A parameter optimization method and system for a nuclear power water chemical ultra-trace detection instrument

Abstract

The present application relates to the technical field of nuclear power water chemistry online monitoring, and discloses a nuclear power water chemistry ultra-trace detector parameter optimization method and system, wherein the method comprises the following steps: constructing an initial state vector; performing full-spectrum scanning analysis based on the initial state vector; quantifying interference based on the spectrum analysis result; performing collision reaction pool parameter optimization; performing verification scanning and local fine tuning; and outputting the isotope concentration result of the element to be measured; compared with the prior art, the collision reaction pool parameters are fixed or artificially set to process multi-atomic ion interference, especially under the condition that the interference intensity fluctuates with the change of boron and lithium concentrations with the core burnup, the technical problem that real-time consideration of interference suppression and sensitivity maintenance cannot be achieved. Due to the full-spectrum peak shape analysis and closed-loop verification fine tuning, the present application realizes the adaptive matching of the collision gas helium flow, the pool bias voltage and the change amount of the quadrupole mass selection window, and improves the accuracy and stability of the iron, nickel and chromium ultra-trace detection result.

Description

Technical Field

[0001] This invention relates to the field of online monitoring technology for nuclear power plant water chemistry, and in particular to a method and system for optimizing the parameters of an instrument for ultra-trace detection of nuclear power plant water chemistry. Background Technology

[0002] In the chemical monitoring of the primary coolant in nuclear power plants, ultra-trace analysis of the isotope concentrations of multiple elements such as iron, nickel, and chromium is required to assess the corrosion of structural materials and the migration of radioactive products. Inductively coupled plasma mass spectrometry (ICP-MS) is the mainstream method for this type of ultra-trace detection. However, the primary coolant contains high concentrations of boron and lithium matrix, which can form complex polyatomic interferences, such as argon-oxygen interference and calcium-oxygen interference, in the plasma and collision reaction cell. Furthermore, the boron and lithium concentrations change with burnup, leading to dynamic drift in the interference response relationship.

[0003] Existing technologies typically employ fixed collision reaction cell parameters and simple interference correction equations for detection, which are ill-suited to adapting to dynamic fluctuations in the boron-lithium matrix and the resulting interference variations. When boron concentration is high or lithium concentration changes significantly, fixed parameters either fail to adequately eliminate interference, leading to overestimation of iron and nickel isotope measurements, or excessively suppress the signal, severely reducing the sensitivity for trace elements and failing to meet the stringent requirements for accuracy and long-term consistency in ultra-trace element monitoring of nuclear power plant primary coolant. Therefore, there is an urgent need for a method that can optimize collision reaction cell parameters in real time based on the matrix condition, achieving highly sensitive, low-background detection of ultra-trace elements even under dynamic interference conditions. Summary of the Invention

[0004] To address the aforementioned technical shortcomings, the present invention aims to propose a parameter optimization method for nuclear power plant water chemistry ultra-trace detection instruments. This method addresses the technical problem in existing technologies where fixed or manually set collision reaction cell parameters are used to handle polyatomic ion interference, particularly when interference intensity fluctuates due to changes in boron and lithium concentrations with core burnup. This method cannot simultaneously address interference suppression and sensitivity maintenance in real time.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument.

[0006] The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument includes:

[0007] Step S10: Obtain the boron-lithium prior data of the primary coolant sample, construct an initial state vector based on the boron-lithium prior data using a wide mass range pre-scan and initial disturbance response relationship, and output the initial state vector;

[0008] Step S20: Based on the initial state vector, perform a full-spectrum scanning analysis task using the Gaussian-Lorentz hybrid peak fitting method, and output the spectral analysis results;

[0009] Step S30: Based on the spectral analysis results, the dynamic interference yield model and recursive least squares method are used to perform the interference quantification task, and the interference evaluation data is output.

[0010] Step S40: Based on the interference evaluation data, the Mamdani fuzzy control and sequential quadratic programming algorithm is used to perform the collision reaction pool parameter optimization task, and the optimized collision reaction pool parameters are output.

[0011] Step S50: Perform verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and output the isotope concentration results of the element to be measured.

[0012] Preferably, step S10, which involves acquiring boron-lithium prior data of the primary coolant sample, constructing an initial state vector based on the boron-lithium prior data using a wide-range pre-scan and initial disturbance response relationship, and outputting the initial state vector, specifically includes:

[0013] Step S101: Obtain the prior values ​​of boron concentration and lithium concentration in the primary coolant sample, set the collision gas helium flow rate, cell bias voltage and quadrupole mass selection window change as safe initial values, and perform a pre-scan within the mass-to-charge ratio range of 6 to 70 to obtain the pre-scan spectrum;

[0014] Step S102: Based on the pre-scan spectrum, extract the intensity of lithium hepta-ion, boron unita-ion, argon forty-ion, calcium forty-four-ion, and the apparent signal intensity at mass-to-charge ratios of 52, 56, and 60 to obtain the matrix ion intensity vector.

[0015] Step S103: Construct the initial disturbance response relationship based on the matrix ion intensity vector to obtain the initial state vector. :

[0016]

[0017] in, The strength of lithium hepta-ion, Boron undecyl ion strength, The ionic strength is argon forty. The ionic strength of calcium is 44. The initial response coefficient for argon-oxygen interference is... The initial response coefficient for calcium-oxygen interference is... The initial safe value for the collision gas helium flow rate. This is the initial safe value for the pool bias voltage. Select a safe initial value for the window change of the quadrupole mass.

[0018] Preferably, step S20, which involves performing a full-spectrum scanning analysis task based on the initial state vector using a Gaussian-Lorentz hybrid peak fitting method and outputting the spectral analysis results, specifically includes:

[0019] Step S201: Read the initial safe values ​​of collision gas helium flow rate, pool bias pressure, and quadrupole mass selection window change from the initial state vector, write them into the collision pool controller and quadrupole power supply, and perform a full spectrum scan on the water sample to be tested in the range of mass-to-charge ratio of 40 to 65 to obtain full spectrum scan data.

[0020] Step S202: Based on the full spectrum scanning data, the Gauss-Lorentz mixed peak shape fitting method is used to analyze the argon-oxygen interference peak, calcium-oxygen interference peak, iron 56 isotope peak, nickel 60 isotope peak and chromium 52 isotope peak respectively, and obtain the peak shape feature vector.

[0021] Step S203: Based on the peak shape feature vector, synchronously update the intensity of lithium hepta-ion, boron eleven-ion, argon forty-ion, and calcium forty-four-ion to obtain the updated matrix ion intensity vector, and use the peak shape feature vector and the updated matrix ion intensity vector as the spectral analysis results.

[0022] Preferably, in step S202, the step of analyzing the argon-oxygen interference peak, calcium-oxygen interference peak, iron 56 isotope peak, nickel 60 isotope peak, and chromium 52 isotope peak respectively using the Gauss-Lorentz mixed peak shape fitting method based on the full spectrum scan data to obtain the peak shape feature vector specifically includes:

[0023] Argon-oxygen interference peaks were identified near a mass-to-charge ratio of 55.957, calcium-oxygen interference peaks were identified near a mass-to-charge ratio of 59.953, iron 56 isotope peaks were identified near a mass-to-charge ratio of 55.935, nickel 60 isotope peaks were identified near a mass-to-charge ratio of 59.931, and chromium 52 isotope peaks were identified near a mass-to-charge ratio of 51.941.

[0024] The following Gaussian-Lorentz mixed peak shape function was used to fit each peak:

[0025]

[0026] in, mass-to-charge ratio The fitted signal intensity at that point, Peak intensity, The Lorentz ratio is the coefficient. The Lorentz half-width parameter, Peak position, The standard deviation is Gaussian.

[0027] Based on the fitting results, a peak shape feature vector containing peak position, full width at half maximum (FWHM), asymmetry factor, and peak intensity is obtained.

[0028] Preferably, step S30, which involves performing interference quantization based on the spectral analysis results using a dynamic interference yield model and recursive least squares method to output interference evaluation data, specifically includes:

[0029] Step S301: Read the peak shape feature vector and the updated matrix ion intensity vector from the spectral analysis results, and establish a dynamic interference yield model for argon-oxygen interference:

[0030]

[0031] in, For the first The intensity of the argon-oxygen interference signal in each analysis cycle. , , and The first Time-varying weighting coefficients for each analysis period;

[0032] Step S302: Using the recursive least squares method with a forgetting factor of 0.92, update the time-varying weight coefficients based on the argon-oxygen interference signal intensity and matrix ion intensity vector of the most recent three analysis periods, and calculate the overlap integral ratio based on the peak position and full width at half maximum of the updated argon-oxygen interference peak and the iron 56 isotope peak.

[0033] Step S303: Calculate the equivalent background of interference at mass-to-charge ratio 56 based on the overlap integral ratio, and introduce the sensitivity attenuation rate, using the equivalent background of interference and the sensitivity attenuation rate as interference evaluation data.

[0034] Preferably, step S40, which involves using the Mamdani fuzzy control and sequential quadratic programming algorithm to perform collision reaction pool parameter optimization based on the interference evaluation data and outputting the optimized collision reaction pool parameters, specifically includes:

[0035] Step S401: Read the equivalent background of interference and the sensitivity attenuation rate at mass-to-charge ratio 56 from the interference evaluation data, compare the equivalent background of interference with the preset target interference threshold to obtain the interference background deviation, and input the interference background deviation and the sensitivity attenuation rate into the Mamdani fuzzy controller to obtain the recommended adjustment amount of collision gas helium flow rate, the recommended adjustment amount of pool bias pressure and the recommended adjustment amount of quadrupole mass selection window;

[0036] Step S402: Based on the recommended adjustment of the collision gas helium flow rate, the recommended adjustment of the pool bias voltage, and the recommended adjustment of the quadrupole mass selection window, establish a sensitivity loss model:

[0037]

[0038] in, To the collision gas helium flow rate Pool bias Sensitivity retention amount Initial sensitivity, The sensitivity loss coefficient, It is the bias stability constant;

[0039] Step S403: Construct an optimization objective function based on the sensitivity loss model and the dynamic interference yield model, and solve the optimization objective function using a sequential quadratic programming algorithm to obtain the optimized collision reaction pool parameters.

[0040] Preferably, step S50, which involves performing verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and outputting the isotope concentration result of the element to be measured, specifically includes:

[0041] Step S501: Write the optimized collision reaction cell parameters into the collision cell controller and quadrupole power supply. After the detection instrument stabilizes, perform a verification scan and recalculate the verification interference equivalent background according to the Gauss-Lorentz mixed peak fitting method in step S20.

[0042] Step S502: Compare the verification interference equivalent background with the preset target interference threshold. If the deviation ratio is greater than the preset convergence ratio, then take the optimized collision reaction cell parameters as the starting point and perform local fine-tuning along the negative gradient direction of the sensitivity loss model with respect to the collision gas helium flow rate and the cell bias pressure to obtain the fine-tuned collision reaction cell parameters.

[0043] Step S503: Based on the fine-tuned collision reaction cell parameters, perform the verification scan again until the deviation ratio of the verification interference equivalent background converges to within the preset convergence ratio, and output the isotope concentration result of the element to be measured.

[0044] This invention also provides a parameter optimization system for nuclear power plant water chemistry ultra-trace detection instruments, comprising:

[0045] The initial state construction module is used to acquire boron-lithium prior data of the primary coolant sample, construct an initial state vector based on the boron-lithium prior data using a wide mass range pre-scan and initial disturbance response relationship, and output the initial state vector.

[0046] The spectral scanning and analysis module is used to perform a full-spectrum scanning and analysis task based on the initial state vector using a Gaussian-Lorentz hybrid peak fitting method, and output the spectral analysis results.

[0047] The interference quantization module is used to perform interference quantization tasks based on the spectral analysis results using a dynamic interference yield model and a recursive least squares method, and output interference evaluation data.

[0048] The parameter optimization module is used to perform the collision reaction pool parameter optimization task based on the interference evaluation data using the Mamdani fuzzy control and sequential quadratic programming algorithm, and output the optimized collision reaction pool parameters.

[0049] The verification output module is used to perform verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and output the isotope concentration results of the element to be tested.

[0050] The present invention also provides a parameter optimization device for a nuclear power plant water chemistry ultra-trace detection instrument. The nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device includes: a memory, a processor, and a nuclear power plant water chemistry ultra-trace detection instrument parameter optimization program stored in the memory and executable on the processor. When the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization program is executed by the processor, the above-mentioned method is implemented.

[0051] The present invention also provides a computer program product, the computer program product including a parameter optimization program for a nuclear power plant water chemistry ultra-trace detection instrument, the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization program implementing the above method when executed by a processor.

[0052] The beneficial effects of this invention are as follows:

[0053] 1. This invention acquires prior boron-lithium data and performs a wide-range pre-scan to construct a state vector containing the relationship between matrix ion strength and initial interference response. This provides an accurate initial base point for subsequent interference modeling and parameter optimization, avoiding the tuning blindness caused by the unknown initial state in traditional methods.

[0054] 2. This invention uses Gauss-Lorentz hybrid peak shape fitting to analyze full-spectrum scanning data, which can accurately separate the overlapping regions of argon-oxygen and calcium-oxygen interference peaks with the peaks of the isotopes to be measured. Combined with a dynamic interference yield model and recursive least squares method to quantify interference in real time, it overcomes the problem of inaccuracy of traditional fixed yield models when boron and lithium vary, and significantly improves the real-time accuracy of interference correction.

[0055] 3. This invention optimizes the collision reaction cell parameters through Mamdani fuzzy control and sequential quadratic programming algorithm. Under the constraints of the sensitivity loss model, it balances the requirements of interference elimination and sensitivity maintenance, and achieves rapid convergence by using negative gradient local fine-tuning. This allows the instrument to maintain its optimal working state continuously during long-term operation, improving the stability and data reliability of ultra-trace detection in nuclear power plant water chemistry. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the first embodiment of a method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument according to the present invention. Detailed Implementation

[0057] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0058] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0059] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument according to the present invention. The first embodiment of the method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument according to the present invention is presented.

[0060] In the first embodiment, the method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument includes:

[0061] Step S10: Obtain the boron-lithium prior data of the primary coolant sample, construct an initial state vector based on the boron-lithium prior data using a wide mass range pre-scan and initial disturbance response relationship, and output the initial state vector;

[0062] The boron-lithium prior data mentioned in this step refers to the prior boron and lithium concentration values ​​obtained from the primary loop chemical sampling system of the nuclear power plant, representing the current principal component state of the coolant matrix. The wide-mass-range pre-scan involves a rapid scan within a mass-to-charge ratio range of 6 to 70, using safe initial values ​​for collisional helium flow rate, pool bias, and quadrupole mass selection window. This scan is used to capture the ion intensities of major matrix components such as lithium-7, boron-11, argon-40, and calcium-44, as well as the apparent signal intensities at mass-to-charge ratios of 52, 56, and 60. The initial interference response relationship is based on the argon-oxygen interference and calcium-oxygen interference initial response coefficients established from the pre-scan spectrum. This initial state vector will serve as the input for subsequent full-spectrum scan analysis and interference dynamic modeling, providing an accurate starting point for the entire parameter optimization process.

[0063] By acquiring matrix ion intensity vectors and constructing initial state vectors through a wide mass range pre-scan, this invention can grasp the approximate distribution of the main matrix and potential interferences before detailed spectral analysis, enabling subsequent high-resolution scanning and interference modeling to be targeted and avoid time-consuming high-resolution scanning under completely unknown conditions, thereby improving analytical efficiency and reducing the blindness of instrument tuning.

[0064] Traditional methods typically set a fixed set of collision cell parameters and perform single-point or narrow-window scans, failing to acquire comprehensive background information on matrix interference, resulting in a lack of basis for subsequent interference correction. This step utilizes pre-scanning to construct an initial state vector, explicitly incorporating prior information on the boron-lithium matrix and multi-atom interference into the state space, significantly enhancing the method's adaptability to matrix variations compared to existing techniques.

[0065] For example, during the process of gradually diluting the boron concentration in the primary loop from 2000 ppm to 1000 ppm after refueling at a nuclear power plant, the lithium concentration is also adjusted accordingly. Through this pre-scan, the instrument can capture significant changes in the signal intensity of lithium-7 and boron-11 within seconds, and update the initial interference response coefficient accordingly. This provides an initial interference relationship adapted to the matrix for the day's detection, ensuring that subsequent optimization is in the correct direction.

[0066] Step S20: Based on the initial state vector, perform a full-spectrum scanning analysis task using the Gaussian-Lorentz hybrid peak fitting method, and output the spectral analysis results;

[0067] The Gauss-Lorentz hybrid peak fitting method in this step is a peak shape analysis model. By adjusting the Lorentz proportion coefficient η, it combines the Gaussian and Lorentz components to more accurately describe the asymmetric peak shape caused by thermal diffusion and collision broadening under collision cell conditions. Performing the full spectrum analysis task means setting the collision cell and quadrupole with safe initial values ​​in the initial state vector within the mass-to-charge ratio range of 40 to 65. After obtaining high-resolution full spectrum data, the target isotope peaks such as argon-oxygen interference peak, calcium-oxygen interference peak, iron 56, nickel 60, and chromium 52 are independently fitted, and features such as peak position, full width at half maximum, asymmetry factor, and peak intensity are extracted. The matrix ion intensity is updated simultaneously to obtain the spectral analysis results.

[0068] Gaussian-Lorentz hybrid peak shape fitting can accurately resolve overlapping interference peaks and the peaks of the analyte isotope. Especially when the mass difference between the interference peak and the target peak is very small, reliable peak intensity and peak position information can still be obtained, providing accurate peak shape feature vectors and matrix ion intensity for subsequent interference yield calculation, thus ensuring the input quality of the interference quantification model.

[0069] Existing techniques often use pure Gaussian or pure Lorentz peak fitting. However, for non-ideal peak fitting within the collision cell, the residual error is large, and it is easy to misjudge the tailing of interfering peaks as the target signal. This step improves the accuracy of peak resolution by using hybrid peak fitting, especially when the argon-oxygen interference peak is close to the iron 56 at high boron concentrations, it can effectively separate them, and significantly reduce the error of overlapping peak resolution compared with traditional methods.

[0070] For example, when the boron concentration is 1500 ppm, the argon-oxygen interference peak (mass-to-charge ratio approximately 55.957) and the iron-56 isotope peak (mass-to-charge ratio 55.935) are only 0.022 amu apart, making it difficult to distinguish the peak shoulder contribution using conventional Gaussian fitting. This step, employing mixed peak shape fitting, clearly decomposes the two peaks, obtaining the true intensity of iron-56. Furthermore, the updated argon-40 and calcium-44 intensities accurately reflect matrix changes, preventing the iron-56 measurement value from being inflated by more than 30% due to the interference peak.

[0071] Step S30: Based on the spectral analysis results, the dynamic interference yield model and recursive least squares method are used to perform the interference quantification task, and the interference evaluation data is output.

[0072] The dynamic interference yield model in this step is a time-varying multiple linear model established for argon-oxygen interference. Its yield coefficients are not only related to argon40, but also include boron11, lithium7, and their cross terms with argon40 to reflect the nonlinear influence of the boron-lithium matrix on the argon-oxygen formation process. The recursive least squares method uses a forgetting factor of 0.92 and recursively updates the time-varying weight coefficients using the interference signals and matrix ion strength of the most recent three analysis periods, enabling the model to adaptively track the interference yield drift. The interference assessment data includes the interference equivalent background at a mass-to-charge ratio of 56 and the sensitivity decay rate based on the rhodium 103 internal standard. Both of these together describe the current severity of interference and the instrument sensitivity status.

[0073] This step combines a dynamic disturbance yield model with a recursive least squares method with a forgetting factor to achieve real-time tracking of disturbance yield changes caused by boron-lithium fluctuations. This avoids the problem of inaccuracy in traditional static yield models after matrix changes, and can provide accurate disturbance cost function input for parameter optimization, ensuring that subsequent optimization is based on a true balance between disturbance and sensitivity.

[0074] Compared to existing technologies that use preset fixed interference equations or periodic calibration, this step can automatically update the interference model in each analysis cycle without human intervention. This is particularly suitable for the dynamic changes in interference caused by frequent boron and lithium adjustments in nuclear power plant online monitoring scenarios. The introduction of a forgetting factor ensures both the model's convergence speed and avoids the continued impact of historical outlier data on the current estimate.

[0075] For example, when unit power adjustments cause a rapid increase of 200 ppm in the primary loop boron concentration, the argon-oxygen interference yield may increase by more than 15% over several analysis cycles. The recursive least squares algorithm in this step, upon receiving new matrix data, can reduce the interference equivalent background prediction error to within 5% within three cycles, thereby promptly triggering collision cell parameter optimization and preventing sudden spikes in iron element measurements.

[0076] Step S40: Based on the interference evaluation data, the Mamdani fuzzy control and sequential quadratic programming algorithm is used to perform the collision reaction pool parameter optimization task, and the optimized collision reaction pool parameters are output.

[0077] In this step, the Mamdani fuzzy controller takes the interference background bias and sensitivity decay rate as inputs, and outputs recommended adjustments for the collision gas helium flow rate, the pool bias pressure, and the quadrupole mass selection window after fuzzification, rule inference, and declarative processing. The sequential quadratic programming algorithm constructs an objective function based on the sensitivity loss model and the dynamic interference yield model, and accurately solves for the optimal combination of collision reaction pool parameters within the constraints of the fuzzy recommended adjustments. The sensitivity loss model describes the exponential effect of changes in helium flow rate and pool bias pressure on the sensitivity retention.

[0078] By combining fuzzy control with sequential quadratic programming, this step can take into account both human experience strategies and precise mathematical models. It uses fuzzy rules to avoid getting stuck in local unsuitable regions and sequential quadratic programming to perform high-precision optimization in the continuous parameter space. This allows it to find an accurate Pareto balance between interference cancellation and sensitivity maintenance, and output truly globally adapted collision reaction pool parameters.

[0079] Traditional methods often employ single-factor rotation or purely empirical rule adjustment, making it difficult to simultaneously optimize multiple collision pool parameters. Furthermore, they lack quantitative constraints on sensitivity loss, easily leading to an excessive pursuit of interference reduction and a significant drop in sensitivity. The fuzzy control-guided sequential quadratic programming algorithm proposed in this paper maximizes interference elimination while satisfying sensitivity loss tolerance, effectively solving the multi-objective optimization problem under multi-parameter coupling.

[0080] For example, when the equivalent background interference exceeds the preset target threshold by 50% and the sensitivity has decreased by 8%, the fuzzy controller will recommend appropriately increasing the helium flow rate and fine-tuning the cell bias. The sequential quadratic programming algorithm iteratively calculates in the neighborhood of this direction to obtain a specific combination of helium flow rate of 4.2 mL / min, cell bias of -18 V, and mass window offset of 0.02 amu, so that the background interference is reduced to within the threshold while the sensitivity decay is controlled within 10%, avoiding the situation where simply increasing the gas flow rate leads to a serious decrease in sensitivity.

[0081] Step S50: Perform verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and output the isotope concentration results of the element to be measured.

[0082] In this step, the verification scan is performed again using the optimized collision reaction cell parameters, and the verification interference equivalent background is recalculated based on the Gauss-Lorentz mixed peak fitting method in the previous step. The negative gradient local fine-tuning is to start with the optimized parameters and make small step adjustments along the negative gradient direction of the sensitivity loss model with respect to the collision gas helium flow rate and the cell bias pressure to quickly correct the verification deviation that may be caused by model error. When the deviation ratio of the verification interference equivalent background converges to within the preset convergence ratio, the concentration results of target isotopes such as iron, nickel, and chromium are output.

[0083] The verification scan and local fine-tuning mechanism constitute a closed-loop feedback correction loop, which can eliminate the residual deviation between the optimized model and the actual instrument response, ensuring that the final analysis results strictly meet the interference control requirements. At the same time, the fine-tuning process is based on gradient information and the step size is controlled, which will not cause large fluctuations in sensitivity, ensuring that the optimized parameters are stable and reliable in actual operation.

[0084] Existing parameter optimization methods typically stop at calculating and directly applying optimal parameters without considering the errors between the model and the actual physical process, which may lead to excessive interference during actual detection. This step, through verification scanning and convergence judgment, and performing local gradient fine-tuning when necessary, is equivalent to adding an online self-calibration function, significantly improving the robustness and parameter effectiveness of the method under the complex operating conditions of water samples from the primary loop of a nuclear power plant.

[0085] For example, after the optimized parameters were written into the execution verification scan, it was found that the verification interference equivalent background was still higher than the preset target threshold of 12%. At this time, the deviation ratio exceeded the convergence requirement of 5%. The system slightly increased the helium flow rate by 0.1 mL / min and finely adjusted the cell bias voltage by -1 V along the negative gradient direction. After verification again, the deviation dropped to 2%, and then the concentrations of iron 56, nickel 60 and chromium 52 isotopes were output to ensure that the detection data had been fully verified before being put into use.

[0086] Example 2: Furthermore, the present invention provides a parameter optimization system for a nuclear power plant water chemistry ultra-trace detection instrument, which employs a parameter optimization method for a nuclear power plant water chemistry ultra-trace detection instrument provided in the above embodiments, and can solve the technical problem of optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument. The beneficial effects of the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization system provided by the present invention are the same as the beneficial effects of the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization method provided in the above embodiments, and other technical features of the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0087] Example 3: This invention provides a parameter optimization device for a nuclear power plant water chemistry ultra-trace detection instrument. The device includes: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which are then executed to enable the at least one processor to perform the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization method described in Example 1. The nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and vehicle terminals (e.g., vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. This nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device is merely an example and should not limit the functionality or scope of the invention. A parameter optimization device for a nuclear power plant water chemistry ultra-trace detection instrument may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory or a program loaded from a storage device into a random access memory. The random access memory also stores various programs and data required for the operation of the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device. The processing unit, the read-only memory, and the random access memory are interconnected via a bus. An I / O interface is also connected to the bus. Typically, the following systems can be connected to the I / O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. The communication device allows the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device to communicate wirelessly or wiredly with other devices to exchange data. Although the figures show a nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device with various systems, it should be understood that implementation or possession of all the systems shown is not required. It can be implemented alternatively or have more or fewer systems.

[0088] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for optimizing parameters of a nuclear power plant water chemistry ultra-trace detection instrument. The computer program product provided by this invention can solve the technical problem of optimizing parameters of a nuclear power plant water chemistry ultra-trace detection instrument. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the above-described method for optimizing parameters of a nuclear power plant water chemistry ultra-trace detection instrument, and will not be repeated here.

[0089] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a read-only memory. When the computer program is executed by a processing device, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0090] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0091] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the present invention and its equivalents, the present invention also intends to include these modifications and variations.

Claims

1. A method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument, characterized in that, The methods include: Step S10: Obtain the boron-lithium prior data of the primary coolant sample, construct an initial state vector based on the boron-lithium prior data using a wide mass range pre-scan and initial disturbance response relationship, and output the initial state vector; Step S20: Based on the initial state vector, perform a full-spectrum scanning analysis task using the Gaussian-Lorentz hybrid peak fitting method, and output the spectral analysis results; Step S30: Based on the spectral analysis results, the dynamic interference yield model and recursive least squares method are used to perform the interference quantification task, and the interference evaluation data is output. Step S40: Based on the interference evaluation data, the Mamdani fuzzy control and sequential quadratic programming algorithm is used to perform the collision reaction pool parameter optimization task, and the optimized collision reaction pool parameters are output. Step S50: Perform verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and output the isotope concentration results of the element to be measured.

2. The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in claim 1, characterized in that, Step S10, which involves acquiring boron-lithium prior data for the primary coolant sample, constructing an initial state vector based on the boron-lithium prior data using a wide-range pre-scan and initial disturbance response relationship, and outputting the initial state vector, specifically includes: Step S101: Obtain the prior values ​​of boron concentration and lithium concentration in the primary coolant sample, set the collision gas helium flow rate, cell bias voltage and quadrupole mass selection window change as safe initial values, and perform a pre-scan within the mass-to-charge ratio range of 6 to 70 to obtain the pre-scan spectrum; Step S102: Based on the pre-scan spectrum, extract the intensity of lithium hepta-ion, boron unita-ion, argon forty-ion, calcium forty-four-ion, and the apparent signal intensity at mass-to-charge ratios of 52, 56, and 60 to obtain the matrix ion intensity vector. Step S103: Construct the initial disturbance response relationship based on the matrix ion intensity vector to obtain the initial state vector. : in, The strength of lithium hepta-ion, Boron undecyl ion strength, The ionic strength is argon forty. The ionic strength of calcium is 44. The initial response coefficient for argon-oxygen interference is... The initial response coefficient for calcium-oxygen interference is... The initial safe value for the collision gas helium flow rate. This is the initial safe value for the pool bias voltage. Select a safe initial value for the window change of the quadrupole mass.

3. The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in claim 1, characterized in that, Step S20, which involves performing a full-spectrum scanning analysis task based on the initial state vector using a Gaussian-Lorentz hybrid peak fitting method and outputting the spectral analysis results, specifically includes: Step S201: Read the initial safe values ​​of collision gas helium flow rate, pool bias pressure, and quadrupole mass selection window change from the initial state vector, write them into the collision pool controller and quadrupole power supply, and perform a full spectrum scan on the water sample to be tested in the range of mass-to-charge ratio of 40 to 65 to obtain full spectrum scan data. Step S202: Based on the full spectrum scanning data, the Gauss-Lorentz mixed peak shape fitting method is used to analyze the argon-oxygen interference peak, calcium-oxygen interference peak, iron 56 isotope peak, nickel 60 isotope peak and chromium 52 isotope peak respectively, and obtain the peak shape feature vector. Step S203: Based on the peak shape feature vector, synchronously update the intensity of lithium hepta-ion, boron eleven-ion, argon forty-ion, and calcium forty-four-ion to obtain the updated matrix ion intensity vector, and use the peak shape feature vector and the updated matrix ion intensity vector as the spectral analysis results.

4. The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in claim 3, characterized in that, In step S202, based on the full-spectrum scanning data, the Gauss-Lorentz mixed peak shape fitting method is used to analyze the argon-oxygen interference peak, calcium-oxygen interference peak, iron 56 isotope peak, nickel 60 isotope peak, and chromium 52 isotope peak to obtain the peak shape feature vector. This step specifically includes: Argon-oxygen interference peaks were identified near a mass-to-charge ratio of 55.957, calcium-oxygen interference peaks were identified near a mass-to-charge ratio of 59.953, iron 56 isotope peaks were identified near a mass-to-charge ratio of 55.935, nickel 60 isotope peaks were identified near a mass-to-charge ratio of 59.931, and chromium 52 isotope peaks were identified near a mass-to-charge ratio of 51.

941. The following Gaussian-Lorentz mixed peak shape function was used to fit each peak: in, mass-to-charge ratio The fitted signal intensity at that point, Peak intensity, The Lorentz ratio is the coefficient. The Lorentz half-width parameter, Peak position, The standard deviation is Gaussian. Based on the fitting results, a peak shape feature vector containing peak position, full width at half maximum (FWHM), asymmetry factor, and peak intensity is obtained.

5. The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in claim 1, characterized in that, Step S30, which involves performing interference quantization based on the spectral analysis results using a dynamic interference yield model and recursive least squares method to output interference evaluation data, specifically includes: Step S301: Read the peak shape feature vector and the updated matrix ion intensity vector from the spectral analysis results, and establish a dynamic interference yield model for argon-oxygen interference: in, For the first The intensity of the argon-oxygen interference signal in each analysis cycle. , , and The first Time-varying weighting coefficients for each analysis period; Step S302: Using the recursive least squares method with a forgetting factor of 0.92, update the time-varying weight coefficients based on the argon-oxygen interference signal intensity and matrix ion intensity vector of the most recent three analysis periods, and calculate the overlap integral ratio based on the peak position and full width at half maximum of the updated argon-oxygen interference peak and the iron 56 isotope peak. Step S303: Calculate the equivalent background of interference at mass-to-charge ratio 56 based on the overlap integral ratio, and introduce the sensitivity attenuation rate, using the equivalent background of interference and the sensitivity attenuation rate as interference evaluation data.

6. The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in claim 5, characterized in that, Step S40, which involves using the Mamdani fuzzy control and sequential quadratic programming algorithm to optimize the collision reaction pool parameters based on the interference evaluation data, and outputting the optimized collision reaction pool parameters, specifically includes: Step S401: Read the equivalent background of interference and the sensitivity attenuation rate at mass-to-charge ratio 56 from the interference evaluation data, compare the equivalent background of interference with the preset target interference threshold to obtain the interference background deviation, and input the interference background deviation and the sensitivity attenuation rate into the Mamdani fuzzy controller to obtain the recommended adjustment amount of collision gas helium flow rate, the recommended adjustment amount of pool bias pressure and the recommended adjustment amount of quadrupole mass selection window; Step S402: Based on the recommended adjustment of the collision gas helium flow rate, the recommended adjustment of the pool bias voltage, and the recommended adjustment of the quadrupole mass selection window, establish a sensitivity loss model: in, To the collision gas helium flow rate Pool bias Sensitivity retention amount Initial sensitivity, The sensitivity loss coefficient, It is the bias stability constant; Step S403: Construct an optimization objective function based on the sensitivity loss model and the dynamic interference yield model, and solve the optimization objective function using a sequential quadratic programming algorithm to obtain the optimized collision reaction pool parameters.

7. The method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in claim 1, characterized in that, Step S50, which involves performing verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and outputting the isotope concentration result of the element to be measured, specifically includes: Step S501: Write the optimized collision reaction cell parameters into the collision cell controller and quadrupole power supply. After the detection instrument stabilizes, perform a verification scan and recalculate the verification interference equivalent background according to the Gauss-Lorentz mixed peak fitting method in step S20. Step S502: Compare the verification interference equivalent background with the preset target interference threshold. If the deviation ratio is greater than the preset convergence ratio, then take the optimized collision reaction cell parameters as the starting point and perform local fine-tuning along the negative gradient direction of the sensitivity loss model with respect to the collision gas helium flow rate and the cell bias pressure to obtain the fine-tuned collision reaction cell parameters. Step S503: Based on the fine-tuned collision reaction cell parameters, perform the verification scan again until the deviation ratio of the verification interference equivalent background converges to within the preset convergence ratio, and output the isotope concentration result of the element to be measured.

8. A parameter optimization system for a nuclear power plant water chemistry ultra-trace detection instrument, applied to the parameter optimization method for a nuclear power plant water chemistry ultra-trace detection instrument as described in any one of claims 1 to 7, characterized in that, The nuclear power plant water chemistry ultra-trace detection instrument parameter optimization system includes: The initial state construction module is used to acquire boron-lithium prior data of the primary coolant sample, construct an initial state vector based on the boron-lithium prior data using a wide mass range pre-scan and initial disturbance response relationship, and output the initial state vector. The spectral scanning and analysis module is used to perform a full-spectrum scanning and analysis task based on the initial state vector using a Gaussian-Lorentz hybrid peak fitting method, and output the spectral analysis results. The interference quantization module is used to perform interference quantization tasks based on the spectral analysis results using a dynamic interference yield model and a recursive least squares method, and output interference evaluation data. The parameter optimization module is used to perform the collision reaction pool parameter optimization task based on the interference evaluation data using the Mamdani fuzzy control and sequential quadratic programming algorithm, and output the optimized collision reaction pool parameters. The verification output module is used to perform verification scanning and negative gradient local fine-tuning based on the optimized collision reaction cell parameters, and output the isotope concentration results of the element to be tested.

9. A parameter optimization device for ultra-trace water chemistry detection instruments in nuclear power plants, characterized in that, The nuclear power plant water chemistry ultra-trace detection instrument parameter optimization device includes: a memory, a processor, and a nuclear power plant water chemistry ultra-trace detection instrument parameter optimization program stored in the memory and executable on the processor. When the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization program is executed by the processor, it implements a nuclear power plant water chemistry ultra-trace detection instrument parameter optimization method according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a parameter optimization program for a nuclear power plant water chemistry ultra-trace detection instrument. When the nuclear power plant water chemistry ultra-trace detection instrument parameter optimization program is executed by a processor, it implements a method for optimizing the parameters of a nuclear power plant water chemistry ultra-trace detection instrument as described in any one of claims 1 to 7.

Images