Method for vortex rigidity feature extraction and compensation to inhibit valve geometry interference

Abstract

The application discloses a vortex hardness feature extraction and compensation method for inhibiting valve geometry interference, relates to the technical field of nondestructive testing and signal processing, and comprises the following steps: acquiring an original impedance voltage signal and key geometry parameters of a valve under a preset multi-frequency sinusoidal excitation signal; constructing a geometry-electromagnetic coupling interference compensation model based on the key geometry parameters, excitation frequencies and electromagnetic parameters, calculating a multivariate coupling geometry interference coefficient, stripping geometry interference from the original signal to obtain a pure impedance voltage signal; performing three-dimensional feature extraction on the pure signal to obtain amplitude features, phase features and hardness sensitive features; introducing a dynamic temperature compensation model to perform nonlinear compensation on a three-dimensional feature vector according to an environmental temperature; and outputting the compensated three-dimensional hardness feature vector. The application inhibits the influence of valve geometry differences on the eddy current signal by constructing a geometry-electromagnetic coupling interference compensation model, thereby improving the accuracy and robustness of eddy current hardness detection.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing and signal processing technology, specifically to a method for extracting and compensating for eddy current hardness features to suppress valve geometry interference. Background Technology

[0002] Eddy current testing, as a non-destructive testing method, has the advantages of short testing cycle, high efficiency, low cost and no damage to the surface quality of components in the field of material hardness sorting. It is widely used in the quality inspection of key components such as aero-engine valves.

[0003] Eddy current detection signals are the result of the coupling of multiple factors, including component geometry, electromagnetic parameters, and detection conditions. For valves, differences in geometric parameters such as stem diameter, overall length, and neck curvature can severely interfere with hardness feature extraction, leading to different eddy current response signals for the same hardness value and reducing sorting accuracy. Traditional methods typically use fixed thresholds or simple regression models for hardness sorting, which are insufficient to effectively suppress geometric interference. Furthermore, changes in the detection coil's own heating and ambient temperature can alter coil resistance and material electromagnetic parameters, causing eddy current impedance signal drift and further exacerbating feature instability. While some existing studies have introduced machine learning methods for hardness classification, these often require a large number of training samples and struggle to achieve online adaptive updates.

[0004] Therefore, there is an urgent need for a method that can remove geometric interference from the original eddy current signal, extract stable features sensitive to hardness, and compensate for temperature effects in real time, in order to solve the above-mentioned technical problems in valve hardness eddy current detection. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a method for extracting and compensating eddy current hardness features to suppress valve geometry interference.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for extracting and compensating eddy current hardness features to suppress valve geometry interference, characterized by comprising the following steps:

[0008] Step S1: Obtain the original impedance voltage signal of the valve under the preset multi-frequency sinusoidal excitation signal through the eddy current detection system, and simultaneously collect the key geometric parameters of the valve.

[0009] Step S2: Based on the key geometric parameters, eddy current excitation frequency, and valve electromagnetic parameters, construct a geometric-electromagnetic coupling interference compensation model and calculate the multivariable coupled geometric interference coefficient.

[0010] Step S3: Based on the multivariable coupling geometric interference coefficient, the interference voltage signal caused by the geometric structure change is removed from the original impedance voltage signal to obtain the pure impedance voltage signal;

[0011] Step S4: Perform three-dimensional feature extraction on the pure impedance voltage signal to obtain a three-dimensional feature vector including amplitude features, phase features and hardness-sensitive features;

[0012] Step S5: Introduce a dynamic temperature compensation model, and perform nonlinear compensation on the three-dimensional feature vector according to the current detection ambient temperature to obtain the compensated three-dimensional hardness feature vector.

[0013] Step S6: Output the compensated three-dimensional hardness feature vector.

[0014] Furthermore, in step S1, the key geometric parameters include the valve stem diameter, the total valve length, and the valve neck radius of curvature.

[0015] Furthermore, step S2 specifically includes the following steps:

[0016] Step S2.1: Construct a main effect term of diameter in Gaussian function form based on the difference between the valve stem diameter and the standard valve stem diameter; construct a main effect term of length in hyperbolic tangent function form based on the relative difference between the total valve length and the standard total valve length; construct a main effect term of neck curvature in Gaussian function form based on the difference between the valve neck curvature radius and the standard valve neck curvature radius; multiply the main effect terms of diameter, length, and neck curvature by the first comprehensive calibration coefficient to obtain the geometric main effect coefficient.

[0017] Step S2.2: Calculate the equivalent skin depth considering diameter variation based on the excitation frequency, valve permeability, valve conductivity, and valve stem diameter. Construct the geometric-electromagnetic coupling coefficient based on the ratio of the equivalent skin depth to the standard skin depth and the ratio of the valve permeability to the standard permeability.

[0018] Step S2.3: Multiply the geometric main effect coefficient with the geometric-electromagnetic coupling coefficient to obtain the multivariable coupled geometric interference coefficient.

[0019] Further, step S3 specifically includes the following steps: multiplying the multivariable coupling geometric interference coefficient with the original impedance voltage signal to obtain the interference voltage signal caused by the change in geometric structure, and subtracting the interference voltage signal from the original impedance voltage signal to obtain the pure impedance voltage signal.

[0020] Furthermore, step S4 specifically includes the following steps:

[0021] Step S4.1: Perform cross-correlation calculations on the pure impedance voltage signal with the sine reference signal and cosine reference signal that are in the same frequency as the excitation signal to obtain the resistance component and the reactance component;

[0022] Step S4.2: Calculate the square root of the sum of squares of the resistance component and the reactance component to obtain the amplitude characteristics;

[0023] Step S4.3: Calculate the arctangent function of the ratio of the resistance component and the reactance component to obtain the phase characteristics;

[0024] Step S4.4: Based on the ratio of the resistance component to the reactance component, the difference between the amplitude characteristic and the standard amplitude characteristic, and the difference between the phase characteristic and the standard phase characteristic, the hardness-sensitive characteristic is constructed.

[0025] Further, step S4.4 specifically includes the following steps: calculating the ratio of the resistive component to the reactive component; constructing a Gaussian function-form amplitude attenuation term with the difference between the amplitude feature and the standard amplitude feature as the independent variable; constructing a sinusoidal function-form phase modulation term with the difference between the phase feature and the standard phase feature as the independent variable; multiplying the ratio by the amplitude attenuation term and then by the phase modulation term, and adding a preset regularization constant to obtain the hardness-sensitive feature.

[0026] Furthermore, step S5 specifically includes the following steps:

[0027] Step S5.1: Calculate the effect function of temperature on valve conductivity based on the difference between the current ambient temperature and the standard temperature;

[0028] Step S5.2: Calculate the effect function of temperature on valve permeability based on the difference between the current ambient temperature and the standard temperature;

[0029] Step S5.3: Calculate the effect function of temperature on the resistance of the detection coil based on the difference between the current ambient temperature and the standard temperature;

[0030] Step S5.4: Based on the conductivity influence function, magnetic permeability influence function, and coil resistance influence function, construct the amplitude characteristic compensation coefficient, phase characteristic compensation coefficient, and hardness sensitivity characteristic compensation coefficient, respectively.

[0031] Step S5.5: Multiply the amplitude feature compensation coefficient, phase feature compensation coefficient and hardness sensitivity feature compensation coefficient by the corresponding amplitude feature, phase feature and hardness sensitivity feature respectively to obtain the compensated three-dimensional hardness feature vector.

[0032] Further, in step S5.4, constructing the amplitude characteristic compensation coefficient specifically includes the following steps: calculating the ratio of conductivity at the current temperature to conductivity at the standard temperature to obtain the conductivity temperature influence factor; calculating the ratio of the upper resistance at the current temperature to the upper resistance at the standard temperature to obtain the coil resistance temperature influence factor; using the difference between the current detection ambient temperature and the standard temperature as the independent variable, constructing a temperature attenuation term in the form of a Gaussian function; multiplying the conductivity temperature influence factor by the coil resistance temperature influence factor and then by the temperature attenuation term to obtain the amplitude characteristic compensation coefficient.

[0033] Further, in step S5.4, constructing the phase characteristic compensation coefficient specifically includes the following steps: calculating the ratio of the permeability at the current temperature to the permeability at the standard temperature to obtain the permeability temperature influence factor; calculating the ratio of the upper resistance at the current temperature to the upper resistance at the standard temperature to obtain the coil resistance temperature influence factor; multiplying the permeability temperature influence factor by the coil resistance temperature influence factor to obtain the phase characteristic compensation coefficient.

[0034] Further, in step S5.4, constructing the hardness-sensitive feature compensation coefficient specifically includes the following steps: calculating the ratio of conductivity at the current temperature to conductivity at the standard temperature to obtain the conductivity temperature influence factor; calculating the ratio of magnetic permeability at the current temperature to magnetic permeability at the standard temperature to obtain the magnetic permeability temperature influence factor; using the difference between the current detection ambient temperature and the standard temperature as the independent variable, constructing a temperature modulation term in the form of a sine function; multiplying the conductivity temperature influence factor by the magnetic permeability temperature influence factor and then by the temperature modulation term to obtain the hardness-sensitive feature compensation coefficient.

[0035] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0036] 1. This invention simultaneously considers the coupling effect of valve geometry, including diameter, length, neck curvature, permeability, and conductivity, and constructs a composite interference coefficient that includes a Gaussian function, a hyperbolic tangent function, and an equivalent skin depth considering diameter changes. This effectively removes interference signals caused by differences in geometric structure and improves the correlation between the extracted pure impedance signal and hardness.

[0037] 2. Based on the traditional amplitude and phase characteristics, this invention constructs a hardness-sensitive characteristic based on the electromagnetic loss characteristics of valve materials. By combining the ratio of the resistance component to the reactance component with the amplitude Gaussian attenuation term and the phase sinusoidal modulation term, and introducing a regularization constant to prevent division by zero, a high-sensitivity response to hardness changes is achieved.

[0038] 3. This invention comprehensively considers the nonlinear effects of temperature on conductivity, permeability and coil resistance, and designs compensation coefficients with different temperature-dependent function forms for different feature dimensions, which can eliminate the influence of temperature drift on feature vectors in real time. Attached Figure Description

[0039] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0040] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram of the three-dimensional feature extraction process according to an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of the dynamic temperature compensation process according to an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] like Figure 1 As shown, the method for extracting and compensating eddy current hardness features to suppress valve geometry interference includes the following steps:

[0045] Step S1: Obtain the original impedance voltage signal of the valve under the preset multi-frequency sinusoidal excitation signal through the eddy current detection system, and simultaneously collect the key geometric parameters of the valve.

[0046] Step S2: Based on the key geometric parameters, eddy current excitation frequency, and valve electromagnetic parameters, construct a geometric-electromagnetic coupling interference compensation model and calculate the multivariable coupled geometric interference coefficient.

[0047] Step S3: Based on the multivariable coupling geometric interference coefficient, the interference voltage signal caused by the geometric structure change is removed from the original impedance voltage signal to obtain the pure impedance voltage signal;

[0048] Step S4: Perform three-dimensional feature extraction on the pure impedance voltage signal to obtain a three-dimensional feature vector including amplitude features, phase features and hardness-sensitive features;

[0049] Step S5: Introduce a dynamic temperature compensation model, and perform nonlinear compensation on the three-dimensional feature vector according to the current detection ambient temperature to obtain the compensated three-dimensional hardness feature vector.

[0050] Step S6: Output the compensated three-dimensional hardness feature vector.

[0051] In step S1, the key geometric parameters include the valve stem diameter, the total valve length, and the valve neck radius of curvature.

[0052] Eddy current detection is used to obtain the original impedance voltage signal of the valve under a preset multi-frequency sinusoidal excitation signal, and the key geometric parameters of the valve are collected simultaneously, including the valve stem diameter, the total valve length, and the valve neck curvature radius. At the same time, the ambient temperature is collected and detected in real time by a temperature sensor. The frequency range of the preset multi-frequency sinusoidal excitation signal is 25Hz to 25kHz, and at least 8 frequency points are selected according to a geometric series.

[0053] Step S2 specifically includes the following steps:

[0054] Step S2.1: Construct a main effect term of diameter in Gaussian function form based on the difference between the valve stem diameter and the standard valve stem diameter; construct a main effect term of length in hyperbolic tangent function form based on the relative difference between the total valve length and the standard total valve length; construct a main effect term of neck curvature in Gaussian function form based on the difference between the valve neck curvature radius and the standard valve neck curvature radius; multiply the main effect terms of diameter, length, and neck curvature by the first comprehensive calibration coefficient to obtain the geometric main effect coefficient.

[0055] Step S2.2: Calculate the equivalent skin depth considering diameter variation based on the excitation frequency, valve permeability, valve conductivity, and valve stem diameter. Construct the geometric-electromagnetic coupling coefficient based on the ratio of the equivalent skin depth to the standard skin depth and the ratio of the valve permeability to the standard permeability.

[0056] Step S2.3: Multiply the geometric main effect coefficient with the geometric-electromagnetic coupling coefficient to obtain the multivariable coupled geometric interference coefficient.

[0057] The specific formula for the geometric main effect coefficient is as follows:

[0058]

[0059] in, This represents the geometric main effect coefficient, characterizing the degree of interference of valve geometry parameters alone on the eddy current signal. This represents the first overall calibration coefficient. This indicates the stem diameter of the valve currently being tested. Indicates the stem diameter of a standard valve. The diameter influences the scale factor, which is obtained by fitting experimental data. This represents the second comprehensive calibration coefficient. This indicates the total length of the valve currently being tested. Indicates the total length of a standard valve. Indicates the neck curvature radius of a standard valve. This indicates the radius of curvature of the valve neck currently being tested. The curvature influence scaling factor represents the sensitivity of the Gaussian function to curvature differences, and is obtained by fitting experimental data.

[0060] The specific formula for the geometric-electromagnetic coupling coefficient is as follows:

[0061]

[0062] in, The geometric-electromagnetic coupling coefficient represents the degree of interference between the interaction of geometric and electromagnetic parameters on the eddy current signal. This represents the coupling adjustment coefficient, obtained through experimental calibration. This represents the equivalent skin depth considering changes in diameter. Indicates the frequency of the excitation signal. This indicates the skin depth of a standard valve under standard conditions. This indicates the relative permeability of the current valve. Represents the relative permeability of a standard valve;

[0063] The specific formula for equivalent skin depth is:

[0064]

[0065] in, This represents the angular frequency of the excitation signal. Represents the permeability of free space. Indicates the current valve conductivity. This represents the diameter-skin depth coupling coefficient, obtained through experimental calibration.

[0066] The specific formula for the multivariable coupling geometric interference coefficient is as follows:

[0067]

[0068] in, It represents the multivariable coupled geometric interference coefficient, which comprehensively characterizes the total interference degree of the coupling effect between valve geometry and electromagnetic parameters on the eddy current signal.

[0069] Step S3 specifically includes the following steps: multiplying the multivariable coupling geometric interference coefficient by the original impedance voltage signal to obtain the interference voltage signal caused by the geometric structure change; subtracting the interference voltage signal from the original impedance voltage signal to obtain the pure impedance voltage signal, as shown in the following formula:

[0070]

[0071] in, This represents the pure impedance voltage signal after removing geometric interference. Represents the original impedance voltage signal. This represents the interference voltage signal caused by changes in geometric structure.

[0072] like Figure 2 As shown, step S4 specifically includes the following steps:

[0073] Step S4.1: Perform cross-correlation calculations on the pure impedance voltage signal with the sinusoidal reference signal and cosine reference signal of the same frequency as the excitation signal to obtain the resistance component and reactance component. The specific formulas are as follows:

[0074]

[0075] in, This represents the resistance component, reflecting the resistive characteristics of the valve material to the vortex field. This represents the reactance component, reflecting the reactance characteristics of the valve material to the eddy current field. This represents a discrete sampling sequence of a pure impedance voltage signal. This represents the number of sampling points within one period. Indicates the sampling period;

[0076] Step S4.2: Calculate the square root of the sum of squares of the resistive and reactive components to obtain the amplitude characteristic. The specific formula is as follows:

[0077]

[0078] in, This indicates the amplitude characteristic, used to characterize the intensity of the eddy current signal;

[0079] Step S4.3: Calculate the arctangent function of the ratio of the resistive and reactive components to obtain the phase characteristics. The specific formula is as follows:

[0080]

[0081] in, This indicates the phase characteristic, used to characterize the phase shift of the eddy current signal relative to the excitation signal;

[0082] Step S4.4: Based on the ratio of the resistive component to the reactive component, the difference between the amplitude characteristic and the standard amplitude characteristic, and the difference between the phase characteristic and the standard phase characteristic, the hardness-sensitive characteristic is constructed. The specific formula is as follows:

[0083]

[0084] in, This indicates a hardness-sensitive characteristic; the value shows a monotonic relationship with valve hardness. This represents the preset regularization constant, which is set to 10. -6 V is used to ensure numerical stability and prevent the denominator from being zero. The amplitude characteristics of a standard valve are obtained through calibration using standard parts. The scaling factor represents the effect of amplitude, controlling the sensitivity of the Gaussian function to amplitude differences. It is obtained by fitting experimental data. This represents the phase modulation coefficient, obtained through experimental calibration, used to adjust the nonlinear contribution of phase shift to hardness-sensitive characteristics. The phase characteristics of a standard valve are obtained through calibration using standard parts. This indicates the preset maximum phase offset.

[0085] Step S4.4 specifically includes the following steps: calculating the ratio of the resistive component to the reactive component; constructing a Gaussian function-form amplitude attenuation term with the difference between the amplitude feature and the standard amplitude feature as the independent variable; constructing a sinusoidal function-form phase modulation term with the difference between the phase feature and the standard phase feature as the independent variable; multiplying the ratio by the amplitude attenuation term and then by the phase modulation term, and adding a preset regularization constant to obtain the hardness-sensitive feature.

[0086] like Figure 3 As shown, step S5 specifically includes the following steps:

[0087] Step S5.1: Based on the difference between the current ambient temperature and the standard temperature, calculate the effect function of temperature on valve conductivity. The specific formula is as follows:

[0088]

[0089] in, This represents the conductivity at the current temperature. Indicates conductivity at standard temperature. The temperature coefficient of conductivity is obtained through experimental calibration. Indicates the current ambient temperature. This indicates the standard ambient temperature, usually set to 20℃.

[0090] Step S5.2: Based on the difference between the current ambient temperature and the standard temperature, calculate the effect function of temperature on valve permeability. The specific formula is as follows:

[0091]

[0092] in, This represents the relative permeability at the current temperature. This represents the linear temperature coefficient of magnetic permeability. It represents the second temperature coefficient of permeability, used to describe the nonlinear change of permeability with temperature;

[0093] Step S5.3: Based on the difference between the current ambient temperature and the standard temperature, calculate the effect function of temperature on the resistance of the detection coil. The specific formula is as follows:

[0094]

[0095] in, This indicates the coil resistance at the current temperature. This indicates the coil resistance at standard temperature. Indicates the temperature coefficient of coil resistance;

[0096] Step S5.4: Based on the conductivity influence function, magnetic permeability influence function, and coil resistance influence function, construct the amplitude characteristic compensation coefficient, phase characteristic compensation coefficient, and hardness sensitivity characteristic compensation coefficient, respectively.

[0097] Step S5.5: Multiply the amplitude feature compensation coefficient, phase feature compensation coefficient and hardness sensitivity feature compensation coefficient by the corresponding amplitude feature, phase feature and hardness sensitivity feature respectively to obtain the compensated three-dimensional hardness feature vector.

[0098] In step S5.4, constructing the amplitude characteristic compensation coefficient specifically includes the following steps: calculating the ratio of conductivity at the current temperature to conductivity at the standard temperature to obtain the conductivity temperature influence factor; calculating the ratio of the upper resistance at the current temperature to the upper resistance at the standard temperature to obtain the coil resistance temperature influence factor; using the difference between the current detection ambient temperature and the standard temperature as the independent variable, constructing a temperature decay term in the form of a Gaussian function; multiplying the conductivity temperature influence factor by the coil resistance temperature influence factor and then by the temperature decay term to obtain the amplitude characteristic compensation coefficient.

[0099] In step S5.4, constructing the phase characteristic compensation coefficient specifically includes the following steps: calculating the ratio of the permeability at the current temperature to the permeability at the standard temperature to obtain the permeability temperature influence factor; calculating the ratio of the upper resistance at the current temperature to the upper resistance at the standard temperature to obtain the coil resistance temperature influence factor; and multiplying the permeability temperature influence factor by the coil resistance temperature influence factor to obtain the phase characteristic compensation coefficient.

[0100] In step S5.4, constructing the hardness-sensitive feature compensation coefficient specifically includes the following steps: calculating the ratio of conductivity at the current temperature to conductivity at the standard temperature to obtain the conductivity temperature influence factor; calculating the ratio of magnetic permeability at the current temperature to magnetic permeability at the standard temperature to obtain the magnetic permeability temperature influence factor; using the difference between the current detection ambient temperature and the standard temperature as the independent variable, constructing a temperature modulation term in the form of a sine function; multiplying the conductivity temperature influence factor by the magnetic permeability temperature influence factor and then by the temperature modulation term to obtain the hardness-sensitive feature compensation coefficient.

[0101] The specific formula for the amplitude characteristic compensation coefficient is as follows:

[0102]

[0103] in, Indicates the amplitude characteristic compensation coefficient. This represents the temperature-dependent scaling factor, controlling the sensitivity of the Gaussian decay term to temperature shifts.

[0104] The specific formula for the phase characteristic compensation coefficient is as follows:

[0105]

[0106] in, Indicates the phase characteristic compensation coefficient;

[0107] The specific formula for the hardness-sensitive feature compensation coefficient is as follows:

[0108]

[0109] in, This represents the compensation coefficient for hardness-sensitive characteristics. The temperature modulation coefficient representing the hardness-sensitive characteristic is obtained through experimental calibration. This indicates the preset maximum operating temperature.

[0110] Multiply the amplitude feature compensation coefficient, phase feature compensation coefficient, and hardness sensitivity feature compensation coefficient by their respective amplitude feature, phase feature, and hardness sensitivity feature to obtain the compensated three-dimensional hardness feature vector. The specific formula is as follows:

[0111]

[0112] in, This represents the compensated three-dimensional hardness feature vector. The compensated three-dimensional hardness feature vector is output as the final hardness feature for subsequent hardness analysis or sorting.

[0113] This invention relates to several key coefficients, and the methods for obtaining each coefficient are as follows:

[0114] 1. Method for obtaining the first and second comprehensive calibration coefficients: At least five groups of valves with different geometric dimensions are selected. Each group includes standard parts with a gradient distribution of diameter, length, and radius of curvature. The hardness value of each valve is known. The original impedance voltage signal of each valve is acquired using an eddy current testing system, and its actual geometric parameters are obtained through precise measurement. A least-squares method is used for multivariate nonlinear fitting, aiming to minimize the error between the compensated signal and the standard signal, and the value that minimizes the error is obtained.

[0115] 2. Methods for obtaining the diameter and curvature influence scaling factors: Valve groups with gradient diameter distributions and identical other geometric parameters are selected, and their vortex response signals are measured. A Gaussian function is used to fit a curve with the diameter difference as the x-axis and the signal change rate as the y-axis. The standard deviation of the fitted Gaussian function is the diameter influence scaling factor. Similarly, the curvature influence scaling factor is obtained by experimenting with valve groups with gradient curvature radii and identical other geometric parameters. A nonlinear least squares method is used for fitting, and the initial value can be set to half of the tolerance range of the corresponding geometric parameters.

[0116] 3. Methods for obtaining the coupling adjustment coefficient and the diameter-skin depth coupling coefficient: Valve groups with gradient diameter distributions and identical other geometric parameters are selected, and their eddy current response signals are measured at different excitation frequencies. Based on the measured data, the product of the ratio of the equivalent skin depth to the standard skin depth and the relative change in permeability is used as the independent variable, and the signal change rate as the dependent variable. Linear regression analysis is employed, and the slope of the regression line is used as the coupling adjustment coefficient. The intercept of the regression line should be close to 1. The diameter-skin depth coupling coefficient is obtained as follows: With a fixed excitation frequency and permeability, the equivalent skin depth of valves with different diameters is measured. The relative change in diameter is used as the independent variable, and the normalized skin depth is used as the dependent variable. Linear regression is used to solve for the slope, which is the diameter-skin depth coupling coefficient.

[0117] 4. Method for obtaining the temperature coefficient of conductivity: Take a standard valve and place it in a constant temperature chamber. Increase the temperature in 5℃ increments within the range of 10℃ to 60℃, holding each temperature point for a sufficient time to ensure uniform valve temperature. Measure the conductivity value at each temperature point using a high-precision conductivity meter. Plot the temperature difference on the x-axis and the relative change in conductivity on the y-axis, and perform a linear fit. The slope of the fitted line is the temperature coefficient of conductivity.

[0118] 5. Methods for obtaining the linear and quadratic temperature coefficients of magnetic permeability: A standard valve is placed in a constant temperature chamber, and the temperature is increased in 5°C increments within the range of 10°C to 60°C, with sufficient holding time at each temperature point. The inductance value at each temperature point is measured using a high-precision impedance analyzer, and the relative permeability is calculated based on the coil parameters. A quadratic polynomial is used for curve fitting, with the temperature difference as the abscissa and the rate of change of relative permeability as the ordinate. The coefficients of the first term of the fitted polynomial are the linear temperature coefficient of magnetic permeability, and the coefficients of the quadratic term are the quadratic temperature coefficient of magnetic permeability.

[0119] 6. Method for obtaining the temperature coefficient of coil resistance: Place the eddy current detection coil in a constant temperature chamber and increase the temperature in 5℃ increments within the range of 10℃ to 60℃, holding each temperature point for a sufficient time. Use a high-precision multimeter to measure the DC resistance value of the coil at each temperature point. Plot the temperature difference on the x-axis and the rate of change of resistance on the y-axis, and perform linear fitting. The slope of the fitted line is the temperature coefficient of coil resistance.

[0120] 7. Method for obtaining the amplitude influence scale factor: Take valve groups with gradient hardness distribution and other identical conditions, and measure their amplitude characteristics. Plot the difference between the amplitude characteristics and the standard amplitude characteristics on the x-axis and the amplitude attenuation term on the y-axis. Adjust the amplitude influence scale factor to ensure that the relationship between the Gaussian attenuation term and the hardness calibration value shows a reasonable attenuation. Usually, 1 to 2 times the standard deviation of the amplitude difference is taken as the initial value of the amplitude influence scale factor.

[0121] 8. Method for obtaining the phase modulation coefficient: Take a valve group with a gradient distribution of hardness and measure its phase characteristics. Plot the difference between the phase characteristics and the standard phase characteristics on the x-axis and the rate of change of hardness on the y-axis. Fit a phase modulation term in the form of a sinusoidal function and solve for the phase modulation coefficient using the nonlinear least squares method. The value range is usually from 0 to 1.

[0122] 9. Method for obtaining the temperature influence scaling factor: Take a standard valve and place it in a constant temperature chamber to measure its amplitude characteristics at different temperatures. Plot the temperature difference as the abscissa and the normalized amplitude change rate as the ordinate. Use a Gaussian function to perform curve fitting. The standard deviation of the fitted Gaussian function is the temperature influence scaling factor, which is used to describe the degree of additional attenuation of the amplitude characteristics after the temperature exceeds the standard temperature.

[0123] 10. Method for obtaining the temperature modulation coefficient of hardness-sensitive characteristics: Take a standard valve and place it in a constant temperature chamber to measure its hardness-sensitive characteristic value at different temperatures. Plot the temperature difference as the abscissa and the normalized rate of change of hardness-sensitive characteristics as the ordinate. Fit a temperature modulation term in the form of a sine function. Solve the temperature modulation coefficient of hardness-sensitive characteristics by nonlinear least squares method. The value range is usually 0 to 0.5.

[0124] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0125] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for extracting and compensating eddy current hardness features to suppress valve geometry interference, characterized in that, Includes the following steps: Step S1: Obtain the original impedance voltage signal of the valve under the preset multi-frequency sinusoidal excitation signal through the eddy current detection system, and simultaneously collect the key geometric parameters of the valve. Step S2: Based on the key geometric parameters, eddy current excitation frequency, and valve electromagnetic parameters, construct a geometric-electromagnetic coupling interference compensation model and calculate the multivariable coupled geometric interference coefficient. Step S3: Based on the multivariable coupling geometric interference coefficient, the interference voltage signal caused by the geometric structure change is removed from the original impedance voltage signal to obtain the pure impedance voltage signal; Step S4: Perform three-dimensional feature extraction on the pure impedance voltage signal to obtain a three-dimensional feature vector including amplitude features, phase features and hardness-sensitive features; Step S5: Introduce a dynamic temperature compensation model, and perform nonlinear compensation on the three-dimensional feature vector according to the current detection ambient temperature to obtain the compensated three-dimensional hardness feature vector. Step S6: Output the compensated three-dimensional hardness feature vector; In step S1, the key geometric parameters include the valve stem diameter, the total valve length, and the valve neck radius of curvature. Step S2 specifically includes the following steps: Step S2.1: Construct a main effect term of diameter in Gaussian function form based on the difference between the valve stem diameter and the standard valve stem diameter; construct a main effect term of length in hyperbolic tangent function form based on the relative difference between the total valve length and the standard total valve length; construct a main effect term of neck curvature in Gaussian function form based on the difference between the valve neck curvature radius and the standard valve neck curvature radius; multiply the main effect terms of diameter, length, and neck curvature by the first comprehensive calibration coefficient to obtain the geometric main effect coefficient. Step S2.2: Calculate the equivalent skin depth considering diameter variation based on the excitation frequency, valve permeability, valve conductivity, and valve stem diameter. Construct the geometric-electromagnetic coupling coefficient based on the ratio of the equivalent skin depth to the standard skin depth and the ratio of the valve permeability to the standard permeability. Step S2.3: Multiply the geometric main effect coefficient with the geometric-electromagnetic coupling coefficient to obtain the multivariable coupling geometric interference coefficient; Step S4 specifically includes the following steps: Step S4.1: Perform cross-correlation calculations on the pure impedance voltage signal with the sine reference signal and cosine reference signal that are in the same frequency as the excitation signal to obtain the resistance component and the reactance component; Step S4.2: Calculate the square root of the sum of squares of the resistance component and the reactance component to obtain the amplitude characteristics; Step S4.3: Calculate the arctangent function of the ratio of the resistance component and the reactance component to obtain the phase characteristics; Step S4.4: Based on the ratio of the resistance component to the reactance component, the difference between the amplitude characteristic and the standard amplitude characteristic, and the difference between the phase characteristic and the standard phase characteristic, the hardness-sensitive characteristic is constructed.

2. The method according to claim 1, characterized in that, Step S3 specifically includes the following steps: multiplying the multivariable coupling geometric interference coefficient with the original impedance voltage signal to obtain the interference voltage signal caused by the change in geometric structure, and subtracting the interference voltage signal from the original impedance voltage signal to obtain the pure impedance voltage signal.

3. The method according to claim 2, characterized in that, Step S4.4 specifically includes the following steps: calculating the ratio of the resistive component to the reactive component; constructing a Gaussian function-form amplitude attenuation term with the difference between the amplitude feature and the standard amplitude feature as the independent variable; constructing a sinusoidal function-form phase modulation term with the difference between the phase feature and the standard phase feature as the independent variable; multiplying the ratio by the amplitude attenuation term and then by the phase modulation term, and adding a preset regularization constant to obtain the hardness-sensitive feature.

4. The method according to claim 3, characterized in that, Step S5 specifically includes the following steps: Step S5.1: Calculate the effect function of temperature on valve conductivity based on the difference between the current ambient temperature and the standard temperature; Step S5.2: Calculate the effect function of temperature on valve permeability based on the difference between the current ambient temperature and the standard temperature; Step S5.3: Calculate the effect function of temperature on the resistance of the detection coil based on the difference between the current ambient temperature and the standard temperature; Step S5.4: Based on the conductivity influence function, magnetic permeability influence function, and coil resistance influence function, construct the amplitude characteristic compensation coefficient, phase characteristic compensation coefficient, and hardness sensitivity characteristic compensation coefficient, respectively. Step S5.5: Multiply the amplitude feature compensation coefficient, phase feature compensation coefficient and hardness sensitivity feature compensation coefficient by the corresponding amplitude feature, phase feature and hardness sensitivity feature respectively to obtain the compensated three-dimensional hardness feature vector.

5. The method according to claim 4, characterized in that, In step S5.4, constructing the amplitude characteristic compensation coefficient specifically includes the following steps: calculating the ratio of conductivity at the current temperature to conductivity at the standard temperature to obtain the conductivity temperature influence factor; calculating the ratio of the upper resistance at the current temperature to the upper resistance at the standard temperature to obtain the coil resistance temperature influence factor; using the difference between the current detection ambient temperature and the standard temperature as the independent variable, constructing a temperature decay term in the form of a Gaussian function; multiplying the conductivity temperature influence factor by the coil resistance temperature influence factor and then by the temperature decay term to obtain the amplitude characteristic compensation coefficient.

6. The method according to claim 4, characterized in that, In step S5.4, constructing the phase feature compensation coefficient specifically includes the following steps: Calculate the ratio of the permeability at the current temperature to the permeability at the standard temperature to obtain the permeability temperature influence factor; calculate the ratio of the upper resistance at the current temperature to the upper resistance at the standard temperature to obtain the coil resistance temperature influence factor; multiply the permeability temperature influence factor by the coil resistance temperature influence factor to obtain the phase characteristic compensation coefficient.

7. The method according to claim 4, characterized in that, In step S5.4, constructing the hardness-sensitive feature compensation coefficient specifically includes the following steps: Calculate the ratio of conductivity at the current temperature to conductivity at the standard temperature to obtain the conductivity temperature influence factor; Calculate the ratio of magnetic permeability at the current temperature to that at the standard temperature to obtain the magnetic permeability temperature influence factor; construct a temperature modulation term in the form of a sinusoidal function with the difference between the current detection ambient temperature and the standard temperature as the independent variable; multiply the conductivity temperature influence factor by the magnetic permeability temperature influence factor and then by the temperature modulation term to obtain the hardness sensitivity feature compensation coefficient.

Images