A method for quantifying the wear depth of finger sleeves applicable to neutron flux measurement in nuclear power plants
By using frequency mixing simulation calculations and signal transformation matrix optimization, the error problem of wear depth and angle quantization in eddy current detection was solved, and higher precision wear quantization of finger sleeves was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- LINGAO NUCLEAR POWER
- Filing Date
- 2023-09-28
- Publication Date
- 2026-06-30
AI Technical Summary
In existing technologies, eddy current detection methods cannot effectively consider the effects of phase angle and circumferential wear angle when quantifying the wear depth of the neutron flux measurement finger sleeve. Furthermore, single-frequency simulation calculations cannot eliminate the influence of support structure noise, resulting in large errors in the quantification results.
A simulation model of finger sleeves with ring and wedge defects was established by using frequency mixing simulation. The external sleeve structure signal was eliminated by dual-frequency mixing. The signal amplitude and phase relationship diagram was plotted by combining defect parameters. The transformation matrix was solved iteratively by the least squares method and the conjugate gradient method to achieve accurate signal quantization.
It improves the accuracy and reliability of quantizing the wear depth of the finger sleeve, enables the quantization of the wear angle of unilateral wedge wear, and reduces the quantization error of eddy current signals.
Smart Images
Figure CN117517448B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of nuclear power testing equipment, and specifically relates to a method for quantifying the wear depth of a finger sleeve suitable for measuring neutron flux in nuclear power plants. Background Technology
[0002] The in-core flux measurement finger sleeve, also known as the finger sleeve, is the channel through which the neutron flux detector enters and exits the reactor fuel assembly. During use, the finger sleeve is worn down by the fluids in the reactor. Currently, the industry mainly uses the Bobbin probe eddy current detection method to periodically track and inspect finger sleeve wear defects; the quantification of the finger sleeve wear depth is based solely on the amplitude of the eddy current signal. However, this method has two significant drawbacks: first, the wear depth is not only affected by the eddy current amplitude, but its phase angle is also related to the wear depth; second, the eddy current amplitude is affected not only by the wear depth but also by the circumferential angle of the wear defect. Based on field experience with finger sleeve wear, it can be found that, due to the relatively stable wear environment, most finger sleeves exhibit a wedge-shaped or annular wear pattern, which can be used to improve the quantification of its depth.
[0003] In the actual detection of the depth of wear defects in finger sleeves, a mixing channel is required to eliminate the influence of the external sleeve structure. Mature commercial pipe eddy current detection simulation software can only calculate using a single frequency method. However, single frequency simulation will collect signals from other parts besides the finger sleeve (such as the support structure) and cannot eliminate the influence of noise from this part. Furthermore, there is an error in frequency between single frequency and mixing. Therefore, single frequency simulation will affect the quantification results. Summary of the Invention
[0004] The purpose of this invention is to provide a method for quantifying the wear depth of finger sleeves used in nuclear power neutron flux measurement. This method is based on frequency mixing simulation calculations and can improve calculation accuracy and data accuracy.
[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: a method for quantifying the wear depth of a finger sleeve suitable for nuclear power neutron flux measurement, comprising the following steps:
[0006] a. Measure the actual specifications and dimensions of the finger sleeves in nuclear power plants, study the actual wear defect morphology and specifications, use the equipment parameters when performing eddy current measurements on the finger sleeves, establish simulation models for finger sleeves with annular and wedge-shaped defects respectively, and perform simulation calculations;
[0007] b. Establish a unified calibration process using experimental data, calibrate simulation signals, and use models to calculate signals at multiple frequencies for different finger sleeve defects;
[0008] c. Select the mixing channel, perform mixing calculations on the multi-frequency simulation signal, eliminate the external sleeve structure signal of the finger sleeve, and the collected frequency signal can eliminate the external sleeve structure signal and enhance the signal after dual-frequency mixing.
[0009] d. Based on the specific parameters of the defect, plot the relationship between the defect depth and circumferential angle and the amplitude and phase of the simulation signal;
[0010] e. Import the measured signal values into the plotted image, and the signal can be quantized into defect parameters based on the coordinates and legend information.
[0011] In another implementation, in step b, a sample is selected as a calibration reference, and after measuring its signal, the calibration coefficients for the amplitude and phase of the simulated signal are calculated.
[0012] In another implementation, the algorithm for dual-frequency mixing in step c to eliminate the influence of the signal from the external sheath structure of the finger sheath is as follows:
[0013] To obtain the eddy current signals at the main frequency and the auxiliary frequency at the support structure, an affine transformation is performed on the auxiliary frequency: first, a magnitude displacement transformation is performed to ensure the support signals of both frequencies have the same equilibrium position; second, an angle rotation transformation is performed to ensure the support signals of both frequencies are in phase; finally, a scaling transformation is performed to ensure the amplitudes of the support signals of both frequencies are consistent. The signal transformation matrix obtained after the affine transformation is then used to derive the error function. Using the least squares method, the affine matrix parameter A that minimizes the error function is determined. This yields the functional relationship between the mixed data Cmi x and the eddy current data Cb of the main frequency and Ca of the auxiliary frequency: Cmi x = Cb - A·Ca.
[0014] In another implementation, in step c, let Sb be the dominant eddy current signal at the support structure, and Sa be the auxiliary eddy current signal at the support structure. An affine transformation is performed on the auxiliary frequency: first, a displacement transformation of Tx and Ty is performed to ensure the support signals of the two frequencies have the same equilibrium position; second, a rotation transformation of θ is performed to ensure the support signals of the two frequencies are in phase; finally, a scaling transformation is performed using Sx and Sy to ensure the amplitudes of the support signals of the two frequencies are consistent.
[0015] The signal obtained after the affine transformation is: Sa'=SaA(Tx,Ty,Sx,Sy)
[0016] The transformation matrix is:
[0017]
[0018] Where Tx and Ty are displacement transformation coefficients, θ is rotation transformation coefficient, and Sx and Sy are scaling transformation coefficients. Sa' is derived from [x'y'] = [xy 1]A
[0019] The error function is:
[0020] E = ||S b -S′ a || 2
[0021] According to the least squares method, we need to find the affine matrix parameters T that minimize the error function E. x T y ,θ,S x S y This problem can be transformed into solving the following system of equations:
[0022]
[0023] The system of equations can be solved iteratively using the conjugate gradient method. Assume C... b Eddy current data at the dominant frequency, C a To obtain the eddy current data for the auxiliary frequency, after calculating the transformation matrix, we can substitute it into the equation Cmix=Cb-A·Ca to obtain the mixed data.
[0024] The beneficial effects of this invention are as follows: This patent can effectively reduce the quantization error of the finger sleeve wear depth of the Bobbin probe eddy current signal, improve the reliability of quantization, and at the same time, it can quantify the wear angle of unilateral wedge wear. Attached Figure Description
[0025] Figure 1 This is a schematic diagram of the wedge-shaped defect in the simulation model of this patent;
[0026] Figure 2 This is a schematic diagram of the ring-shaped defect in the simulation model of this patent;
[0027] Figure 3 The Lisa figure refers to the signal after mixing of 160kHz-80kHz signals due to sleeve wear defects;
[0028] Figure 4 This refers to the quantization curve of the bushing wear defect under the mixing condition of 160kHz-40kHz. Detailed Implementation
[0029] The present invention will now be described in detail with reference to the embodiments shown in the accompanying drawings:
[0030] The method for quantifying the wear depth of the finger sleeve applicable to nuclear power neutron flux measurement includes the following steps:
[0031] a. Measure the actual specifications and dimensions of the finger sleeves in nuclear power plants, study the actual wear defect morphology and specifications, use the equipment parameters when performing eddy current measurements on the finger sleeves, establish simulation models for finger sleeves with annular and wedge-shaped defects respectively, and perform simulation calculations;
[0032] b. Establish a unified calibration process using experimental data, calibrate simulation signals, and use models to calculate signals at multiple frequencies for different finger sleeve defects;
[0033] c. Select the mixing channel, perform mixing calculations on the multi-frequency simulation signal, eliminate the external sleeve structure signal of the finger sleeve, and the collected frequency signal can eliminate the external sleeve structure signal and enhance the signal after dual-frequency mixing.
[0034] d. Based on the specific parameters of the defect, plot the relationship between the defect depth and circumferential angle and the amplitude and phase of the simulation signal;
[0035] e. Import the measured signal values into the plotted image, and the signal can be quantized into defect parameters based on the coordinates and legend information.
[0036] In step b, a sample is selected as the calibration reference, and after measuring its signal, the calibration coefficients for the amplitude and phase of the simulated signal are calculated.
[0037] In step c, let Sb be the dominant eddy current signal at the support structure, and Sa be the auxiliary eddy current signal at the support structure. An affine transformation is performed on the auxiliary frequency: first, a displacement transformation of Tx and Ty is performed to ensure the support signals of both frequencies have the same equilibrium position; second, a rotation transformation of θ is performed to ensure the support signals of both frequencies are in phase; finally, a scaling transformation is performed using Sx and Sy to ensure the amplitudes of the support signals of both frequencies are consistent.
[0038] The signal obtained after the affine transformation is: Sa'=SaA(Tx,Ty,Sx,Sy)
[0039] The transformation matrix is:
[0040]
[0041] Where Tx and Ty are displacement transformation coefficients, θ is rotation transformation coefficient, and Sx and Sy are scaling transformation coefficients. Sa' is derived from [x' y'] = [xy 1]A
[0042] The error function is:
[0043] E = ||S b -S′ a || 2
[0044] According to the least squares method, we need to find the affine matrix parameters T that minimize the error function E. x T y ,θ,S x S y This problem can be transformed into solving the following system of equations:
[0045]
[0046] The system of equations can be solved iteratively using the conjugate gradient method. Assume C... b Eddy current data at the dominant frequency, C a To obtain the eddy current data for the auxiliary frequency, after calculating the transformation matrix, we can substitute it into the equation Cmix=Cb-A·Ca to obtain the mixed data.
[0047] Specifically, such as Figure 1-2 As shown, in step a, a basic cylindrical finger sleeve model is first established according to the actual dimensions. Then, the model is refined based on the defect morphology. The characteristics of annular defects are: the axial width of the defect is basically fixed, starting from a certain point on the outer surface of the finger sleeve and extending circumferentially to both sides. The defect depth gradually decreases as it extends, and the maximum depth should not exceed 80% of the sleeve wall thickness. The characteristics of wedge-shaped defects are: the defect starts from a certain point on the outer surface of the finger sleeve and extends circumferentially to both sides while also extending axially to one side. The defect depth gradually decreases as it extends, and the maximum depth should not exceed 80% of the sleeve wall thickness. Finally, the probe information is added to complete the establishment of the entire model.
[0048] In step b, the selected calibration sample has a circular through-hole defect. The phase angles of each channel are adjusted to the noise level; an appropriate measurement channel amplitude is selected and adjusted to a suitable value, and then transferred to the other channels. For example, when the phase angles of each channel are adjusted to 20°, and the amplitude of the 80kHz absolute channel is set to 7V, the noise cancellation effect is optimal, resulting in an amplitude calibration factor of 2080.92 and a phase calibration factor of -89.79. These two calibration factors are then applied to the simulation data obtained from model a.
[0049] In step c, the signal values obtained from the simulation are first statistically analyzed. For each signal with different defect parameters, the real and imaginary parts of the strongest signal at different frequencies are extracted and retained. Then, dual-frequency mixing is performed on the signals of the same defect at different frequencies, with the high-frequency channel as the main frequency channel and the low-frequency channel as the auxiliary frequency channel, such as... Figure 3 As shown, the main frequency is a 160kHz channel, and the auxiliary frequency is a 40kHz channel. Finally, the mixing process is completed by converting the amplitude and phase values. The conversion formula is as follows:
[0050]
[0051] Where V is the amplitude, P is the phase, a is the real part, and b is the imaginary part.
[0052] Step d involves, after mixing, plotting the mixed signals of defects with different parameters onto the coordinate axes, with phase as the X-axis and amplitude as the Y-axis. Simultaneously, connecting defects of the same depth with those of the same wear angle yields the following result: Figure 4 The defect quantification curve shown.
[0053] After obtaining the quantization curve, step e can be performed, where the actual measured amplitude and phase are used as the horizontal and vertical axes. Importing these into the graph allows the identification of the corresponding depth and wear angle. This example uses six defective samples to verify the quantization accuracy; the specific results are shown in Table 1, with an average error of 9.75%.
[0054] Actual Defect Number 1 2 3 4 5 6 Measured amplitude (V) 13.03 10.06 27.08 11.55 36.36 7 Measured phase (°) 58 47 67 44 78 31 Actual wear angle (°) 165 165 255 225 285 195 Actual wear depth (mm) 0.85 0.68 1.02 0.68 1.19 0.51 Quantization depth (mm) 0.884 0.765 0.918 0.714 0.986 0.561 relative error 4％ 12.5％ 10％ 5％ 17％ 10％
[0055] Table 1
[0056] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit of the present invention should be covered within the scope of protection of the present invention.
Claims
1. A method for quantifying the wear depth of a finger sleeve suitable for nuclear power neutron flux measurement, characterized in that: It includes the following steps: a. Measure the actual specifications and dimensions of the finger sleeves in nuclear power plants, study the actual wear defect morphology and specifications, use the equipment parameters when performing eddy current measurements on the finger sleeves, establish simulation models for finger sleeves with annular and wedge-shaped defects respectively, and perform simulation calculations; b. Establish a unified calibration process using experimental data, calibrate simulation signals, and use models to calculate signals at multiple frequencies for different finger sleeve defects; c. Select the mixing channel, perform mixing calculations on the multi-frequency simulation signal, eliminate the external sleeve structure signal of the finger sleeve, and the collected frequency signal can eliminate the external sleeve structure signal and enhance the signal after dual-frequency mixing. d. Based on the specific parameters of the defect, plot the relationship between the defect depth and circumferential angle and the amplitude and phase of the simulation signal; e. Import the measured signal values into the plotted image, and the signal can be quantized into defect parameters based on the coordinates and legend information.
2. The method for quantifying the wear depth of a finger sleeve used in nuclear power plant neutron flux measurement according to claim 1, characterized in that: In step b, a sample is selected as the calibration reference, and after measuring its signal, the calibration coefficients for the amplitude and phase of the simulated signal are calculated.
3. The method for quantifying the wear depth of a finger sleeve applicable to nuclear power neutron flux measurement according to claim 1, characterized in that, In step c, the dual-frequency mixing algorithm for eliminating the influence of signals from the external sheath structure of the finger sheath is as follows: To obtain the eddy current signals at the main frequency and the auxiliary frequency at the support structure, an affine transformation is performed on the auxiliary frequency: first, a magnitude displacement transformation is performed to ensure the support signals of both frequencies have the same equilibrium position; second, an angle rotation transformation is performed to ensure the support signals of both frequencies are in phase; finally, a scaling transformation is performed to ensure the amplitudes of the support signals of both frequencies are consistent. The signal transformation matrix obtained after the affine transformation is then used to derive the error function. Using the least squares method, the affine matrix parameter A that minimizes the error function is determined. This yields the functional relationship between the mixed data Cmix, the eddy current data Cb of the main frequency, and the eddy current data Ca of the auxiliary frequency: Cmix = Cb - A·Ca.
4. The method for quantifying the wear depth of a finger sleeve for measuring neutron flux in nuclear power plants according to claim 3, characterized in that: In step c, let Sb be the dominant eddy current signal at the support structure, and Sa be the auxiliary eddy current signal at the support structure. An affine transformation is performed on the auxiliary frequency: first, a displacement transformation of Tx and Ty is performed to ensure the support signals of both frequencies have the same equilibrium position; second, a rotation transformation of θ is performed to ensure the support signals of both frequencies are in phase; finally, a scaling transformation is performed using Sx and Sy to ensure the amplitudes of the support signals of both frequencies are equal. The signal obtained after the affine transformation is: Sa'=SaA(Tx,Ty,Sx,Sy) The transformation matrix is: Where Tx and Ty are displacement transformation coefficients, θ is rotation transformation coefficient, and Sx and Sy are scaling transformation coefficients; Sa' is derived from [x'y'] = [xy 1]A The error function is: E=||S b -S′ a || 2 According to the least squares method, we need to find the affine matrix parameters T that minimize the error function E. x T y ,θ,S x S y This problem can be transformed into solving the following system of equations: The system of equations can be solved iteratively using the conjugate gradient method. Assume C... b Eddy current data at the dominant frequency, C a To obtain the eddy current data for the auxiliary frequency, after calculating the transformation matrix, we can substitute it into the equation Cmix=Cb-A·Ca to obtain the mixed data.