange window is opened on the sample pipe, and the test solution is located in the sample pipe. The light source emits a light beam, which passes through a number of optical devices and is focused on the test solution through a focusing lens, and Raman scattered light is excited at the test solution. The Raman scattered light is finally received by the Raman probe in the detector, thereby obtaining the Raman spectrum.

[0184] As an optional implementation method, the embodiment of the present application sets the calibration object in the optical path of the detector, and the calibration object is not in the solution to be tested; as the focal position of the detector changes, the distance between the focus of the detector and the liquid surface of the solution to be tested changes, and the distance between the focus of the detector and the calibration object changes.

[0185] Specifically, the solution to be tested includes the target substance, and the calibration object is placed at an appropriate position in the optical path. For example, in the embodiment of the present application, a flange window is specially selected as the calibration object. For example, the material used for the window is one of quartz, sapphire, aluminum oxynitride, zinc sulfide, zinc selenide, and calcium fluoride.

[0186] It is easy to understand that the obtained Raman spectrum includes the Raman characteristic spectrum intensity of the target substance and the Raman characteristic spectrum intensity of the calibration object. The Raman spectrum obtained by the first detector includes: the Raman characteristic spectrum intensity of the target substance at a preset concentration detected by the first detector under standard conditions and the Raman characteristic spectrum intensity of the calibration object. The Raman spectrum obtained by the second detector includes: the Raman characteristic spectrum intensity of the target substance to be tested and the Raman characteristic spectrum intensity of the calibration object detected by the second detector under standard conditions.

[0187] By selecting an object other than the solution to be tested as a calibration object, the influence of the introduction of the calibration object on the solution to be tested is avoided, and the calibration object is not affected by factors such as the concentration and flow rate of the solution to be tested. Therefore, the Raman characteristic spectrum intensity of the calibration object can be avoided from being affected by environmental factors, laser power changes, system errors and other factors. The calibration object can be used to calibrate the consistency and accuracy of the Raman spectra obtained at different focal positions, thereby making the quantitative concentration detection of the solution to be tested using the detector more accurate and reliable.

[0188] Moreover, when the focal position of the detector changes, the distance between the focus of the detector and the liquid surface of the solution to be tested and the distance between the focus of the detector and the calibration object change. This change in distance will affect the Raman characteristic spectrum intensity. In this way, the Raman characteristic spectrum intensities of the solution to be tested and the calibration object at a series of different focal positions can be obtained. These data facilitate machine learning training to obtain the first model and the second model.

[0189] As an optional implementation method, the concentration measurement model migration method of the embodiment of the present application also includes: selecting a device capable of stimulating Raman light in the detector as a calibration object; or inserting a device capable of stimulating Raman light between the detector and the solution to be tested, and using the device as a calibration object.

[0190] Specifically, in the process of the light source passing through the optical path to the solution to be tested, there may be some devices that can also excite Raman light. In the relevant technology, the Raman light excited by these devices is generally regarded as stray light, and will not be considered for application during the test process. Even in the design and construction of the detector, optical devices with Raman activity will be avoided as much as possible to avoid interference with the Raman signal of the substance to be tested. Therefore, in this embodiment, the original optical device with Raman activity in the system when the Raman spectrum of the solution is excited is specially selected as a calibration object. The spectral signal of the calibration object is usually regarded as a stray light signal, but it will not be affected by factors such as solution concentration and flow rate, so that the spectrum calibration can be more accurate.

[0191] In addition, the original optical device in the detector is used as a calibration object, and the structural parameters and optical parameters of the calibration object are fixed, which is more convenient for calibrating the Raman spectrum. In addition, this method does not require the introduction of additional optical devices, so there is no need to consider the installation of the device when introducing additional calibration objects. This solution is low in cost and easy to operate.

[0192] It is easy to understand that in addition to using the original optical device in the detector as a calibration object, a substance with Raman activity can be inserted between the detector and the solution to be tested as a calibration object. In this way, the accurate calibration of the Raman spectrum can also be achieved.

[0193] More specifically, Fig.13 As shown, it is a schematic diagram of the application of the first detector provided in the embodiment of the present application to migrate the concentration measurement model to the second detector. In the first detector, the Raman spectrum acquired by the first detector includes the Raman characteristic spectrum intensity of the target substance with a preset concentration and the Raman characteristic spectrum intensity of the calibration object. The functional relationship between the Raman characteristic spectrum intensity of the target substance with a preset concentration and the focal position of the first detector under standard conditions is used as the first model, and the functional relationship between the Raman characteristic spectrum intensity of the calibration object and the focal position of the first detector under standard conditions is used as the second model.

[0194] As an optional implementation, Fig.14 As shown, step S12 of the concentration measurement model migration method in the embodiment of the present application includes:

[0195] Step S121, under standard conditions, when the focus of the first detector is located at the first focus position, obtaining the Raman characteristic spectrum intensity of the target substance of preset concentration detected by the first detector as the Raman characteristic spectrum intensity of the standard target substance;

[0196] Step S122, under standard conditions, adjusting the focus position in the first detector to obtain the Raman characteristic spectrum intensity of the target substance of preset concentration detected by the first detector under different focus positions;

[0197] Step S123, using a regression algorithm to perform machine training, obtain a first functional relationship between the ratio of the Raman characteristic spectrum intensity of the target substance of preset concentration detected by the first detector under different focal positions and the Raman characteristic spectrum intensity of the standard target substance and the focal position, and use the first functional relationship as the first model.

[0198] To explain the process of obtaining the first model in more detail, Fig.15 As shown, a schematic diagram of training a first model on a first detector provided by an embodiment of the present application. Under standard conditions, the focus of the first detector is placed at a first focus position, the Raman characteristic spectrum intensity of a target substance of a preset concentration is detected and recorded, and the Raman characteristic spectrum intensity of the target substance of the preset concentration of the first detector is marked as the Raman characteristic spectrum intensity of the standard target substance, which is the benchmark for subsequent acquisition of the first model. Under the same standard conditions, the focus position of the first detector is adjusted so that the focus is at different positions. At each focus position, the Raman characteristic spectrum intensity of the target substance of the preset concentration is detected and recorded.

[0199] Then, the collected data is machine trained using a regression algorithm to obtain a first functional relationship, which can describe the relationship between the ratio between the Raman characteristic spectrum intensity of the target substance of the preset concentration and the Raman characteristic spectrum intensity of the standard target substance by the first detector at different focal positions and the focal position. The first functional relationship, i.e., the first model, can be used to preset or correct the Raman characteristic spectrum intensity at different focal positions, and to infer the Raman characteristic spectrum intensity correction coefficient.

[0200] As an optional implementation, the first model is expressed by the following formula:

[0201] Formula (1);

[0202] In the formula, I A0 It represents the intensity of the Raman characteristic spectrum detected by the first detector for the target substance of preset concentration under standard conditions and when the focus of the first detector is located at the first focus position, I An Indicates that under standard conditions, the focus of the first detector is at then In the case of the first detector detecting the target substance at the preset concentration, the Raman characteristic spectrum intensity is: f n Indicates that the focus of the first detector is at n At each location, F 1(*) indicates the first functional relationship.

[0203] As an optional implementation, Fig.16 As shown, step S12 of the concentration measurement model migration method in the embodiment of the present application also includes:

[0204] Step S124, under the standard condition, when the focus of the first detector is located at the first focus position, obtaining the Raman characteristic spectrum intensity of the calibration object detected by the first detector as the Raman characteristic spectrum intensity of the standard calibration object;

[0205] Step S125, under the standard condition, adjusting the focus position in the first detector to obtain the Raman characteristic spectrum intensity detected by the first detector on the calibration object under different focus positions;

[0206] Step S126, using a regression algorithm to perform machine training, obtain a second functional relationship between the ratio of the Raman characteristic spectrum intensity detected by the first detector on the calibration object and the Raman characteristic spectrum intensity of the standard calibration object under different focal positions and the focal position, and use the second functional relationship as the second model.

[0207] To explain the process of obtaining the second model in more detail, Fig.17 As shown, a schematic diagram of training the second model on the first detector provided by an embodiment of the present application. Under standard conditions, the focus of the first detector is placed at the first focal position, the Raman characteristic spectrum intensity of the calibration object is detected and recorded, and the Raman characteristic spectrum intensity of the calibration object of the first detector is marked as the Raman characteristic spectrum intensity of the standard calibration object, which is the basis for the subsequent acquisition of the second model. Under the same standard conditions, the focus position of the first detector is adjusted so that the focus is at different positions. At each focal position, the Raman characteristic spectrum intensity of the calibration object is detected and recorded.

[0208] Then, the collected data is machine trained using a regression algorithm to obtain a second functional relationship, which can describe the relationship between the ratio between the Raman characteristic spectrum intensity of the calibration object and the Raman characteristic spectrum intensity of the standard calibration object of the first detector at different focal positions and the focal position. The second functional relationship is the second model, which can be used to combine the Raman characteristic spectrum intensity of the calibration object of the second detector to be migrated and applied to invert and obtain the focal position of the second detector.

[0209] As an optional implementation, the second model is expressed by the following formula:

[0210] Formula (2);

[0211] In the formula, I B0 It means that under the standard condition, the focus of the first detector is located at the first focus position, and the Raman characteristic spectrum intensity detected by the first detector on the calibration object is I Bn Indicates that under standard conditions, the focus of the first detector is at the n In the case of the first detector detecting the Raman characteristic spectrum intensity of the calibration object, f n Indicates that the focus of the first detector is at n At each location, F 2(*) indicates the second functional relationship.

[0212] As an optional implementation method, any one of the following regression algorithms is used for machine training to obtain the first model and the second model: multivariate linear regression, polynomial regression, Lasso regression, ridge regression, elastic network regression, Bayesian ridge regression, Huber regression.

[0213] The regression algorithm is a supervised learning algorithm. The regression algorithm describes the relationship between the input features and the target variable by establishing a model. This relationship can be represented by a function that maps the input features to the target variable. For example, in the first model, the input features can be the Raman characteristic spectrum intensity and focal position of the target substance, and the target variable can be the concentration of the target substance. In the second model, the input features can be the Raman characteristic spectrum intensity and focal position of the calibration object, and the target variable can be the concentration of the target substance.

[0214] Specifically, multivariate linear regression can predict the linear relationship between input features and target variables; polynomial regression is an extension of multivariate linear regression, and polynomial regression can be used to capture more complex relationships in Raman spectroscopy, such as obtaining nonlinear relationships by introducing high-order terms of input features; Lasso regression and ridge regression are variants of linear regression, and Lasso regression can help identify the most important Raman spectral features for concentration prediction, and ridge regression can improve the stability of the trained model when there is a strong correlation between input features; elastic network regression is a hybrid of Lasso regression and ridge regression, and elastic network regression is very useful when the number of input features is greater than the number of samples or there are multiple collinearities between input features; Bayesian ridge regression can provide a prior distribution of regression coefficients and perform parameter estimation by maximizing the marginal log-likelihood; Huber regression is a regression algorithm that is resistant to outliers, which modifies the loss function so that square loss is used when the prediction error is small, and linear loss is used when the prediction error is large.

[0215] In practical applications, multiple regression algorithms can be tested multiple times, and the performance of the obtained models can be evaluated through methods such as cross-validation. Finally, one or two or more regression algorithms with the best effect can be selected for machine training to obtain the first model and the second model.

[0216] As an optional implementation, according to the obtained focal position of the second detector, combined with the first model, the obtained Raman characteristic spectrum intensity correction coefficient is expressed by the following formula:

[0217] Formula (3);

[0218] In the formula, F 1(*) indicates the first functional relationship, f’ Indicates the focus position of the second detector.

[0219] After the focus position of the second detector is obtained, the first model can be used to estimate the change in Raman characteristic spectrum intensity caused by the different focus positions. The first model is established during the training process, and it associates the relationship between the Raman characteristic spectrum intensity of the target substance of the preset concentration and the focus position.

[0220] By inputting the focal position of the second detector into the first model, the first model will output a correction coefficient, namely the Raman characteristic spectrum intensity correction coefficient, which is used to adjust the Raman characteristic spectrum intensity measured by the second detector to compensate for the influence of the change in the focal position. The Raman characteristic spectrum intensity correction coefficient reflects the proportional relationship between the Raman characteristic spectrum intensity measured by the second detector and the Raman characteristic spectrum intensity at the focal position.

[0221] As an optional implementation, the first concentration measurement model is expressed by the following formula:

[0222] Formula (4);

[0223] In the formula, Q A represents the regression model under standard conditions in the first detector, I A It represents the intensity of the Raman characteristic spectrum detected by the first detector on the target substance under standard conditions and when the focus of the first detector is located at the first focus position.

[0224] The Raman characteristic spectrum intensity correction coefficient is combined with the first concentration measurement model to obtain the second concentration measurement model represented by the following formula:

[0225] Formula (5);

[0226] In the formula, Q A represents the regression model under standard conditions in the first detector, M represents the correction coefficient of the Raman characteristic spectrum intensity, I’ A It indicates the intensity of the Raman characteristic spectrum detected by the second detector for the target substance under standard conditions.

[0227] The second concentration measurement model can calculate the concentration of the target substance based on the corrected Raman characteristic spectrum intensity measured by the second detector. The first concentration measurement model is trained under standard conditions and associates the relationship between the Raman characteristic spectrum intensity and the target substance concentration. By combining the Raman characteristic spectrum intensity correction coefficient with the first concentration measurement model, it can be ensured that the concentration of the target substance can be accurately measured even on different detectors.

[0228] In summary, the concentration measurement model migration method for quantitative detection based on Raman spectroscopy provided in the embodiment of the present application selects a device other than the solution to be tested as a calibrant for calibration, and sets the calibrant in the optical path of the detector. Under standard conditions, a first concentration measurement model, a first model, and a second model are obtained through machine training in a first detector, and these models are migrated and applied to a second detector to obtain a second concentration measurement model, thereby calculating the concentration of the target substance, so that the concentration measurement model established based on the first detector can be migrated and used between different detectors, thereby improving the accuracy and universality of the concentration measurement model trained in the detector, effectively reducing the impact of differences between detectors on concentration measurement results, and improving the measurement accuracy of Raman spectroscopy in quantitative analysis.

[0229] It should be noted that the detector provided by the present application is aimed at detecting the concentration of boric acid solution in a pressurized water reactor, and can achieve a measurement limit as low as 30 mg / L, which is significantly improved compared with the measurement limit of the boron meter based on a neutron source in the related art (the boron meter based on a neutron source is no longer accurate when the concentration is lower than 100 mg / L), with small measurement error and high reliability. In addition, depending on the different detection objects, the monitoring system based on Raman spectrometer provided by the present application can also be promoted and applied to biopharmaceuticals, physical and chemical analysis, environmental monitoring, agricultural product inspection, etc., and has broad application prospects.

[0230] In one embodiment, the optical path system includes an outer shell, and the Raman probe assembly and the signal sampling unit are both arranged in the outer shell; the pressurized water reactor neutron-free radiation source boric acid concentration detector also includes a flange window and a planar super lens installed on the flange window.

[0231] like Fig.18As shown, the planar superlens 14 is fixedly mounted on the flange window 13, so that the planar superlens 14 isolates the space between the sampling tube component 100 and the outer shell, and the planar superlens 14 acts on the laser beam, so that the laser beam acts on the solution to be tested after being focused to form a Raman scattering signal, and the pressure resistance of the planar superlens 14 is not less than 3MPa.

[0232] As a new type of optical element, the main advantage of superlens in early applications is that it can achieve imaging resolution that exceeds the diffraction limit of traditional optics. Traditional convex lenses are limited by the wavelength of light waves and cannot distinguish two points that are too close, which is the so-called diffraction limit. Superlenses can overcome this limitation, making imaging at the nanoscale possible. Superlenses mainly use the microscopic structural units (meta-surface elements) on the surface to accurately control the phase of the incident light, achieve high-precision control of the light wavefront, and achieve the deflection of light at different positions through the control of the wavefront, ultimately achieving the focusing function.

[0233] According to the focusing principle of the above-mentioned superlens, the superlens can be designed as a planar pressure-resistant structure. It should be noted that the sampling tube components and probe integration device for Raman spectroscopy detection provided by this application are applied to the boric acid concentration detector of the neutron-free radiation source of the pressurized water reactor. The planar superlens has a pressure resistance of not less than 3MPa, and can be in contact with the solution in the high-pressure detection environment of the pressurized water reactor, and serve as a mechanical structural element to achieve solution isolation and pipeline pressure resistance. This application realizes the focusing function through the design of the superlens while taking into account the mechanical capacity of the optical window, thereby integrating the functions of two discrete optical elements in the Raman spectrum real-time detection system of the online solution, thereby improving the collection efficiency and sensitivity of the Raman optical signal.

[0234] like Fig.19 As shown, it shows a comparison of the spectrum collected by the Raman system integrating the super lens and the flange window in the embodiment of the present application and the spectrum collected by the traditional Raman system. Fig.19 It can be seen that at wave number 400cm -1 and 650cm -1 The peak intensity measured by the Raman system with the integrated superlens and flange window provided by the present application is higher than that of the traditional Raman system. The Raman system with the integrated superlens and flange window provided by the present application can reduce the loss of the laser beam when it passes through the optical element during transmission, thereby improving the stability and accuracy of the Raman scattering signal.

[0235] The above disclosure is only the preferred embodiment of the present application, but it is not intended to limit the scope of rights of the present application. A person of ordinary skill in the art can understand that without departing from the spirit and scope of the present application and the appended claims, changes, modifications, substitutions, combinations, and simplifications should all be equivalent replacement methods and still fall within the scope of the invention.

Claims

1. A pressurized water reactor neutron-free radiation source boric acid concentration detector, characterized in that: include: A sampling tube component, used for holding the solution to be tested; An optical system connected to the sampling tube component, the optical system comprising a Raman probe assembly and a signal sampling unit, the Raman probe assembly is used to emit a laser beam to the solution to be tested, and receive a Raman scattering signal generated when the laser beam acts on the solution to be tested, and the signal sampling unit receives the Raman scattering signal through the Raman probe assembly and forms a Raman spectrum; A signal processing system connected to the optical path system, and acquiring a data signal sent by the signal sampling unit that characterizes the Raman spectrum, and processing the data signal to determine the boric acid concentration in the solution to be tested; Wherein, a calibration object is provided in the optical path system; The signal processing system comprises: A migration model construction unit, the migration model construction unit is used to obtain a first model and a second model trained in a first pressurized water reactor neutronless radiation source boric acid concentration detector, the first model is a functional relationship between the Raman characteristic spectrum intensity of a target substance at a preset concentration under standard conditions and the focal position of the first pressurized water reactor neutronless radiation source boric acid concentration detector, and the second model is a functional relationship between the Raman characteristic spectrum intensity of a calibration object under standard conditions and the focal position of the first pressurized water reactor neutronless radiation source boric acid concentration detector; A focus position inversion unit, which inverts and obtains the focus position of the second pressurized water reactor neutronless radiation source boric acid concentration detector based on the second model and the Raman characteristic spectrum intensity of the calibration object collected from the second pressurized water reactor neutronless radiation source boric acid concentration detector under standard conditions; A concentration measurement model building unit is used to obtain a Raman characteristic spectrum intensity correction coefficient according to the focal position of the second pressurized water reactor neutronless radiation source boric acid concentration detector in combination with the first model, and combine the Raman characteristic spectrum intensity correction coefficient with the first concentration measurement model to obtain a second concentration measurement model, wherein the first concentration measurement model is obtained by training the first pressurized water reactor neutronless radiation source boric acid concentration detector under standard conditions, and the second pressurized water reactor neutronless radiation source boric acid concentration detector calculates the concentration of the target substance based on the second concentration measurement model.

2. The pressurized water reactor neutron-free radiation source boric acid concentration detector according to claim 1, characterized in that: The signal processing system is used to set a pixel interval, and the characteristic peak of the standard spectrum of the calibration object is located in the pixel interval; the signal processing system is also used to obtain a standard value and a reference value, wherein the standard value is the Raman scattering light intensity in the pixel interval calculated at any solution concentration, and the reference value is the Raman scattering light intensity in the pixel interval calculated at other solution concentrations; The signal processing system calculates a ratio between the standard value and the reference value to obtain calibration data, and calibrates the original Raman measurement spectrum at the corresponding solution concentration based on the calibration data.

3. The PWR neutronless radiation source boric acid concentration detector according to claim 2, characterized in that: The standard spectrum and the original Raman measurement spectrum are both spectra without baseline correction.

4. The PWR neutronless radiation source boric acid concentration detector according to claim 1, characterized in that: The signal processing system comprises: A sample extraction unit, used for obtaining a first Raman spectrum formed by the interaction between a boric acid solution of a first preset concentration and the laser beam; A feature extraction unit, configured to extract a first interval covering a characteristic peak characterizing boric acid from the first Raman spectrum, and use the pixel points included in the first interval and the light intensity values ​​corresponding to the pixel points as first spectrum data; A model building unit, wherein the model building unit builds a regression model based on the first spectral data and the first preset concentration and based on a Lasso regression algorithm; The concentration detection unit is used to detect the actual concentration of the boric acid solution of unknown concentration according to the regression model.

5. The pressurized water reactor neutron-free radiation source boric acid concentration detector according to claim 4, characterized in that: The sample extraction unit is further used to obtain a second Raman spectrum formed by the interaction between the boric acid solution of a second preset concentration and the laser beam; the feature extraction unit is further used to extract a second interval covering a characteristic peak of boric acid in the second Raman spectrum, and use the pixel points included in the second interval and the light intensity values ​​corresponding to the pixel points as second spectrum data; The signal processing system also includes a result verification unit, which is used to obtain the second spectral data and input the second spectral data into the regression model. The result verification unit compares the calculated data generated by the regression model with the second spectral data to obtain a fit index of the regression model and determines whether the fit index is greater than a fit threshold. The model building unit outputs the regression model when the fit index is greater than the fit threshold, and adjusts the penalty coefficient in the Lasso regression algorithm when the fit index is less than or equal to the fit threshold to generate an updated regression model.

6. The pressurized water reactor neutron-free radiation source boric acid concentration detector according to claim 5, characterized in that: The sample extraction unit is used to extract the number of the pixel points in the first spectral data or the second spectral data, and the sample extraction unit is also used to store the first preset concentration and the second preset concentration, and extract the number of concentrations of the boric acid solution with different preset concentrations; The model building unit generates the updated regression model according to the number of pixels, the number of concentrations and the updated penalty coefficient, and the objective function of the updated regression model satisfies the following relationship: ； in, X represents the first spectral data and the second spectral data, y represents the first preset concentration and the second preset concentration, α represents the penalty coefficient, β represents the regression coefficient, N represents the number of concentrations, P Indicates the number of pixels.

7. The PWR neutronless radiation source boric acid concentration detector according to claim 5, characterized in that: The first Raman spectrum and the second Raman spectrum are both Raman spectra that have not been processed by background subtraction data.

8. The PWR neutronless radiation source boric acid concentration detector according to claim 1, characterized in that: The first model is expressed by the following formula: ； In the formula, I A0 It indicates that under the standard condition, when the focus of the first pressurized water reactor neutronless radiation source boric acid concentration detector is located at the first focus position, the Raman characteristic spectrum intensity detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector for the target substance of the preset concentration, I An It means that under the standard condition, the focus of the boric acid concentration detector of the first pressurized water reactor neutronless radiation source is at the n In the case of a position, the intensity of the Raman characteristic spectrum detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector on the target substance of the preset concentration, f n Indicates that the focus of the boric acid concentration detector of the first pressurized water reactor neutron-free radiation source is at the n At each location, F 1(*) indicates the first functional relationship; The second model is expressed by the following formula: ； In the formula, I B0 It indicates that under the standard condition, and the focus of the first pressurized water reactor neutronless radiation source boric acid concentration detector is located at the first focus position, the Raman characteristic spectrum intensity detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector on the calibration object, I Bn It means that under the standard condition, the focus of the boric acid concentration detector of the first pressurized water reactor neutronless radiation source is at the n In the case of a position, the intensity of the Raman characteristic spectrum detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector on the calibration object is, f n Indicates that the focus of the boric acid concentration detector of the first pressurized water reactor neutron-free radiation source is at the n At each location, F 2(*) indicates the second functional relationship.

9. The PWR neutronless radiation source boric acid concentration detector according to claim 8, characterized in that: The Raman characteristic spectrum intensity correction coefficient is expressed by the following formula: ； In the formula, F 1(*) represents the first functional relationship, f’ Indicates the focal position of the second pressurized water reactor neutron-free radiation source boric acid concentration detector; The first concentration measurement model is expressed by the following formula: ； In the formula, Q A represents the regression model under standard conditions in the boric acid concentration detector of the first pressurized water reactor neutronless radiation source, I A represents the intensity of the Raman characteristic spectrum detected by the first pressurized water reactor neutronless radiation source boric acid concentration detector on the target substance under the standard condition and when the focus of the first pressurized water reactor neutronless radiation source boric acid concentration detector is located at the first focus position; The second concentration measurement model is expressed by the following formula: ； In the formula, Q A represents the regression model under standard conditions in the boric acid concentration detector of the first pressurized water reactor neutronless radiation source, M represents the Raman characteristic spectrum intensity correction coefficient, I’ A It represents the intensity of the Raman characteristic spectrum detected by the second pressurized water reactor neutronless radiation source boric acid concentration detector on the target substance under the standard condition.

10. The pressurized water reactor neutron-free radiation source boric acid concentration detector according to claim 1, characterized in that: The optical path system comprises an outer shell, and the Raman probe assembly and the signal sampling unit are both arranged in the outer shell; The pressurized water reactor neutronless radiation source boric acid concentration detector also includes a flange window and a planar metalens installed on the flange window, the flange window is arranged between the outer shell and the sampling tube component, the outer shell is connected to the sampling tube component through the flange window, the planar metalens is arranged between the Raman probe assembly and the sampling tube component, the planar metalens acts on the laser beam so that the laser beam is focused and then acts on the solution to be tested to form the Raman scattering signal, and the pressure resistance of the planar metalens is not less than 3MPa.

11. The pressurized water reactor neutron-free radiation source boric acid concentration detector according to claim 10, characterized in that: The surface of the planar superlens has a micro-nano hole structure, so that the phase of the laser beam is changed sequentially from the center point of the planar superlens to the edge thereof.

Images