A method, apparatus, medium, and product for generating randomly distributed meshes for backing materials.
By determining the minimum surface spacing and the constraints of the three-dimensional volume fraction field in the backing material, a randomly distributed mesh that meets the acoustic performance requirements is generated. This solves the shortcomings of the traditional RSA algorithm in simulating inter-particle scattering coupling and material inhomogeneity, and realizes the controllability and efficient modeling of acoustic simulation.
Patent Information
- Application Number
- CN202611264944.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-19
- Publication Date
- 2026-11-17
AI Technical Summary
Traditional RSA algorithms cannot effectively control near-field scattering coupling between particles and simulate the non-uniformity of macroscopic materials when generating random distribution grids for backing materials. This results in uncontrollable acoustic attenuation spectra in simulation results and difficulty in achieving accurate modeling of gradient materials.
By acquiring the target acoustic parameter set, determining the minimum surface spacing value and the three-dimensional target volume fraction field, and combining multiple constraint detection to generate a material identification matrix, we can ensure that the particle distribution meets the requirements of non-overlapping, controllable spacing and macroscopic acoustic impedance gradient, thereby realizing the controllable acoustic response of the particle distribution grid and the direct modeling of the gradient matching layer.
This approach achieves controllability and accuracy in acoustic simulation results, reduces the number of simulation design iterations, improves the accuracy and efficiency of acoustic modeling of backing materials, and lowers simulation trial-and-error costs.
Smart Images

Figure CN123116576A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of backing material design technology for super-powered transducers, specifically to a method, apparatus, medium, and product for generating a randomly distributed grid for backing materials. Background Technology
[0002] The backing material in high-energy transducers is typically made of epoxy resin substrate mixed with high-impedance metal particles and cured. Its main function is to rapidly attenuate rearward-radiated ultrasonic waves and achieve a certain degree of acoustic impedance matching. When performing acoustic simulation of this type of material using the finite element method, the first step is to construct a random distribution model of the particles in a three-dimensional mesh space.
[0003] The RSA (Random Sequential Adsorption) algorithm can generate non-overlapping random structures. The basic process is as follows: randomly generate particle positions one by one in the matrix grid. If the position of the particle to be placed does not overlap with the position of the already placed particle, then the corresponding grid is occupied; otherwise, the position is randomly generated again until all particles are placed.
[0004] However, directly applying the traditional RSA algorithm to backing material simulation has the following significant drawbacks: 1. Inability to control near-field scattering coupling between particles: Traditional RSA algorithm only considers the absence of geometric overlap when placing particles. However, when the particle surfaces are infinitely close or even tangent, they are in a certain ultrasonic frequency band of acoustic radiation field. The small particle spacing will cause a strong near-field scattering coupling effect, which will cause the simulated acoustic attenuation peak frequency and amplitude to deviate from the design target value, affecting the attenuation spectrum prediction and making the acoustic attenuation spectrum uncontrollable. 2. Inability to simulate macroscopic material inhomogeneity: Traditional RSA algorithms generate statistically uniform random structures, while actual backing materials have macroscopic gradients in particle volume fraction due to manufacturing processes or acoustic gradient design requirements. This results in a lack of local volume fraction control mechanisms, making it difficult to accurately model gradient materials and simulate volume fraction gradients.
[0005] Therefore, there is an urgent need for a method that can transform the acoustic performance requirements of the backing material into microstructural constraints and generate a randomly distributed mesh that meets the expected acoustic response in one go. Summary of the Invention
[0006] This invention provides a method for generating randomly distributed meshes for backing materials, in order to solve the problems that traditional RSA algorithms cannot control near-field scattering coupling between particles and cannot simulate macroscopic material inhomogeneities.
[0007] In a first aspect, the present invention provides a method for generating a randomly distributed mesh for a backing material, the method comprising: Obtain the target acoustic parameter set of the target backing material; based on the target acoustic parameter set, determine the minimum surface spacing value of each particle in the target backing material and the corresponding three-dimensional target volume fraction field of the target backing material. The distribution of the three-dimensional target volume fraction field satisfies the acoustic impedance gradient requirement; using multiple minimum surface spacing values and the three-dimensional target volume fraction field as constraints, sequentially detect the overlap, spacing, and local volume fraction of each particle in the target backing material, and obtain a material identification matrix. The material identification matrix includes the material identification of multiple particles that meet all detection requirements; based on the material identification matrix, generate the particle distribution grid of the target backing material.
[0008] The random distribution mesh generation method for backing materials provided by this invention determines the minimum surface spacing value and the three-dimensional target volume fraction field by combining the target acoustic parameter set. This transforms macroscopic acoustic performance indicators into two types of hard constraints: microscopic particle arrangement. By forcibly applying the minimum spacing constraint, the near-field scattering coupling effect between particles can be effectively reduced, making the acoustic attenuation spectrum of the simulated particle distribution mesh highly consistent with the target acoustic attenuation spectrum, thus achieving controllable acoustic response. Furthermore, by using the target volume fraction field to constrain the placement of local particles, a backing material model with a target acoustic impedance gradient can be directly generated, providing a direct modeling method for the simulation design of gradient matching layers of backing materials. Further, by combining the two constraints to sequentially perform triple detection of overlap, spacing, and local volume fraction and generate a material identification matrix, particle arrangement schemes that do not meet acoustic requirements are filtered layer by layer. Invalid particle positions are eliminated from three dimensions: geometric non-overlap, compliant scattering spacing, and local particle concentration matching gradient, achieving full-dimensional compliance verification of the microstructure. Furthermore, a particle distribution mesh is generated based on the material identification matrix. A three-dimensional mesh model suitable for finite element acoustic simulation is directly output based on the verified compliant particle data, eliminating the need for multiple iterative adjustments and providing a simulation mesh for the backing material that matches the acoustic design requirements in one step. Therefore, by implementing this invention, the acoustic performance requirements of the backing material are directly transformed into multiple constraints on the arrangement of micro-particles, and a random mesh that simultaneously satisfies no geometric overlap, controllable particle spacing, and macroscopic acoustic impedance gradient distribution is generated in one step. This significantly reduces the number of simulation design iterations, eliminates the unpredictable acoustic attenuation spectrum problem, lowers the trial-and-error cost of acoustic simulation, and improves the accuracy and design efficiency of acoustic modeling for the backing material.
[0009] In one optional implementation, the minimum surface spacing value of each particle in the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material are determined based on the target acoustic parameter set, including: Based on the upper limit of the working frequency band of the target acoustic parameter set and the longitudinal wave velocity of the matrix, the minimum surface spacing value of each particle in the target backing material is determined; based on the target acoustic attenuation spectrum of the target acoustic parameter set, the volume fraction and particle size distribution of each particle in the target backing material are determined through multiple scattering model inversion; based on the acoustic impedance gradient requirements, volume fraction and particle size distribution of the target acoustic parameter set, a three-dimensional target volume fraction field is constructed.
[0010] The random distribution mesh generation method for backing materials provided by this invention solves for the minimum surface spacing value by using the upper limit of the working frequency band and the longitudinal wave velocity of the matrix. This can accurately suppress the near-field scattering coupling effect between particles in the corresponding frequency band, ensuring that the simulated acoustic attenuation peak frequency and amplitude closely match the design value. Furthermore, by combining the multiple scattering model to invert the volume fraction and particle size distribution of each particle, the basic parameters of micro-particles are derived from the desired macroscopic attenuation performance, realizing a quantitative mapping from acoustic attenuation index to particle microscopic properties, avoiding errors caused by manually setting particle parameters based on experience. Furthermore, by combining the acoustic impedance gradient requirement, volume fraction, and particle size distribution to construct a three-dimensional target volume fraction field, the gradient impedance design requirement is transformed into a spatially continuously changing particle concentration constraint field, accurately replicating the macroscopic non-uniform structure caused by the gravity settlement of the actual backing material and the gradient layer, realizing quantitative modeling of gradient materials.
[0011] In one optional implementation, constrained by multiple minimum surface spacing values and a three-dimensional target volume fraction field, the overlap, spacing, and local volume fraction of each particle in the target backing material are sequentially detected, resulting in a material identification matrix, including: Select any target particle from among the particles and determine whether there are other particles that overlap with the target particle. If there are no other particles that overlap with the target particle, use multiple minimum surface spacing values as constraints to detect the surface spacing between the target particle and any remaining particle, and determine whether the surface spacing between the target particle and any remaining particle meets the spacing detection requirements. If the surface spacing between the target particle and any remaining particle meets the spacing detection requirements, use a three-dimensional target volume fraction field as constraints to detect the local volume fraction of the target particle, and determine whether the local volume fraction of the target particle meets the local volume fraction detection requirements. If the local volume fraction of the target particle meets the local volume fraction detection requirements, return to the step of selecting any target particle from among the particles, and iterate repeatedly until multiple target particles whose local volume fraction meets the local volume fraction detection requirements are obtained. Construct a material identification matrix based on multiple target particles.
[0012] The random distribution mesh generation method for backing materials provided by this invention, by randomly selecting target particles and verifying particle overlap, can eliminate candidate particle positions that would cause mesh intersection failure in advance, reducing the amount of invalid calculations in subsequent spacing and volume fraction detection, and improving the overall mesh generation efficiency. Furthermore, under the premise of no overlap, the particle surface spacing is verified based on the minimum surface spacing constraint. By strictly controlling the minimum particle spacing, near-field scattering coupling caused by excessive particle proximity is prevented, stabilizing the acoustic attenuation spectrum characteristics of the simulation output and ensuring the predictability of the acoustic simulation results. Further, after the spacing detection meets the standard, the local volume fraction is verified based on the three-dimensional volume fraction field. By matching the spatial gradient concentration constraint, the particle proportion in the local region of each particle is controlled to conform to the impedance gradient distribution, accurately reproducing the macroscopic non-uniform microstructure of the backing material. Furthermore, after all three detections are qualified, the particles are iteratively filled. Through a layered progressive verification logic, only particle positions that simultaneously meet the geometric, scattering, and gradient concentration requirements are retained, and the complete three-dimensional space is gradually filled, ensuring that all particles ultimately meet all acoustic constraints. Finally, a material identification matrix was constructed based on all qualified particles. By uniformly storing the spatial and attribute identifiers of all compliant particles in a standardized matrix carrier, the structured output of particle data was achieved, providing well-organized and directly callable underlying data for mesh generation.
[0013] In one optional implementation, selecting a particle that satisfies any target particle from among the particles, and determining whether there are other particles that overlap with the target particle, includes: Construct spherical voxel masks for each particle in the target backing material and calculate the spherical voxel mask value for each particle. The spherical voxel mask is a binary matrix. Select any target particle among the particles and randomly generate candidate coordinates of the target particle's center according to preset boundary conditions. Based on the candidate coordinates of the center, construct a bounding box matrix centered on the center of the target particle and calculate the bounding box matrix value. When both the spherical voxel mask value and the bounding box matrix value of the target particle meet the preset requirements, determine that there are no particles overlapping with the target particle in the bounding box matrix.
[0014] The random distribution mesh generation method for backing materials provided by this invention constructs a spherical voxel binary mask for the particles and calculates the mask values. By quantifying the three-dimensional space occupied by the particles using a voxel matrix, it achieves a digital representation of the particle geometric boundaries, providing a standardized numerical calculation basis for overlap determination. Furthermore, it randomly generates candidate coordinates for the center of the target particles, achieving a disordered particle distribution and preserving the natural random arrangement characteristics of the backing material particles, thus making it conform to the microscopic morphology of the real material. Further, it constructs a bounding box matrix based on the center coordinates and calculates the matrix values. By limiting the overlap detection range through local bounding boxes, it eliminates the need to traverse all existing particles, significantly reducing the computational load for overlap determination and improving the mesh generation speed. Finally, when both the mask and bounding box values meet preset requirements, it determines that there is no overlap, achieving standardized determination logic without manual intervention, avoiding the problems of missed or false judgments of geometric overlap.
[0015] In one optional implementation, when no other particles overlap with the target particle, the surface distance between the target particle and any remaining particle is detected using multiple minimum surface distance values as constraints, and it is determined whether the surface distance between the target particle and any remaining particle meets the distance detection requirements, including: If there are no other particles overlapping with the target particle, calculate the Euclidean distance between the centers of any remaining particles; based on multiple Euclidean distance values, calculate the surface distance between the target particle and any remaining particle; if the surface distance between the target particle and any remaining particle is less than the minimum surface distance of the target particle, determine that the surface distance between the target particle and any remaining particle meets the distance detection requirements.
[0016] The random distribution grid generation method for backing materials provided by this invention combines the center-to-center distance of spheres to calculate the particle surface spacing value. By eliminating the influence of particle radius, the surface spacing of the particles can be directly obtained, which can accurately match the surface spacing judgment index required to suppress near-field scattering, avoiding the judgment bias caused by using only the center-to-center distance. Furthermore, when all surface spacings are not less than the minimum surface spacing threshold, the spacing is judged to meet the standard, ensuring that there are no near-distance particles around the target particles that would cause scattering coupling, thus stabilizing the intensity of scattering interference under the control of the ultrasonic frequency band. Therefore, by implementing this invention, by ensuring that the spacing between each particle in the final randomly distributed grid meets the minimum surface spacing value constraint, the influence of near-field scattering coupling effect is avoided, thereby enabling the target backing material to meet the target acoustic parameter requirements. This completely avoids the problem of acoustic attenuation spectrum distortion caused by near-field scattering coupling at the microscopic particle spacing level.
[0017] In one optional implementation, when the surface distances between the target particle and any remaining particle meet the distance detection requirements, the local volume fraction of the target particle is detected using a three-dimensional target volume fraction field as a constraint, and it is determined whether the local volume fraction of the target particle meets the local volume fraction detection requirements, including: When the surface distance between the target particle and any remaining particle meets the distance detection requirements, calculate the side length of the cube bounding box centered on the center of the target particle; determine the total number of grids in the preset statistical window based on the side length; calculate the local volume fraction of the target particle within the preset statistical window based on the total number of grids; obtain the target volume fraction of the target particle in the three-dimensional target volume fraction field, and determine the target threshold based on the target volume fraction and the preset positive tolerance value; when the local volume fraction is greater than the target threshold, determine that the local volume fraction of the target particle meets the local volume fraction detection requirements.
[0018] The random distribution grid generation method for backing materials provided by this invention calculates the side length of the cubic bounding box of the particle center after the spacing meets the standard, and determines the total number of grids in the statistical window. This limits the concentration statistics to only the local area around the particle, which aligns with the logic of local micro-concentration characterization of the material. It quantifies the grid base of the local statistical area, provides a unified denominator for volume fraction calculation, and ensures that the local concentration calculation standard for particles at different locations is consistent. Furthermore, calculating the local volume fraction within the window based on the total number of grids can quantitatively characterize the actual particle proportion in the local area where the particle is located, thus intuitively reflecting the local micro-concentration distribution state. Furthermore, the target volume fraction at the corresponding location is extracted from the three-dimensional volume fraction field, and a judgment threshold is generated by combining it with the positive tolerance value. This takes into account both the gradient field standard and reasonable process tolerance, avoiding the direct rejection of qualified particles due to small concentration fluctuations, and adapting to the small non-uniformity characteristics of actual material preparation. Furthermore, when the local volume fraction is higher than the threshold, the concentration is judged to meet the standard, ensuring that the global particle distribution conforms to the preset macroscopic non-uniform gradient structure. Therefore, by implementing this invention, it is possible to ensure that the volume fraction of each particle in the randomly distributed grid satisfies the three-dimensional target volume fraction field, thereby enabling the particle distribution grid of the generated target backing material to also meet the target acoustic parameter requirements of the target backing material. Under the premise of strictly following the acoustic impedance gradient distribution requirements, it is compatible with the small concentration deviation caused by the actual preparation, and realizes the automated and accurate verification of local particle concentration under macroscopic gradient constraints.
[0019] In one alternative implementation, the method further includes: If no valid location is found in the target particle within the preset maximum number of attempts, adjust the preset positive tolerance value or the three-dimensional target volume fraction field. The absence of a valid location indicates that any detection result of the target particle overlap, spacing, or local volume fraction does not meet the requirements.
[0020] The random distribution mesh generation method for backing materials provided by this invention, by presetting a maximum number of attempts, prevents the target particles from repeatedly failing to meet a certain detection requirement when undergoing triple detection at the selected effective location, thus solving the problem of the random distribution mesh generation process getting stuck and resulting in wasted computing power, manpower, and time. Furthermore, by adjusting the preset positive tolerance value or the three-dimensional target volume fraction field in real time, it is beneficial to generate the random distribution mesh in one go, reducing iterations and saving resources.
[0021] In one alternative implementation, the method further includes: The scattering cross section of each particle in the target backing material is obtained, and the target scattering cross section of any two particles in the target backing material is calculated. The target scattering cross section is the sum of the scattering cross sections of any two particles. The surface spacing value of each particle in the target backing material is gradually increased, and the normalized scattering cross section of any two particles in the target backing material is calculated using a two-sphere scattering model. Based on the normalized scattering cross section and the target scattering cross section, the rate of change of the scattering cross section of any two particles is calculated. When the rate of change of the scattering cross section is less than a preset threshold, the surface spacing value of any two particles is obtained, and the minimum surface spacing value of each particle in the target backing material is calculated.
[0022] The random distribution mesh generation method for backing materials provided by this invention obtains the scattering cross section of a single particle and calculates the scattering cross section of a dual-particle target, thus establishing a quantitative evaluation benchmark for scattering coupling effects using the total scattering cross section of the dual particles as an uncoupled ideal benchmark. Furthermore, by progressively increasing the particle surface spacing and calculating the normalized scattering cross section using a dual-sphere scattering model, the actual scattering intensity of the dual particles under different particle spacings can be quantitatively characterized, revealing the influence of spacing changes on ultrasonic scattering coupling. Further, by comparing the normalized scattering cross section with the ideal target scattering cross section and solving for the rate of change of the scattering cross section, the near-field scattering coupling degree is transformed into a quantifiable rate of change index, achieving a numerical evaluation of the coupling effect. Furthermore, when the rate of change is lower than a preset threshold, the corresponding surface spacing is extracted as the minimum surface spacing value. The minimum spacing is then calculated in reverse with the scattering coupling intensity meeting the standard, ensuring that the scattering coupling effect between particles is negligible at this spacing. This allows for determining the constraint threshold from the acoustic scattering mechanism level rather than simply relying on empirical settings, further improving the matching degree between the acoustic simulation results of the mesh model and the design objectives. Therefore, by implementing this invention, appropriate methods for determining the minimum surface spacing value can be selected according to different actual situations, demonstrating strong versatility.
[0023] In a second aspect, the present invention provides a device for generating a randomly distributed grid for a backing material, comprising: The parameter acquisition module is used to acquire the target acoustic parameter set of the target backing material, as well as the parameters of the substrate material and the metal particles. The data processing module determines the minimum surface spacing value of each particle of the target backing material and the corresponding three-dimensional target volume fraction field of the target backing material based on the acquired target acoustic parameter set. The constraint detection module performs constrained placement of the metal particles by detecting the overlap, spacing, and local volume fraction of each particle. The verification output module is used to record, form, and output the placed material identification matrix, which can be used for simulation verification of the material identification matrix.
[0024] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the backing material random distribution grid generation method of the first aspect or any corresponding embodiment described above.
[0025] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the backing material random distribution grid generation method of the first aspect or any corresponding embodiment described above.
[0026] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the method for generating a random distribution grid for backing material according to the first aspect or any corresponding embodiment described above. Attached Figure Description
[0027] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0028] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first process of a method for generating a random distribution grid for backing material according to an embodiment of the present invention; Figure 3 This is a slice comparison diagram of the random distribution grid generation method for backing material according to an embodiment of the present invention and the traditional RSA method; Figure 4 This is a schematic diagram of the second process of the method for generating a random distribution grid for backing material according to an embodiment of the present invention; Figure 5This refers to the target volume fraction field and local statistical window of the random distribution grid generation method for backing material according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the third process of the method for generating a random distribution grid for backing material according to an embodiment of the present invention; Figure 7 This is a spherical voxel mask constructed using a random distribution grid generation method for backing material according to an embodiment of the present invention; Figure 8 This is a spacing detection diagram of a random distribution grid generation method for backing material according to an embodiment of the present invention; Figure 9 This is a three-dimensional orthogonal slice of the gradient backing material generated according to an embodiment of the present invention; Figure 10 This is a comparison chart of acoustic attenuation spectra generated from model simulation results according to an embodiment of the present invention; Figure 11 This is a structural block diagram of a backing material random distribution grid generation device according to an embodiment of the present invention; Figure 12 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation
[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0030] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.
[0031] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0032] As an optional application scenario of this invention, considering the specific application environment architecture or specific hardware architecture upon which the random distribution mesh generation method for backing materials depends, the specific application environment architecture or specific hardware architecture is described here. For example... Figure 1 As shown, the architecture system may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.
[0033] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.
[0034] This invention provides a method for generating a randomly distributed mesh for backing materials. By directly converting the acoustic performance requirements of the backing material into multiple constraints on the arrangement of micro-particles, a random mesh that simultaneously satisfies the requirements of no geometric overlap, controllable particle spacing, and macroscopic acoustic impedance gradient distribution is generated in one go. This reduces the number of simulation design iterations, eliminates the problem of unpredictable acoustic attenuation spectrum, reduces the cost of acoustic simulation trial and error, and improves the accuracy and design efficiency of acoustic modeling of backing materials.
[0035] According to an embodiment of the present invention, a method for generating a random distribution grid of a backing material is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.
[0036] This embodiment provides a method for generating a randomly distributed mesh for a backing material, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 2 This is a flowchart of a method for generating a randomly distributed mesh for backing material according to an embodiment of the present invention, as shown below. Figure 2 As shown, the process includes the following steps: Step S201: Obtain the target acoustic parameter set of the target backing material.
[0037] In an alternative embodiment, the target backing material refers to a composite functional material assembled inside the ultrasonic transducer, which is a mixture of an epoxy resin matrix and high acoustic impedance metal particles such as tungsten powder / copper powder, cured and molded.
[0038] Furthermore, the target acoustic parameter set represents a pre-defined set of acoustic design parameters used to constrain the arrangement of micro-particles in the backing material, and may include the acoustic attenuation spectrum of the target material. Acoustic impedance gradient requirements, operating frequency band .
[0039] Step S202: Based on the target acoustic parameter set, determine the minimum surface spacing value of each particle in the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material.
[0040] In one optional embodiment, the minimum surface spacing value of each particle represents the minimum safe distance between the solid surfaces of any two metal filler particles within the ultrasonic transducer backing material. Furthermore, at this spacing, the near-field scattering coupling effect between the particles is sufficiently weak to prevent the acoustic attenuation spectrum from deviating from a preset specification.
[0041] Furthermore, the distribution of the three-dimensional target volume fraction field satisfies the acoustic impedance gradient requirement, that is, the spatial distribution of the three-dimensional target volume fraction field should make the acoustic impedance estimated by the equivalent density approach the target value.
[0042] Step S203: Using multiple minimum surface spacing values and a three-dimensional target volume fraction field as constraints, the overlap, spacing and local volume fraction of each particle in the target backing material are detected sequentially, and a material identification matrix is obtained.
[0043] In one optional embodiment, the material identification matrix includes material identifications for multiple particles that meet all testing requirements, and further, the material identifications are used to distinguish different particles.
[0044] In one alternative embodiment, the traditional RSA algorithm can only avoid the geometric overlap of particles. It cannot limit the near-field scattering coupling caused by the close proximity of particles, which would cause the acoustic attenuation spectrum to deviate from the design value. It also cannot control the proportion of particles in local areas to replicate the volume fraction gradient of the actual backing material.
[0045] In this embodiment, by superimposing the minimum surface spacing and the three-dimensional target volume fractional field as dual constraints, the unqualified particle arrangement is screened out from three dimensions: geometry, acoustic scattering, and macroscopic gradient structure. It is possible to screen out qualified particles that simultaneously meet the requirements of geometric non-interference, particle spacing matching acoustic scattering, and local particle concentration matching impedance gradient. Furthermore, the spatial and attribute information of all qualified particles can be uniformly organized into a standardized material identification matrix, which solves the problems of uncontrollable simulation results and inability to model gradient backing materials in the original algorithm.
[0046] Step S204: Generate the particle distribution grid of the target backing material based on the material identification matrix.
[0047] In one optional embodiment, the particle distribution grid represents a three-dimensional discrete mesh model adapted for finite element acoustic simulation of ultrasonic backing materials, consisting of matrix mesh elements and high-impedance metal particle elements filling the mesh. Furthermore, each element within the mesh is distinguished by a material identification matrix (epoxy resin matrix / metal particles).
[0048] The material identification matrix is stored only in a two-dimensional digital matrix and does not have the ability to visualize and simulate three-dimensional spatial geometry, so it cannot be used to directly carry out ultrasonic acoustic finite element simulation.
[0049] In one optional embodiment, an index mapping rule is established to match the sphere center coordinates of particles in the material identification matrix to the spatial coordinate system of the three-dimensional matrix mesh of the target backing material, and to generate the spherical spatial occupancy range of each particle according to its radius. Simultaneously, all three-dimensional voxel mesh elements within the spherical range are marked.
[0050] Furthermore, by traversing all voxel mesh elements within the 3D matrix, if an element falls within the area occupied by any qualified spherical particle, it is marked as a metal particle element based on its material identifier. Furthermore, the remaining mesh elements not occupied by particles are uniformly assigned the value of epoxy resin matrix elements. Furthermore, this operation allows for the classification of all mesh elements into material categories.
[0051] Furthermore, based on the constructed three-dimensional target volume fraction field data, the data can be written into each local grid region according to the spatial coordinates, and the standard particle volume fraction corresponding to each spatial location can be recorded in the grid model. At the same time, the acoustic impedance gradient distribution characteristics are preserved, so that the grid has the ability to represent macroscopic non-uniform structures.
[0052] Furthermore, it is possible to traverse all mesh cells, verify whether the cell assignments correspond one-to-one with the qualified particle information in the material identification matrix, and remove invalid mesh cells that exceed the three-dimensional boundary of the backing and delete abnormal cells that conflict with the triple detection rules, so as to ensure that all particles inside the mesh meet the three constraints of no overlap, minimum surface spacing, and local volume fraction.
[0053] Furthermore, the completed and verified 3D voxel mesh is integrated and exported into a general mesh format for finite element simulation, forming the final target backing material particle distribution mesh. This mesh can simultaneously contain geometric spatial information, particle material properties, and spatial volume fraction gradient data, and can be directly imported into finite element simulation tools to perform acoustic performance simulation calculations such as ultrasonic attenuation and near-field scattering.
[0054] In one instance, such as Figure 3As shown, a slice comparison diagram of the random distribution grid generation method for backing material according to an embodiment of the present invention and the traditional RSA method is provided. The first row represents a three-dimensional slice generated by random distribution grid based on the traditional RSA method. It can be further seen that the surface spacing between a large number of particles is too small, significantly smaller than the minimum surface spacing determined by the present invention based on acoustic design requirements. At ultrasonic operating frequencies, such excessively small particle spacing is very likely to cause strong near-field scattering coupling effects, thereby seriously affecting the prediction accuracy of the acoustic attenuation spectrum.
[0055] Furthermore, Figure 3 The second row represents a three-dimensional slice generated by the random distribution grid method of the backing material in this embodiment of the invention. It can be further seen that all particles in the second row are placed with strict constraints on the preset minimum surface spacing based on their particle type, which can effectively avoid the influence caused by excessive near-field scattering coupling effect.
[0056] This embodiment effectively reduces near-field scattering coupling between particles by forcibly applying the minimum surface spacing between each particle and the three-dimensional target volume fractional field determined based on the acoustic parameter set, making the acoustic response controllable. At the same time, it enables the macroscopic gradient structure of the target backing material model to be generated in accordance with the actual target requirements, providing a direct modeling method for the design of the gradient matching layer. Furthermore, it achieves full-process constraints during implementation, allowing the randomly distributed mesh generated in the first instance to approximate the target acoustic performance indicators, significantly shortening the trial-and-error cycle of material development and reducing a large amount of design costs.
[0057] This embodiment provides a method for generating a randomly distributed mesh for a backing material, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 4 This is a flowchart of a method for generating a randomly distributed mesh for backing material according to an embodiment of the present invention, as shown below. Figure 4 As shown, the process includes the following steps: Step S401: Obtain the target acoustic parameter set of the target backing material. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.
[0058] Step S402: Based on the target acoustic parameter set, determine the minimum surface spacing value of each particle in the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material.
[0059] Specifically, step S402 includes: Step S4021: Determine the minimum surface spacing value of each particle in the target backing material based on the upper limit of the working frequency band of the target acoustic parameters and the longitudinal wave velocity of the matrix.
[0060] In an alternative embodiment, the upper limit of the operating frequency band of the target acoustic parameter set can be incorporated. and matrix longitudinal wave velocity Calculate the minimum wavelength during operation. .
[0061] Furthermore, combined This allows us to determine the minimum surface spacing of each particle in the target backing material. Specifically, Pick to .
[0062] Step S4022: Based on the target acoustic attenuation spectrum in the target acoustic parameter set, the volume fraction and particle size distribution of each particle in the target backing material are determined by inversion using a multiple scattering model.
[0063] In one optional embodiment, the multiple scattering model is a concept widely used in physics, optics, computer graphics, and remote sensing. It refers to a mathematical model that describes a wave (such as a light wave or a particle wave) undergoing two or more scattering processes in a medium. In this embodiment, the model is an acoustic mathematical model describing the propagation and multiple reciprocating scattering of ultrasonic waves in a mixed medium of epoxy resin matrix and metal particles.
[0064] In one optional embodiment, using the target acoustic attenuation spectrum in the target acoustic parameter set as input, the multiple scattering model is used to back-calculate the overall proportion and size distribution parameters of the backing filler particles required from the desired acoustic attenuation performance, that is, to output the volume fraction and particle size distribution of various metal particles, providing data support for the subsequent construction of a three-dimensional target volume fraction field.
[0065] For example, the target acoustic attenuation spectrum is input into a multiple scattering model, and the theoretical acoustic attenuation spectrum after multiple scatterings of ultrasound between the matrix and metal particles is calculated using the model.
[0066] Furthermore, the difference between the theoretical attenuation spectrum output by the model and the target acoustic attenuation spectrum input is calculated at each frequency point to obtain the attenuation spectrum deviation value, so as to quantify the degree of matching between the current particle parameters and the design requirements.
[0067] Furthermore, the particle size distribution combination consisting of the overall volume fraction of particles and the proportion of particles in each grade is continuously adjusted, and the forward calculation and deviation solution are repeatedly performed to continuously reduce the error between the theoretical spectrum and the target spectrum until the attenuation spectrum deviation is less than the preset error threshold. At this point, the iteration is determined to be converged, and the corresponding overall volume fraction of particles and particle size distribution results are output.
[0068] Step S4023: Construct a three-dimensional target volume fraction field based on the acoustic impedance gradient requirements, volume fraction, and particle size distribution of the target acoustic parameters.
[0069] In one alternative embodiment, a three-dimensional spatial coordinate mesh covering the entire backing model can be divided according to the finite element simulation dimensions. Furthermore, each spatial mesh point corresponds to a local region and is used as the basic unit for assigning volume fraction values.
[0070] Furthermore, a conversion formula can be established based on the intrinsic acoustic impedance parameters of the particles and the matrix, and the target acoustic impedance values at each point in the gradient requirement can be converted into the local target particle volume fractions at the corresponding spatial locations.
[0071] Furthermore, along the defined gradient direction, values are assigned to each coordinate within the three-dimensional mesh domain, thereby achieving a smooth and continuous change in volume fraction within the gradient interval. Moreover, uniformly filling the inverted baseline total integral in gradient-free regions can accommodate the needs of both gradually changing and uniform regions.
[0072] Furthermore, the globally unified particle size distribution parameters are bound to the three-dimensional spatial field, so that each coordinate point in the field simultaneously carries the local particle volume fraction allowed at that location and the suitable particle size range.
[0073] Furthermore, by integrating all three-dimensional coordinates, corresponding local volume fractions, and binding particle size information, a three-dimensional target volume fraction field can be formed that can be directly invoked by the particle detection step. .
[0074] In one instance, such as Figure 5 As shown, a target volume fraction field and local statistical window obtained by the random distribution grid generation method for backing material according to an embodiment of the present invention are provided. The grid on the left represents the target acoustic impedance gradient, specifically reflecting the change of the target volume fraction field at different levels. The horizontal axis represents the grid layer number. In this example, the entire volume distribution field is divided into 200 layers. The vertical axis represents the target volume fraction on the corresponding grid layer. The diagram on the right represents a slice example of the actual layer 100 in this embodiment. Different shapes in the diagram represent different types of metal particles, and the gray background represents the matrix material. The volume fraction of all particle sets in the diagram on the right satisfies the target volume fraction of the corresponding level in the diagram on the left. The three-dimensional target volume fraction field determined in this step also provides a constraint basis for subsequent detection.
[0075] Step S403: Using multiple minimum surface spacing values and a three-dimensional target volume fraction field as constraints, the overlap, spacing, and local volume fraction of each particle in the target backing material are detected sequentially to obtain the material identification matrix. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.
[0076] Step S404: Generate the particle distribution mesh of the target backing material based on the material identification matrix. For details, please refer to [link to relevant documentation]. Figure 2Step S204 of the illustrated embodiment will not be described again here.
[0077] The method for generating a randomly distributed mesh for the backing material provided in this embodiment is based on the target acoustic parameter set. It calculates the minimum surface spacing of each particle using the upper limit of the working frequency band and the longitudinal wave velocity of the matrix. Based on the acoustic attenuation spectrum, it obtains the volume fraction and particle size distribution of various metal particles through back-calculation using a multiple scattering model. Based on the target acoustic impedance gradient, the volume distribution, and the particle size distribution of various metal particles, a three-dimensional target volume fraction field is constructed. This embodiment, through the acoustic parameter set and the multiple scattering model, achieves a target expectation-based inversion technique, and then uses the inversion results to guide the generation of the randomly distributed mesh for the backing material towards the target expectation.
[0078] This embodiment provides a method for generating a randomly distributed mesh for a backing material, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 6 This is a flowchart of a method for generating a randomly distributed mesh for backing material according to an embodiment of the present invention, as shown below. Figure 6 As shown, the process includes the following steps: Step S601: Obtain the target acoustic parameter set of the target backing material. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.
[0079] Step S602: Based on the target acoustic parameter set, determine the minimum surface spacing of each particle in the target backing material and the corresponding three-dimensional target volume fraction field of the target backing material. The distribution of the three-dimensional target volume fraction field satisfies the acoustic impedance gradient requirement. For details, please refer to... Figure 4 Step S402 of the illustrated embodiment will not be described again here.
[0080] Step S603: Using multiple minimum surface spacing values and a three-dimensional target volume fraction field as constraints, the overlap, spacing and local volume fraction of each particle in the target backing material are detected sequentially, and a material identification matrix is obtained. The material identification matrix includes the material identification of multiple particles that meet all detection requirements.
[0081] Specifically, step S603 includes: Step S6031: Select any target particle from among the particles, and determine whether there are other particles that overlap with the target particle.
[0082] In this step, each particle refers to all the metal particles in the substrate material to be placed onto the target backing material. This step selects any one target metal particle from all the metal particles to be placed and places it, ensuring that the placed target metal particle does not overlap with any already placed metal particles, so that subsequent detection and placement operations can be performed.
[0083] In some optional implementations, step S6031 above includes: Step a1: Construct spherical voxel masks for each particle in the target backing material and calculate the spherical voxel mask value for each particle.
[0084] In an optional embodiment, the spherical voxel mask represents a binary three-dimensional voxel matrix (binary matrix) used to digitally characterize the three-dimensional space occupied by a single spherical metal particle. It is a digital geometric characterization tool and contains only two values: 0 and 1. The value 1 represents that the voxel unit is covered by the particle sphere, and the value 0 represents that the voxel unit belongs to the matrix and is not occupied by the particle.
[0085] In an alternative embodiment, first construct a structure with a size of The 3D matrix mesh is such that every mesh element is a cube, and the side length h of the cube satisfies: ,in, Set the initial material identification matrix for the radius of the smallest particle among various metal particles. And set all elements to 0, representing a pure matrix.
[0086] Furthermore, spherical voxel masks are constructed for each type of metal particle. Specifically, for particles with a radius of... The particles, their spherical voxel mask It is a scale of A binary matrix of size , whose elements are defined by the following relation (1):
[0087] in, .
[0088] In one embodiment, such as Figure 7 As shown, this is a spherical voxel mask with a radius of R = 5 grids. The spherical voxel mask occupies the length of its diameter in all three dimensions, which is 10 grids, and is characterized as spherical.
[0089] Step a2: Select any target particle from among the particles, and randomly generate candidate coordinates of the target particle's center according to preset boundary conditions.
[0090] In an alternative embodiment, for any target particle Randomly generate candidate coordinates of its sphere center. Each component must satisfy the boundary conditions, as shown in the following relation (2):
[0091] in, For target particles The particle radius is used to ensure that any target particle is located inside the three-dimensional matrix mesh.
[0092] Step a3: Based on the candidate center coordinates, construct a bounding box matrix centered on the center of the target particle and calculate the bounding box matrix value.
[0093] In an alternative embodiment, candidate coordinates of the sphere center are extracted and calculated. centered bounding box matrix The following relation (3) is shown:
[0094] This step is based on the candidate coordinates of the sphere's center. With particle radius The bounding box matrix is calculated. .
[0095] Step a4: When the spherical voxel mask value and the bounding box matrix value of the target particle both meet the preset requirements, it is determined that there are no particles overlapping with the target particle in the bounding box matrix.
[0096] In an alternative embodiment, an index is set. If an index exists that makes and If an index does not exist, then an overlap is determined, and the process returns to step a2 to restart generating candidate coordinates for the sphere center; if no index exists that allows for this, then the process continues. and If no overlap occurs, proceed with the spacing detection operation according to the procedure.
[0097] Step S6032: When there are no other particles overlapping with the target particle, the surface distance between the target particle and any remaining particle is detected using multiple minimum surface distance values as constraints, and it is determined whether the surface distance between the target particle and any remaining particle meets the distance detection requirements.
[0098] In one alternative embodiment, after the overlap detection passes, the surface distance between the target particle and any remaining particle is further detected using multiple minimum surface distance values as constraints.
[0099] In some optional implementations, step S6032 above includes: Step b1: When there are no other particles overlapping with the target particle, calculate the Euclidean distance value of the center of any remaining particle.
[0100] In an optional embodiment, when there are no other particles overlapping with the target particle, the Euclidean distance between the center of the sphere and any remaining particle is first calculated, as shown in the following equation (4):
[0101] in, Let be the Euclidean distance between the center of the sphere and the center of any remaining sphere. Let the coordinates be the center coordinates of the target particle. Let be the coordinates of the center of any remaining particle. The Euclidean distance between the centers of the particles is calculated in this step. It can provide data support for subsequent calculations of the surface distance between the target particle and any remaining particles.
[0102] Step b2: Calculate the surface distance between the target particle and any remaining particle based on the multiple Euclidean distance values between the centers of the spheres.
[0103] In an optional embodiment, based on the calculated Euclidean distance value of the sphere's center The surface distance between the target particle and any remaining particle is calculated as shown in the following equation (5):
[0104] in, The surface distance value between the target particle and any remaining particle. Let be the particle radius of the target particle. The particle radius of any of the remaining selected particles.
[0105] In one instance, such as Figure 8 As shown, a particle spacing detection diagram is provided, where dashed circles represent target particles to be placed, and solid circles represent any remaining particles that have already been placed. The actual surface distance between the target particle and any remaining particle is shown in the figure. The intersection of the dashed and solid circles indicates that they have overlapped. Located in the overlapping area.
[0106] Step b3: When the surface distance between the target particle and any remaining particle is less than the minimum surface distance between the target particle, it is determined that the surface distance between the target particle and any remaining particle meets the distance detection requirements.
[0107] in, Represents the target particle to be placed Particle type, Represents any remaining particle that has been placed. Particle type, This represents the smallest surface spacing value among multiple minimum surface spacing values between two types of particles. If Less than If the spacing detection fails, return to step a2 to regenerate candidate coordinates for the sphere center; Greater than or equal to If the distance detection is successful, then the distance detection is considered passed.
[0108] Furthermore, this embodiment calculates the surface distance between the target particle and any remaining particle, and combines it with the minimum surface distance value from the previous step to ensure that the surface distance between any two particles meets the expected requirements. This avoids the occurrence of near-field scattering coupling effects caused by unreasonable particle distribution spacing in the target backing material, and improves the stability of the material's acoustic performance.
[0109] Step S6033: When the surface distance between the target particle and any remaining particle meets the distance detection requirements, the local volume fraction of the target particle is detected using the three-dimensional target volume fraction field as a constraint, and it is determined whether the local volume fraction of the target particle meets the local volume fraction detection requirements.
[0110] In some optional implementations, step S6033 includes: Step c1: When the surface distance between the target particle and any remaining particle meets the distance detection requirements, calculate the side length of the cubic bounding box centered on the center of the target particle.
[0111] In an optional embodiment, after the target particle passes the spacing detection, the side length of the cube bounding box centered on the center of the target particle is calculated. The cube bounding box is used to calculate the local volume fraction centered on the center of the target particle, which is the basis for the local volume fraction detection of the target particle, as shown in the following relationship (6):
[0112] in, Let be the side length of the bounding box of the cube. The coefficient is 2 to 4. The maximum particle radius is defined as . This step constructs a cubic bounding box and calculates its side length, defining the boundary for subsequent calculations of the local volume fraction within the cubic bounding box.
[0113] Step c2: Determine the total number of grids in the preset statistical window based on the side length.
[0114] In an optional embodiment, the side length of the cube bounding box obtained in step c1 is used as a reference. Determine the preset statistical window, which is the total number of grid cells in the cube bounding box. .
[0115] Step c3: Calculate the local volume fraction of the target particles within the preset statistical window based on the total number of grids.
[0116] In an alternative embodiment, based on the total number of grids The local volume fraction of the target particles within the preset statistical window is calculated as shown in the following equation (7):
[0117] in, To determine the local volume fraction of target particles within a preset statistical window, This represents the total number of grid cells in the cube's bounding box. This is an indicator function. If the result within the parentheses is yes, it outputs 1; if the result within the parentheses is no, it outputs 0. This is used to count the total number of occupied grid cells.
[0118] Step c4: Obtain the target volume fraction of the target particles in the three-dimensional target volume fraction field, and determine the target threshold based on the target volume fraction and the preset positive tolerance value.
[0119] In an optional embodiment, based on the location of the target particle to be placed, the target volume fraction at that location can be obtained using a three-dimensional target volume fraction field, and combined with a preset positive tolerance value, the target threshold can be determined, as shown in the following relationship (8):
[0120] in, The target threshold representing the volume fraction detection is... Target volume fraction, The preset positive tolerance limit is typically set to 0.02. This embodiment calculates the target threshold based on the target volume fraction and the preset positive tolerance limit, providing a basis for volume fraction detection.
[0121] Step c5: When the local volume fraction is greater than the target threshold, it is determined that the local volume fraction of the target particles meets the local volume fraction detection requirements.
[0122] In an optional embodiment, the magnitude of the local volume fraction and the target threshold are determined. If the local volume fraction... Greater than the preset threshold Right now If the local volume fraction is less than or equal to the preset threshold, the local volume fraction constraint is violated, the volume fraction detection is deemed unsuccessful, and the process returns to step a2 to regenerate candidate coordinates for the sphere center; if the local volume fraction is less than or equal to the preset threshold, the local volume fraction constraint is followed, the volume fraction detection is deemed successful, and the process continues according to the procedure.
[0123] Furthermore, in this embodiment, a statistical window is obtained by defining a cubic bounding box centered on the target particle, and the local volume fraction within the statistical window is calculated. By combining the obtained target volume fraction with the preset positive tolerance value, the local volume fraction is detected, ensuring that the random distribution grid of the backing material is arranged with the target volume fraction field as the guide, which has a certain degree of controllability.
[0124] Step S6034: When the local volume fraction of the target particle meets the local volume fraction detection requirements, return to the step of selecting any target particle that meets the requirements among the particles, and iterate repeatedly until multiple target particles whose local volume fraction meets the local volume fraction detection requirements are obtained.
[0125] In an optional embodiment, a placement operation is performed when the local volume fraction of the target particles meets the local volume fraction detection requirements. Further, for all particles in the mask that meet the requirements... Let the index be represented by the following relation (9):
[0126] in, The material identifier for the target particle is then determined. Returning to step a2, the process of generating candidate center coordinates is repeated iteratively until no target particles requiring placement are found, signifying that all particles have been placed.
[0127] Step S6035: Construct a material identification matrix based on multiple target particles.
[0128] In an optional embodiment, the material identification of the resulting plurality of target particles is performed. Integrate and construct the corresponding material identification matrix. .
[0129] Step S604: Generate the particle distribution mesh of the target backing material based on the material identification matrix. For details, please refer to [link to relevant documentation]. Figure 2 Step S204 of the illustrated embodiment will not be described again here.
[0130] In one example, a method for generating random distribution meshes for backing materials based on improved RSA is provided, including the following steps: Step S1: Obtain the target acoustic parameters, including the desired acoustic attenuation spectrum. Acoustic impedance gradient requirements and operating frequency band To obtain the physical properties of the matrix material and at least two types of metal particles, such as density and sound velocity.
[0131] Step S2: Using a multiple scattering model, the volume fraction of various particles is inferred from the target acoustic attenuation spectrum. and particle size distribution; based on the upper limit of the operating frequency band With matrix longitudinal wave velocity Determine the minimum surface spacing between particles , Pick to ,in .
[0132] Step S3: Define the three-dimensional target volume fraction field according to the acoustic impedance gradient requirement. Its spatial distribution should make the acoustic impedance estimated from the equivalent density approach the target value.
[0133] Step S4: Build a structure with the following dimensions. A three-dimensional matrix mesh, where the mesh element is a cube with a side length of... Satisfy: For the smallest particle radius ,have Set the initial material identification matrix. And set all elements to 0, representing a pure matrix.
[0134] Step S5: Pre-generate a spherical voxel mask for each type of particle. For a radius of... Particles (in terms of grid number), mask It is a scale of A binary matrix of size, whose elements are defined as shown in the above relation (1).
[0135] Step S6: Perform a constrained randomized sequential adsorption process. Construct a total sequence of all particles to be placed according to type and quantity. After randomly shuffling the sequence, perform the following sub-steps for each particle in the sequence: S6.1: For particles Randomly generate candidate coordinates of its sphere center. Each component must satisfy the boundary conditions, as shown in the above relation (2), in order to ensure that the entire sphere is located inside the matrix.
[0136] S6.2 Overlap Detection: Extracting... centered bounding box matrix As shown in relation (3) above, if an index exists... Make and If an overlap occurs, the process returns to step S6.1 to regenerate candidate coordinates.
[0137] S6.3 Spacing Detection: For each particle in the placed particle list The Euclidean distance between the center of the sphere is calculated according to the above relationship (4), and the distance between the candidate particle and the sphere is calculated according to the above relationship (5). Surface spacing, if for any ,exist If the spacing constraint is violated, the process returns to step S6.1.
[0138] S6.4 Local Volume Fraction Detection: Using the center of the candidate particle Define the statistical window centered on the data. The side length of its cube bounding box is as shown in the above relation (6), and the current particle volume fraction in the window is defined as shown in the above relation (7). If the judgment condition of combining the above relation (8) and step c5 is met, it is determined that the local volume fraction constraint is violated, and then return to step S6.1.
[0139] S6.5 If all the above checks pass, perform the following placement operation: For all masks that meet the requirements... Based on the above relation (9), the index of the particles is obtained. Material labeling At the same time, the particles The center of the ball ,radius and type Record it in the list of placed particles.
[0140] Step S7: Repeat step S6 until all particles have been placed, and output the material identification matrix. If a particle fails to find a valid location within the preset maximum number of attempts, a prompt message will be issued, and the tolerance can be increased as needed. Alternatively, the target field can be adjusted and the process can be repeated.
[0141] As a preferred embodiment, the method for determining the minimum surface spacing in step S2 further includes: calculating the normalized scattering cross section using a two-sphere scattering model. , to increase surface spacing Gradually increase, when When the rate of change relative to the sum of the scattering cross sections of two isolated particles is less than 5%, the spacing at this point is recorded as... and take .
[0142] As a preferred option, the target volume fraction field in step S3 , is defined as a function that varies linearly along a certain spatial direction (e.g., the Z direction), as shown in the following relation (10):
[0143] in and Let be the expected total integrals for the bottom and top surfaces, respectively. The desired number of volume fraction layers is set. This is the actual floor number of the target location. Or, It can also be generated by a three-dimensional random field generator, with a specified correlation length. and root mean square deviation A stationary Gaussian random field is used to simulate the random inhomogeneities during the curing process.
[0144] As a preferred approach, the calculation of the current volume fraction within the window in step S6.4 can be incrementally updated using a block accumulator. Specifically, the matrix mesh is pre-divided into several non-overlapping coarse blocks, and each coarse block maintains a counter for the number of occupied meshes within it. After placing the particles, only the counter values of the coarse blocks affected by the mask are updated. The current total integral is then calculated. At this time, the count value of the coarse blocks covered by the window is directly superimposed and divided by the total number of window grids, thus avoiding grid-by-grid traversal.
[0145] Based on the above examples, a specific embodiment of the method for generating random distribution grids for backing materials using improved RSA is provided. It should be understood that the embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.
[0146] Example 1: Gradient impedance modeling for two types of metal particle / epoxy resin backing: The objective of this embodiment is to design a medical ultrasound transducer backing material with an operating frequency of 2-5MHz, requiring its acoustic impedance to gradually change from 8MRayl on the bottom surface to 4MRayl on the top surface, while the average longitudinal wave attenuation within the operating frequency band is not less than 15dB / cm.
[0147] Step S1: Material parameters and target definition.
[0148] Matrix epoxy resin: density Longitudinal wave speed .
[0149] Metal Particle 1 (Tungsten Powder): Density Longitudinal wave speed Select particle radius .
[0150] Metal particles 2 (copper powder): density Longitudinal wave speed Select particle radius .
[0151] Take the grid side length The total number of matrix meshes is Corresponding physical dimensions .at this time Corresponding to 3 grids, For two grids, satisfying Precision requirements. Upper limit of operating frequency band. .
[0152] Step S2: Infer particle parameters based on attenuation spectrum.
[0153] Under the independent scattering approximation, the longitudinal wave attenuation coefficient is given by equation (11):
[0154] in, For the first Number density of particles, The scattering cross section is a single-sided particle. The Allegra-Hawley model is used to accurately solve the problem under known phase densities and sound velocities. An optimization problem is established with the attenuation index as the objective, as shown in equation (12):
[0155] The volume fraction of tungsten powder was obtained by iterative solution. copper powder volume fraction Calculate the initial value for the minimum surface spacing: the shortest wavelength of the longitudinal wave in the matrix. ,Pick Based on this, a two-sphere scattering model is used for correction. For tungsten-tungsten particle combinations, when the surface spacing increases to... When (corresponding to 2 grids), the coupling deviation of the normalized scattering cross section has been reduced to below 5%; and for tungsten-copper and copper-copper combinations, the coupling deviation in the contact state is within an acceptable range. Therefore, the minimum surface spacing is set to: tungsten-tungsten particles (2 grids), tungsten-copper particles (Contact permitted), copper-copper particles (Touch permitted).
[0156] Step S3: Define the target volume fraction field.
[0157] Based on the requirement that the acoustic impedance gradually changes linearly along the Z direction, the impedance gradient is mapped to a volume fraction gradient using the equivalent density mixing law. The target volume fraction field is defined as a linear function along the Z direction, as shown in equation (13):
[0158] The venue indicated that in the The expected upper limit of the total metal integral at the horizontal slice of the layer.
[0159] The expected value of the bottom surface is 36%, the expected value of the top surface is 18%, and the overall average is approximately 27%, which is related to the total integral of the two types of particles. The two types of particles are matched. The local volume ratio of the two types of particles remains at tungsten:copper = 2:1.
[0160] Step S4: Initialize the mesh and mask.
[0161] Create a size of integer matrix The initial elements are all 0. A spherical mask is pre-generated according to equation (1): For tungsten particles (R=3): Mask for The matrix, centered at ,satisfy The position is set to 1. For copper particles (R=2): mask for The matrix, centered at ,satisfy The position is set to 1.
[0162] Step S5: Construct and shuffle the particle sequence.
[0163] The required number of particles is calculated based on the volume fraction and the total matrix volume. Considering the slight volume difference between voxelized spheres and ideal spheres, and the potential reduction in effective placement space due to subsequent spacing and local volume fraction constraints, the placement target is set at 95% of the theoretical number in actual operation: tungsten particles. One, copper particle One. Generates containing Particles identified as 1 and A sequence of particles, each labeled 2, is randomly shuffled to obtain the final placement order.
[0164] Step S6: Constrained random sequential adsorption.
[0165] Set the maximum number of attempts per particle to 5000, and repeat the process for each particle in the sequence: 1. Generate the coordinates c of the sphere center uniformly and randomly within the effective range according to step S6.1; 2. Perform overlap detection as per step S6.2. If the detection fails, exit the current attempt. 3. Perform spacing detection according to step S6.3, comparing the surface spacing with each placed particle. If the surface spacing between the candidate particle and any placed tungsten particle is less than 2 grids, it is deemed invalid, and contact with the copper particle is allowed; 4. Perform local volume fraction detection according to step S6.4. Take half the width of the window. Window side length .calculate and In comparison, the local volume fraction tolerance in this embodiment... Take 0.05, if there is Then it is deemed invalid; 5. If all passes, proceed to step S6.5 to assign the mesh within the mask area the material identifier of the particle, and record the particle information in the placed list.
[0166] Step S7: Output and finite element simulation verification.
[0167] After all particles have been placed, output the material identification matrix. As attached Figure 9 As shown in the figure, this diagram illustrates a three-dimensional slice of the backing material. Statistical analysis reveals that the actual generated tungsten powder volume fraction is approximately 17.8%, and the copper powder volume fraction is approximately 9.1%, with a deviation from the design values of less than 0.5%. The matrix... Assign acoustic parameters to epoxy resin, tungsten, and copper according to their respective identifiers. Generate element sets in commercial finite element software, apply periodic boundary conditions along the Z-axis, input broadband longitudinal wave pulses for transient analysis, extract transmission and reflection signals, and calculate the acoustic attenuation coefficient and acoustic impedance. (See attached...) Figure 10 As shown, the dashed line represents the theoretical attenuation spectrum calculated using a multiple scattering model based on the actual volume fraction of the generated structure, while the solid line represents the attenuation spectrum obtained from finite element simulation of the generated structure. The two curves almost completely overlap throughout the entire operating frequency band, with an average longitudinal wave attenuation of 15.2 dB / cm in the 2–5 MHz band, meeting the design requirement of no less than 15 dB / cm, and the acoustic impedance decreasing approximately linearly along the Z-axis. This result fully verifies that the method of this invention can generate a backing material microstructure that highly matches the designed acoustic performance.
[0168] This example provides a method for generating random distribution meshes for backing materials based on improved RSA, which has the following advantages: Controllable acoustic response: By forcibly applying a minimum spacing constraint that depends on frequency, the near-field scattering coupling between particles is effectively reduced, making the acoustic attenuation spectrum obtained by simulation highly consistent with the target spectrum, and the attenuation peak frequency deviation can be controlled within 5%.
[0169] On-demand generation of macroscopic gradient structures: By constraining the placement of local particles using the target volume fractional field, a backing material model with a preset acoustic impedance gradient can be directly generated, providing a direct modeling method for the design of gradient matching layers.
[0170] One-time generation reduces iterations: By internalizing acoustic design principles into the microstructure generation algorithm, the first generated model can approximate acoustic performance indicators, significantly shortening the trial-and-error cycle of material development.
[0171] High versatility: This method can be extended to the synchronous random generation of various fillers and multi-level pores, and is also applicable to the modeling of random media containing scattering particles, such as electromagnetic absorbing materials and thermal interface materials.
[0172] This embodiment also provides a device for generating a random distribution grid of backing material, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.
[0173] This embodiment provides a device for generating a randomly distributed grid for backing material, such as... Figure 11 As shown, the device includes: The parameter acquisition module 1101 is used to acquire the target acoustic parameter set of the target backing material, and to acquire the substrate material parameters and metal particle parameters.
[0174] The data processing module 1102 is used to determine the minimum surface spacing value of each particle of the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material based on the acquired target acoustic parameter set.
[0175] The constraint detection module 1103 is used to constrain the placement of metal particles by detecting the overlap, spacing and local volume fraction of each particle.
[0176] The verification output module 1104 is used to record, form and output the placed material identification matrix, and can perform simulation verification on the material identification matrix.
[0177] The backing material random distribution grid generation device provided in this embodiment of the invention can execute the backing material random distribution grid generation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method. Further functional descriptions of the above modules are the same as in the corresponding embodiments described above, and will not be repeated here.
[0178] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.
[0179] The following is a detailed reference. Figure 12 The diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 1201, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1202 or a program loaded from memory 1208 into random access memory (RAM) 1203. The RAM 1203 also stores various programs and data required for the operation of the electronic device. The processor 1201, ROM 1202, and RAM 1203 are interconnected via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.
[0180] Typically, the following devices can be connected to I / O interface 1205: input devices 1206 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 1207 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 1208 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1209. Communication device 1209 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 12 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.
[0181] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 1209, or installed from a memory 1208, or installed from a ROM 1202. When the computer program is executed by the processor 1201, it performs the functions defined in the backing material random distribution mesh generation method of the embodiments of the present invention.
[0182] Figure 12 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.
[0183] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the random distribution grid generation method for backing material shown in the above embodiments is implemented.
[0184] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.
[0185] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.
Claims
1. A method for generating a randomly distributed mesh for a backing material, characterized in that, The method includes: Obtain the target acoustic parameter set of the target backing material; Based on the target acoustic parameter set, the minimum surface spacing value of each particle in the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material are determined, and the distribution of the three-dimensional target volume fraction field satisfies the acoustic impedance gradient requirement. Constrained by multiple minimum surface spacing values and the three-dimensional target volume fraction field, the overlap, spacing and local volume fraction of each particle in the target backing material are detected sequentially, and a material identification matrix is obtained. The material identification matrix includes the material identification of multiple particles that meet all detection requirements. Based on the material identification matrix, a particle distribution grid of the target backing material is generated.
2. The method according to claim 1, characterized in that, Based on the target acoustic parameter set, determine the minimum surface spacing value of each particle in the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material, including: The minimum surface spacing value of each particle in the target backing material is determined based on the upper limit of the working frequency band of the target acoustic parameters and the longitudinal wave velocity of the matrix. Based on the target acoustic attenuation spectrum in the target acoustic parameter set, the volume fraction and particle size distribution of each particle in the target backing material are determined by inversion using a multiple scattering model. The three-dimensional target volume fraction field is constructed based on the acoustic impedance gradient requirement of the target acoustic parameter set, the volume fraction, and the particle size distribution.
3. The method according to claim 1, characterized in that, Constrained by multiple minimum surface spacing values and the three-dimensional target volume fraction field, the overlap, spacing, and local volume fraction of each particle in the target backing material are detected sequentially, resulting in a material identification matrix, including: Select any particle that satisfies the target particle from among the particles, and determine whether there are other particles that overlap with the target particle; When there are no other particles overlapping with the target particle, the surface distance between the target particle and any remaining particle is detected using the multiple minimum surface distance values as constraints, and it is determined whether the surface distance between the target particle and any remaining particle meets the distance detection requirements. When the surface distance between the target particle and any remaining particle meets the distance detection requirement, the local volume fraction of the target particle is detected using the three-dimensional target volume fraction field as a constraint, and it is determined whether the local volume fraction of the target particle meets the local volume fraction detection requirement. When the local volume fraction of the target particle meets the local volume fraction detection requirement, return to the step of selecting any target particle that meets the requirement among all particles, and iterate repeatedly until multiple target particles whose local volume fraction meets the local volume fraction detection requirement are obtained. The material identification matrix is constructed based on the multiple target particles.
4. The method according to claim 3, characterized in that, Selecting any target particle from among the particles and determining whether there are other particles overlapping with the target particle includes: Construct a spherical voxel mask for each particle in the target backing material and calculate the spherical voxel mask value for each particle. The spherical voxel mask is a binary matrix. Select any target particle from among the particles, and randomly generate candidate coordinates of the center of the target particle according to preset boundary conditions; Based on the candidate center coordinates, construct a bounding box matrix centered on the center of the target particle and calculate the bounding box matrix value; When the spherical voxel mask value and the bounding box matrix value of the target particle both meet the preset requirements, it is determined that there are no particles overlapping with the target particle in the bounding box matrix.
5. The method according to claim 3, characterized in that, When no other particles overlap with the target particle, the surface distance between the target particle and any remaining particle is detected using the plurality of minimum surface distance values as constraints, and it is determined whether the surface distance between the target particle and any remaining particle meets the distance detection requirements, including: If there are no other particles overlapping with the target particle, calculate the Euclidean distance between the centers of any remaining particles; Calculate the surface distance between the target particle and any remaining particle based on multiple Euclidean distance values between the centers of the spheres; When the surface distance between the target particle and any remaining particle is less than the minimum surface distance between the target particle, it is determined that the surface distance between the target particle and any remaining particle meets the distance detection requirement.
6. The method according to claim 3, characterized in that, When the surface distances between the target particle and any remaining particle meet the distance detection requirements, the local volume fraction of the target particle is detected using the three-dimensional target volume fraction field as a constraint, and it is determined whether the local volume fraction of the target particle meets the local volume fraction detection requirements, including: When the surface distance between the target particle and any remaining particle meets the distance detection requirement, calculate the side length of the cubic bounding box centered on the center of the target particle. Based on the side length, determine the total number of grids in the preset statistical window; Based on the total number of grid cells, calculate the local volume fraction of the target particles within the preset statistical window; The target volume fraction of the target particles is obtained in the three-dimensional target volume fraction field, and the target threshold is determined based on the target volume fraction and a preset positive tolerance value. When the local volume fraction is greater than the target threshold, it is determined that the local volume fraction of the target particle meets the local volume fraction detection requirement.
7. The method according to claim 6, characterized in that, The method further includes: If no valid location is found in the target particle within the preset maximum number of attempts, the preset positive tolerance value or the three-dimensional target volume fraction field is adjusted. The absence of a valid location indicates that any detection result of the overlap, spacing, or local volume fraction of the target particles does not meet the requirements.
8. The method according to claim 1, characterized in that, The method further includes: Obtain the scattering cross section of each particle in the target backing material, and calculate the target scattering cross section of any two particles in the target backing material, wherein the target scattering cross section is the sum of the scattering cross sections of any two particles; The surface spacing values of each particle in the target backing material are gradually increased, and the normalized scattering cross section of any two particles in the target backing material is calculated using a two-sphere scattering model. Calculate the rate of change of the scattering cross section of any two particles based on the normalized scattering cross section and the target scattering cross section; When the rate of change of the scattering cross section is less than a preset threshold, the surface spacing value of any two particles is obtained, and the minimum surface spacing value of each particle in the target backing material is calculated.
9. A device for generating a randomly distributed grid for a backing material, characterized in that, The device includes: The parameter acquisition module is used to acquire the target acoustic parameter set of the target backing material, and to acquire the substrate material parameters and metal particle parameters. The data processing module is used to determine the minimum surface spacing value of each particle of the target backing material and the three-dimensional target volume fraction field corresponding to the target backing material based on the acquired target acoustic parameter set. The constraint detection module is used to constrain the placement of metal particles by detecting the overlap, spacing and local volume fraction of each particle. The verification output module is used to record, form and output the placed material identification matrix, and can perform simulation verification on the material identification matrix.
10. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the method for generating a random distribution grid for backing material as described in any one of claims 1 to 8.
