Sound velocity calibration method applied to multilayer variable-thickness structure

Abstract

The present invention relates to the technical field of sound velocity calibration, and in particular to a sound velocity calibration method applied to a multilayer variable-thickness structure, comprising the following calibration steps: firstly, constructing a planar multilayer medium laminate structure, and calculating a sound velocity of each layer in the planar multilayer medium laminate structure; then, constructing a curved multilayer medium laminate structure, arranging a plurality of discrete units in the curved multilayer medium laminate structure, and simultaneously, creating a fluctuation equation for calculating a sound pressure value of each discrete unit in the curved multilayer medium laminate structure; next, creating a loss function between the sound pressure value of each discrete unit calculated on the basis of the fluctuation equation and an actually-measured sound pressure value of each discrete unit; and finally, carrying out gradient descent calculation on the sound pressure value of each discrete unit on the basis of the loss function, to carry out an iterative update to obtain a specific sound velocity value serving as a weight parameter in each discrete unit when the loss function has a minimum value, wherein the sound velocity value is an optimal velocity value. The present invention can achieve accurate measurement of a velocity boundary of a material.

Description

A sound velocity calibration method for multi-layer variable thickness structures

[0001] priority

[0002] This application claims priority to Chinese patent application No. 202311765200.1 filed on December 21, 2023. Technical Field

[0003] The present invention relates to the technical field of sound velocity calibration, and in particular to a sound velocity calibration method applied to a multi-layer variable thickness structure. Background Art

[0004] At present, the measurement of material sound velocity is generally carried out based on the ultrasonic pulse bottom echo method. That is, for a material of a certain thickness, when the ultrasonic pulse emitted by the ultrasonic probe reaches the bottom surface of the material, the pulse will be reflected back to the probe due to the difference in acoustic impedance. The time it takes for the bottom echo to reach the probe is measured to calculate the sound velocity of the material. For single-layer planar structures, this method has high measurement accuracy. However, for actual production equipment, including various pipelines, pressure vessels, etc., they are often curved structures. During use, scaling or residual material may occur, resulting in multiple layers of variable thickness. In the process of applying the above method, the sound velocity change of each layer of the material is often not known in advance. Usually, relevant parameters are estimated through work experience, and then the curved surface structure is combined. This method makes the application of the ultrasonic pulse bottom echo method in the boundary calibration of multi-layer variable thickness structures subject to certain limitations, and the accuracy of the sound velocity measurement is low, so it is urgent to solve it.

[0005] Summary of the Invention

[0006] In order to avoid and overcome the technical problems existing in the prior art, the present invention provides a sound velocity calibration method applied to a multi-layer variable thickness structure. The present invention can more accurately calculate the sound velocity of ultrasound in the material.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A sound velocity calibration method applied to a multi-layer variable thickness structure includes the following calibration steps:

[0009] S1. Construct a planar multilayer dielectric stack structure and solve the sound velocity of each layer in the planar multilayer dielectric stack structure;

[0010] S2. Establishing a curved multilayer dielectric stack structure and providing a plurality of discrete units in the curved multilayer dielectric stack structure; and establishing a wave equation for calculating the sound pressure value of each discrete unit in the curved multilayer dielectric stack structure;

[0011] S3, establishing a loss function between the sound pressure value of each discrete unit calculated by the wave equation and the actual sound pressure value of each discrete unit;

[0012] S4. Perform gradient descent calculation on the sound pressure value of each discrete unit through the loss function, and iteratively update the specific size of the sound speed value as the weight parameter in each discrete unit when the loss function takes the minimum value. The sound speed value is the optimal speed value.

[0013] As a further solution of the present invention, the optimal speed value of the discrete unit is calculated by the optimal speed iteration formula, and the optimal speed iteration formula is as follows:

[0014] Among them, c g m+1 represents the sound velocity value of the g-th discrete unit at the m+1-th iteration; c g m represents the sound velocity value of the g-th discrete unit at the m-th iteration; H α represents the approximate Hessian matrix; α g m represents the iteration step size of the mth iteration of the gth discrete unit; E represents the matrix composed of the loss function; c g Represents the sound velocity value of the g-th discrete unit.

[0015] As a further solution of the present invention: the specific steps of step S1 are as follows:

[0016] S11. Construct a planar multilayer dielectric stack structure, selecting n types of dielectric materials, and stacking the various dielectric materials in a vertical direction from top to bottom, with the contact surface between two adjacent layers of dielectric materials forming a horizontal reflection plane;

[0017] S12. Using an ultrasonic transmitter whose array elements are arranged equidistantly along a straight line, irradiate the planar multilayer dielectric stack structure. The specific irradiation process is as follows:

[0018] S121. Select a rectangular parallelepiped planar multilayer dielectric stack structure, and establish a spatial rectangular coordinate system O-XYZ with the thickness direction of the planar multilayer dielectric stack structure (i.e., the direction in which the various dielectric materials are stacked in sequence from top to bottom along the vertical direction) as the positive direction of the Z axis, the width direction of the planar multilayer dielectric stack structure as the positive direction of the Y axis, and the length direction of the planar multilayer dielectric stack structure as the positive direction of the X axis; the top surface of the planar multilayer dielectric stack structure coincides with the XY plane, and a corner of the top surface coincides with the coordinate origin O;

[0019] S122. Arrange the array elements of the ultrasonic transmitter in sequence and at equal intervals along a straight line; move the ultrasonic transmitter along the positive direction of the Y axis to perpendicularly illuminate the reflection plane, forming a moving trajectory, and during the movement of the ultrasonic transmitter, the arrangement direction of the array elements is parallel to the positive direction of the X axis;

[0020] S123, repeatedly moving the ultrasonic transmitter multiple times along the positive direction of the Y axis to form a corresponding number of movement trajectories, wherein the movement trajectories are parallel to each other along the Y axis, and the distance between adjacent movement trajectories is equal to the distance between adjacent array elements;

[0021] The speed of sound of the ultrasonic wave emitted by the ultrasonic transmitter in the kth layer of dielectric material is c k , the thickness of the kth layer of dielectric material is d k , through the sound velocity of ultrasound and the thickness of the dielectric material, the corresponding time can be calculated using the time calculation formula. The time calculation formula is as follows:

[0022] Among them, t i,j,k It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the k-th layer of dielectric material and be received by the j-th array element after being reflected by the reflection plane of the k-th layer of dielectric material; t i,i,k It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the k-th layer of dielectric material and be received by the i-th array element after being reflected by the reflecting plane of the k-th layer of dielectric material; represents the root mean square of the sound velocity from the first dielectric material to the kth dielectric material; x i,r represents the position coordinate of the i-th array element on the X-axis in the r-th moving trajectory; x j,r represents the position coordinate of the j-th array element on the X-axis in the r-th moving trajectory;

[0023] S13, by Can be calculated The calculation formula is as follows:

[0024] Among them, t k It represents the one-way time of ultrasonic wave propagating vertically in the k-th layer of dielectric material; represents the root mean square of the sound velocity from the first layer of dielectric material to the nth layer of dielectric material; t i,i,k-1 It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the k-1-th layer of dielectric material and be received by the i-th array element after being reflected by the reflecting plane of the k-th layer of dielectric material;

[0025] S14. For n-1 layers of dielectric material, the following equation is obtained:

[0026] in, It represents the root mean square of the sound velocity from the first layer of dielectric material to the n-1th layer of dielectric material;

[0027] S15. c can be obtained by the equation in step S13 and the equation in step S14. n The specific expression formula is as follows:

[0028] Among them, t i,i,n-1 It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the n-1-th layer of dielectric material and be received by the i-th array element after being reflected by the reflection plane of the n-1-th layer of dielectric material; t i,i,n It represents the time taken for the ultrasonic wave emitted by the i-th array element to propagate to the n-th layer of dielectric material and be received by the i-th array element after being reflected by the reflecting plane of the n-th layer of dielectric material;

[0029] It can be measured by the corresponding measuring instrument t i,i,n 、 t i,i,n-1 , t i,i,k and t i,i,k-1 Substitute the value into c n The specific expression formula of can be used to obtain the sound speed of ultrasonic waves propagating in each layer of dielectric materials.

[0030] As a further solution of the present invention, the process of using the wave equation to calculate the sound pressure value of each discrete unit is as follows:

[0031] S21. Constructing a curved multilayer dielectric stack structure, selecting n types of dielectric materials, and stacking the various dielectric materials in a vertical direction from top to bottom, with the contact surface between two adjacent layers of dielectric materials forming a reflective curved surface;

[0032] S22. Using an ultrasonic transmitter, irradiate the curved multilayer dielectric stack structure according to the contents of steps S121 to S123; while each array element moves along the Y-axis, it forms irradiation vertical planes that are equidistant, parallel, and vertically arranged along the X-axis, and each irradiation vertical plane intersects each reflective curved surface;

[0033] S23. During the movement of the ultrasonic transmitter, points are taken on the movement path at set time intervals. The vertical irradiation rays formed by each point in the irradiation vertical plane intersect with the corresponding reflective surfaces to form intersection points. The intersection points are combined to divide the corresponding irradiation vertical plane into multiple grids. The intersection points constitute grid points of the grids, and the grid points of each grid constitute discrete units.

[0034] The wave equation is expressed as follows: A·P=B;

[0035] Where A represents the operator of the wave equation; P represents the sound pressure field matrix composed of the sound pressure values ​​of each discrete unit; B represents the laser emission source matrix; 2 Represents a two-dimensional gradient operator; T represents the time matrix composed of the time combination of each discrete unit receiving the signal; C represents the velocity matrix composed of the velocity combination of each discrete unit;

[0036] S24. By calculating the sound velocity of the ultrasonic wave propagating in each layer of dielectric material, and inputting the values ​​of each element in B and T into the wave equation, the sound pressure value of each discrete unit in P is calculated; and in the wave equation, the sound velocity value of each discrete unit is the weight parameter of the corresponding sound pressure value of each discrete unit.

[0037] As a further solution of the present invention: the loss function is expressed as follows:

[0038] Wherein, E represents the matrix formed by the loss function; ΔD represents the pressure difference matrix formed by the sound pressure difference between the sound pressure value of each discrete unit calculated by the wave equation and the sound pressure value actually measured for each discrete unit.

[0039] As a further solution of the present invention, the specific steps of step S3 are as follows:

[0040] S31. Calculate the partial derivative of the sound speed value through the loss function. The derivative results are as follows:

[0041] Where W represents the weight matrix of ΔD;

[0042] S32. Calculate the partial derivative of the sound velocity using the wave equation. The derivative results are as follows:

[0043] S33. Combining the derivation results of step S31 and step S32, we can obtain:

[0044] Among them, A -1 WΔD represents the wave field propagating in the opposite direction;

[0045] S34. According to the principle of gradient descent, the iteration step size of each discrete unit is set, and the iteration step size is substituted into the optimal speed iteration formula. The optimal sound speed value of each discrete unit is obtained by iteration when the loss function takes the minimum value.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] In order to solve the problem of inaccurate sound velocity measurement of multi-layer variable thickness materials, the present invention divides the entire sound velocity calibration process into two steps: First, the material is assumed to be a planar multi-layer dielectric stack structure, and the sound velocity is preliminarily calibrated by combining the ultrasonic pulse bottom echo method and the collected full matrix; Second, the velocity model obtained by the preliminary calibration is used as the initial model and put into the full waveform inversion iteration process. By continuously iterating the velocity model, it reaches the final convergence state, thereby achieving a refined calibration of the sound velocity boundary of the multi-layer variable thickness material. The ultrasonic pulse bottom echo method is relatively simple to process data and has a fast calculation speed, but the calibration accuracy for non-planar structures is not high; full waveform inversion obtains the final velocity field by utilizing the rich information contained in the entire pulse echo, which has high accuracy, but the convergence speed is very slow. Combining the two methods not only speeds up the convergence speed of the model, but also prevents the full waveform inversion iteration from diverging due to the large gap between the initial model and the actual velocity field. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] FIG1 is a schematic diagram of the main calibration process of the present invention.

[0049] FIG2 is a schematic diagram of an actual velocity model in the present invention.

[0050] FIG3 is a schematic diagram of a planar multilayer dielectric stack structure according to the present invention.

[0051] FIG4 is a schematic diagram of a curved multilayer dielectric stack structure in the present invention. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] Referring to Figures 1 to 4, a sound velocity calibration method applied to a multi-layer variable thickness structure includes the following:

[0054] 1. Preliminary calibration.

[0055] The ultrasonic transmitter used has multiple array elements arranged in sequence along the linear array direction. The array elements can be arranged at equal distances or at variable distances. The present invention mainly verifies the situation of the array elements arranged at equal distances.

[0056] For uniform dielectric materials, ultrasonic transmitters are used to irradiate them. Each array element emits cylindrical ultrasonic waves. The present invention only considers ultrasonic waves located in the same vertical plane. Ultrasonic waves in other directions or other interfaces are not within the calculation scope of the present invention.

[0057] When constructing a uniform dielectric material structure, a rectangular parallelepiped structure is typically selected, and a spatial rectangular coordinate system is established along the length, width, and thickness of the rectangular parallelepiped. As shown in Figure 2, the spatial rectangular coordinate system O-XYZ is established, with the thickness direction of the uniform dielectric material structure (i.e., the vertical direction from top to bottom) as the positive Z-axis, the width direction of the uniform dielectric material structure as the Y-axis, and the length direction of the uniform dielectric material structure as the X-axis. The top surface of the uniform dielectric material structure coincides with the XY plane, and one corner of this top surface coincides with the coordinate origin O. In Figure 2, the array elements are installed in the probe of the ultrasonic transmitter and arranged sequentially along the X-axis. Multiple rows of array elements can be arranged, forming a rectangular array arrangement. The probe moves along the Y-axis, forming a trajectory, each trajectory containing a corresponding number of scan tracks. Repeated movement along the Y-axis generates multiple trajectories, totaling R trajectories. Each moving track is evenly spaced along the X-axis, and the distance between adjacent tracks is the same as the distance between array elements. This distance design ensures that the distances between each scanning track formed by each array element along the X-axis are the same, improving scanning uniformity.

[0058] Calculate the propagation time of an ultrasonic wave. The ultrasonic wave is emitted by the i-th array element. After reaching the bottom surface of the uniform dielectric material structure, it begins to reflect. The reflection direction varies. It can be reflected directly back to the original transmitting array element or reflected to other array elements. The propagation time of the ultrasonic wave emitted by the i-th array element and reflected to the j-th array element is as follows:

[0059] Among them, t i,j It represents the time taken for the ultrasonic wave emitted by the i-th array element to propagate in the uniform dielectric material and be received by the j-th array element after being reflected at the reflection interface; t i,i It represents the time taken for the ultrasonic wave emitted by the i-th array element to propagate in the uniform dielectric material and be received by the i-th array element after being reflected at the reflection interface. The reflection interface of the uniform dielectric material is the transition surface between the uniform dielectric material and another material.

[0060] x i,r represents the position coordinate of the i-th array element on the X-axis in the r-th moving trajectory; x j,r represents the position coordinate of the jth array element on the x-axis in the rth moving trajectory. d is the distance between the reflecting interface and the surface of the array emitting ultrasound, and c is the propagation speed of ultrasound in the medium, i.e., the speed of sound.

[0061] In determining t i,iFinally, different times are obtained using different sound speeds. For combinations with the same spacing, their amplitudes are superimposed, the average value is taken, and then the amplitudes of the corresponding times of the corresponding receiving elements at different sound speeds are superimposed. i,i , the corresponding amplitude can be obtained. By comparing the amplitudes, the sound speed corresponding to the maximum amplitude is taken as the propagation speed of the ultrasonic wave in the uniform medium.

[0062] In a planar multilayer dielectric stack, a variety of different dielectric materials are stacked together from top to bottom to form a stacked structure. The reflection interface of each stacked layer is a horizontal plane, perpendicular to the vertical plane of ultrasonic propagation. This reflection interface is called a reflection plane.

[0063] Since the dielectric materials of each layer are different, the propagation speed of ultrasonic waves in each layer is also different. Assume that the speed of sound in the kth layer is c k , thickness d k , then:

[0064] is the root mean square of the sound velocity in all dielectric layers and is expressed as follows:

[0065] t k It represents the one-way time of ultrasonic wave propagating vertically in the k-th layer of dielectric material. Then we can know that:

[0066] For n-1 layers, we have:

[0067] By subtracting formula (5) from formula (6), we can relate the sound velocity of the nth layer to the corresponding root mean square, that is, formula (7):

[0068] The sound velocity of each layer of dielectric material can be calculated using formula (7).

[0069] 2. Fine calibration.

[0070] In the calculation of the sound velocity above, the reflection interface is set to a horizontal plane. However, in actual dielectric materials, the propagation interface is mostly curved. The velocity calculated by formula (7) can only be used as a reference. Therefore, it is necessary to have a reference velocity before further accurate calculations can be performed.

[0071] Consider the wave equation of an isotropic medium: A·P=B (8)

[0072] Where A represents the operator of the wave equation; P represents the sound pressure field matrix composed of the sound pressure values ​​of each discrete unit; B represents the laser emission source matrix;2 represents a two-dimensional gradient operator; T represents the time matrix composed of the time combination of each discrete unit receiving the signal; C represents the velocity matrix composed of the velocity combination of each discrete unit. The sound velocity value calculated by formula (7) is input into formula (8), and the loss function is defined by the two-norm. The loss function is expressed as follows:

[0073] The inversion problem is the process of minimizing the loss function. The partial derivative of formula (9) with respect to the speed of sound is as follows:

[0074] Where W represents the weight matrix, which expresses the relative contribution of the residual data in the overall inversion unit. It is called the Jacobian matrix J.

[0075] The partial derivatives of the left and right sides of formula (8) with respect to the speed of sound are obtained as follows:

[0076] Formula (11) avoids the direct calculation of the Jacobian matrix J. Substituting formula (11) into formula (10) yields:

[0077] Therefore, within the framework of the least squares norm, the process of obtaining the velocity gradient of the loss function is divided into two steps: 1. Calculating the wave field of the array element under the velocity model; 2. Using the signal residual obtained by the array element as a new signal source for backpropagation, and calculating the residual wave field under the velocity model.

[0078] The optimal speed iteration formula of the overall model is: H a =J T J

[0079] Substituting the numerical value corresponding to the structure into the optimal speed iteration formula, the corresponding sound speed can be calculated.

[0080] As shown in Figure 3, a planar multilayer dielectric stack structure is formed of two dielectric materials. As shown in Figure 4, a curved multilayer dielectric stack structure is formed of two dielectric materials constituting the planar multilayer dielectric stack structure, wherein the reflection surface is approximately a concave parabola. The number of array elements of the ultrasonic generator is 16, the probe frequency is 100Khz, and the sampling frequency is 10Mhz. Calculations according to the present invention show that in the planar multilayer dielectric stack structure, the sound velocity of the upper dielectric material is 2750m / s, and the sound velocity of the lower dielectric material is 2550m / s. In the curved multilayer dielectric stack structure, the sound velocity of the upper dielectric material is 3000m / s, and the sound velocity of the lower dielectric material is 2000m / s.

[0081] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A sound velocity calibration method applied to a multi-layer variable thickness structure, characterized in that: The calibration steps include: S1. Construct a planar multilayer dielectric stack structure and solve the sound velocity of each layer in the planar multilayer dielectric stack structure; S2. Establishing a curved multilayer dielectric stack structure, and setting a plurality of discrete units in the curved multilayer dielectric stack structure; and establishing a wave equation for calculating the sound pressure value of each discrete unit in the curved multilayer dielectric stack structure; S3, establishing a loss function between the sound pressure value of each discrete unit calculated by the wave equation and the measured sound pressure value of each discrete unit; S4. Perform gradient descent calculation on the sound pressure value of each discrete unit through the loss function, so as to iteratively update the specific size of the sound speed value used as the weight parameter in each discrete unit when the loss function takes the minimum value. The sound speed value is the optimal speed value.

2. The method for calibrating the sound velocity of a multilayer variable thickness structure according to claim 1, characterized in that: The optimal speed value of the discrete unit is calculated by the optimal speed iteration formula. The optimal speed iteration formula is as follows: Among them, c g m+1 represents the sound velocity value of the g-th discrete unit at the m+1th iteration; c g m represents the sound velocity value of the g-th discrete unit at the m-th iteration; H α represents the approximate Hessian matrix; α g m represents the iteration step size of the mth iteration of the gth discrete unit; E represents the matrix composed of the loss function; c g Represents the sound velocity value of the g-th discrete unit.

3. The method for calibrating the sound velocity of a multi-layer variable thickness structure according to claim 2, characterized in that: The specific steps of step S1 are as follows: S11, construct a planar multilayer dielectric stacking structure, select n kinds of dielectric materials, and stack the various dielectric materials in order from top to bottom along the vertical direction, and the contact surface between two adjacent layers of dielectric materials forms a water Flat reflective surface; S12, using an ultrasonic transmitter whose array elements are arranged in sequence and at equal intervals along a straight line to irradiate the planar multilayer dielectric stack structure, the specific irradiation process is as follows: S121, select a rectangular parallelepiped planar multilayer dielectric stacking structure, and establish a spatial rectangular coordinate system O-XYZ with the thickness direction of the planar multilayer dielectric stacking structure, that is, the direction in which various dielectric materials are stacked in sequence from top to bottom along the vertical direction as the positive direction of the Z axis, the width direction of the planar multilayer dielectric stacking structure as the positive direction of the Y axis, and the length direction of the planar multilayer dielectric stacking structure as the positive direction of the X axis; the top surface of the planar multilayer dielectric stacking structure coincides with the XY plane, and one corner of the top surface coincides with the coordinate origin O; S122, the array elements on the ultrasonic transmitter are arranged in sequence and equidistantly along a straight line; the ultrasonic transmitter is moved along the positive direction of the Y axis to vertically irradiate the reflection plane and form a moving track, and the arrangement direction of the array elements is parallel to the positive direction of the X axis during the movement of the ultrasonic transmitter; S123, repeatedly moving the ultrasonic transmitter multiple times along the positive direction of the Y axis to form a corresponding number of moving tracks, where each moving track is parallel to each other along the Y axis, and the distance between adjacent moving tracks is equal to the distance between adjacent array elements; The speed of sound propagated by the ultrasonic wave emitted by the ultrasonic transmitter in the kth layer of dielectric material is c k , the thickness of the kth dielectric material is d k , through the sound velocity of ultrasound and the thickness of the dielectric material, the corresponding time can be calculated using the time calculation formula. The time calculation formula is as follows: Among them, t i,j,k It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the k-th layer of dielectric material and be received by the j-th array element after being reflected by the reflection plane of the k-th layer of dielectric material; t i,i,k It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the k-th layer of dielectric material and be received by the i-th array element after being reflected by the reflection plane of the k-th layer of dielectric material; represents the root mean square of the sound velocity from the first dielectric material to the kth dielectric material; x i,r represents the position coordinate of the i-th array element on the X-axis in the r-th moving trajectory; x j,r represents the position coordinate of the j-th array element on the X-axis in the r-th moving trajectory; S13, by It can be calculated The calculation formula is as follows: Among them, t k It represents the one-way time of ultrasonic wave propagating vertically in the k-th layer of dielectric material; represents the root mean square of the sound velocity from the first dielectric material to the nth dielectric material; t i,i,k-1 It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the k-1-th layer of dielectric material and be received by the i-th array element after being reflected by the reflection plane of the k-th layer of dielectric material; S14. For n-1 layers of dielectric material, there is the following equation: in, It represents the root mean square of the sound velocity from the first dielectric material to the n-1th dielectric material; S15, c can be obtained by the equation in step S13 and the equation in step S14. n The specific expression formula is as follows: Among them, t i,i,n-1 It represents the time taken by the ultrasonic wave emitted by the ith array element to propagate to the n-1th layer of dielectric material and be received by the ith array element after being reflected by the reflection plane of the n-1th layer of dielectric material; t i,i,n It represents the time taken by the ultrasonic wave emitted by the i-th array element to propagate to the n-th layer of dielectric material and be received by the i-th array element after being reflected by the reflection plane of the n-th layer of dielectric material; It can be measured by the corresponding measuring instrument t i,i,n 、 t i,i,n-1 ,t i,i,k and t i,i,k-1 Substitute the value into c n From the specific expression formula of , the sound velocity of ultrasonic wave propagating in each layer of dielectric material can be obtained.

4. The method for calibrating the sound velocity of a multi-layer variable thickness structure according to claim 3, characterized in that: The process of using the wave equation to calculate the sound pressure value of each discrete unit is as follows: S21, constructing a curved multilayer dielectric stacking structure, selecting n kinds of dielectric materials, and stacking the various dielectric materials in sequence from top to bottom along the vertical direction, and the contact surface between two adjacent layers of dielectric materials constitutes a reflective curved surface; S22, using an ultrasonic transmitter to irradiate the curved multilayer dielectric stack structure according to the contents of steps S121 to S123; each array element forms an irradiated vertical plane that is equidistant and parallel to each other and arranged vertically along the X-axis direction in the process of moving along the Y-axis, and each irradiated vertical plane intersects with each reflective curved surface; S23, during the movement of the ultrasonic transmitter, points are taken on the moving path at set time intervals, and the vertical irradiation rays formed by each point in the irradiation plumb plane intersect with the corresponding reflection surfaces to form intersection points; the intersection points are combined to separate the corresponding irradiation plumb plane into a plurality of grids, the intersection points constitute grid points of the grids, and the grid points of each grid constitute discrete units; The wave equation is expressed as follows: A·P=B; Wherein, A represents the operator of the wave equation; P represents the sound pressure field matrix composed of the sound pressure values ​​of each discrete unit; B represents the laser emission source matrix; represents a two-dimensional gradient operator; T represents the time matrix composed of the time combination of each discrete unit receiving the signal; C represents the speed matrix composed of the speed combination of each discrete unit; S24. By calculating the sound velocity of the ultrasonic wave propagating in each layer of dielectric material, and inputting it into the wave equation together with the numerical values ​​of each element in the set B and T, the sound pressure value of each discrete unit in P is calculated; and in the wave equation, the sound velocity value of each discrete unit is the weight parameter of the corresponding sound pressure value of each discrete unit.

5. The method for calibrating the sound velocity of a multi-layer variable thickness structure according to claim 4, characterized in that: The loss function is expressed as follows: Wherein, E represents a matrix formed by loss functions; ΔD represents a pressure difference matrix formed by the sound pressure difference between the sound pressure value of each discrete unit calculated by the wave equation and the sound pressure value actually measured for each discrete unit.

6. The method for calibrating the sound velocity of a multi-layer variable thickness structure according to claim 5, characterized in that: The specific steps of step S3 are as follows: S31. The partial derivative of the sound speed value is obtained through the loss function. The derivative result is as follows: Where W represents the weight matrix of ΔD; S32. Use the wave equation to find the partial derivative of the sound velocity value. The derivative result is as follows: S33, combining the derivation results of step S31 and step S32, we can get: Among them, A -1 WΔD represents the wave field propagating in the opposite direction; S34. According to the principle of gradient descent, the iteration step size of each discrete unit is set, and the iteration step size is substituted into the optimal speed iteration formula, and the optimal sound speed value of each discrete unit is obtained by iteration when the loss function takes the minimum value.

Images