Anisotropic substrate mechanical property parameter multi-channel fusion method

Abstract

The invention relates to the technical field of composite substrate material mechanical modeling simulation, and provides an anisotropic substrate mechanical characteristic parameter multi-channel fusion method which comprises the following steps: constructing an anisotropic mechanical parameter library of a base material; mapping substrate bit pixels with material indexes in the anisotropic mechanical parameter library to obtain a material index recording matrix; in combination with the material index record matrix and an anisotropic mechanical parameter library, channel layers corresponding to all mechanical parameters of the substrate are obtained one by one and serve as all mechanical parameter channels of the substrate bitmap; and integrating the mechanical parameter channels of the substrate bitmap to generate multi-channel bitmap tensor data of the substrate. According to the method, information fusion of various mechanical parameters of the material is realized by using the pixels at the same position, the deep convolutional neural network can conveniently learn and extract equivalent material characteristics under the combined action of different mechanical parameter information, and the generalization ability of equivalent mechanical characteristics of anisotropic materials is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical modeling and simulation of composite substrate materials, and in particular to a multi-channel fusion method for mechanical characteristic parameters of anisotropic substrates. Background Art

[0002] In the thermal / mechanical simulation of composite substrates, in order to solve the problem of low efficiency or even infeasibility of cross-scale simulation caused by the fact that the substrate plane dimensions and thickness, as well as the characteristic line width and spacing dimensions span multiple orders of magnitude (usually 4 to 5 orders of magnitude), the industry mainly adopts the homogenized material equivalence method to avoid fine meshing of the microstructure, thereby greatly reducing the scale of finite element simulation units and nodes. For the prediction of material parameters of the mesoscopic representative volume element (RVE) used in the homogenization equivalent method, the papers "TSV wafer warpage simulation by machine learning-based anisotropic equivalent modeling method", "An RDL Modeling and Thermo-Mechanical Simulation Method of 2.5D / 3D Advanced Package Considering the Layout Impact Based on Machine Learning", and "Utilizing CNN to predict homogeneous thermo-mechanical properties of conductive layers for reliability numerical analysis in electronics" provide an equivalent division of a thin substrate into multiple grid blocks and pixelate the grid blocks into binary images to represent the via distribution information or metal copper ratio in the grid blocks, which is then used for fully connected neural network modeling.

[0003] Patent 202211088300.0 discloses a batch calculation method for the equivalent thermal conductivity of the wiring area of ​​integrated circuit products. The circuit image is binarized and converted into a two-dimensional array containing position and pixel information. An equivalent thermal resistance network and weighted average processing are used to calculate the anisotropic equivalent thermal conductivity of the area corresponding to the image.

[0004] Patent 202111602122.4 discloses a simulation method for the press-forming process of a printed circuit board. By partitioning the circuit board and identifying the copper foil represented by the binary image information of each partition or numerically calculating the corresponding equivalent performance parameters, it achieves high-precision and efficient automatic modeling and simulation analysis of the press-forming process of complex, large-size and multi-layer PCBs.

[0005] Patent 202211132414.0 discloses a method for calculating the equivalent performance parameters of the wiring area of ​​an integrated circuit product. It uses finite element analysis to solve the equivalent performance parameters of the microscopic model of the substrate circuit diagram partition. Different colors are used to distinguish different materials in the partition, but it does not involve the information fusion representation of multiple material parameters and anisotropic parameters.

[0006] Patent 201810410242.6 discloses a method for predicting the macroscopic properties of short fiber reinforced composite materials based on deep learning, which converts the fiber distribution image into a binary image containing fibers and matrix resin, and then establishes a deep learning agent model.

[0007] Patent US18102669 discloses a material structure generation and analysis method based on material parameters and a machine learning model. This method uses voxels to encode material proportions or structures, thereby providing input to the machine learning model. However, it does not involve the fusion of information on multiple material parameters and anisotropy parameters.

[0008] Existing methods do not involve the embedding of material anisotropic parameter information and model modeling prediction, resulting in research that is obviously inconsistent with reality, affecting the accuracy of using homogenization methods to simplify simulations, and unable to achieve modeling agency of equivalent mechanical properties of different materials through the same deep convolutional neural network model. Summary of the Invention

[0009] The present invention aims to provide a multi-channel fusion method for the mechanical property parameters of anisotropic substrates to solve the problems that previous solutions are only used for isotropic substrate materials, fail to consider the anisotropy of the materials, and lack the embedding of the material mechanical parameter information. The image data input to the model cannot fuse multiple mechanical parameter information, resulting in poor model universality and inability to generalize the characterization of substrate materials with different mechanical properties, and cannot model the equivalence of anisotropic materials.

[0010] The present invention provides a multi-channel fusion method for anisotropic substrate mechanical property parameters, comprising:

[0011] Construct anisotropic mechanical parameter library of matrix materials;

[0012] Mapping substrate bitmap pixels to material indices in the anisotropic mechanical parameter library to obtain a material index record matrix;

[0013] Combining the material index record matrix and the anisotropic mechanical parameter library, obtaining channel layers corresponding to the mechanical parameters of the substrate one by one as the mechanical parameter channels of the substrate bitmap;

[0014] The mechanical parameter channels of the substrate bitmap are integrated to generate substrate multi-channel bitmap tensor data.

[0015] In some embodiments, the anisotropic mechanical parameter library of the matrix material is constructed, comprising:

[0016] Constructing multiple unit matrices corresponding to each material, where the unit matrices are used to store various anisotropic mechanical parameters corresponding to each material;

[0017] All element matrices of a material are grouped together and the corresponding material index is established to construct the anisotropic mechanical parameter library of the matrix material.

[0018] In some embodiments, in the anisotropic mechanical parameter library of the matrix material:

[0019] The number of rows and columns of each unit matrix is ​​the same as the number of pixels in height and width of the substrate bitmap used;

[0020] The number of unit matrices corresponding to each matrix material is the same as the number of anisotropic mechanical parameters;

[0021] The element assignment of the unit matrix is ​​based on the corresponding mechanical parameters. The minimum and maximum values ​​of the mechanical parameters in the material are often linearly normalized.

[0022] In some embodiments, if the anisotropic mechanical parameter library includes an isotropic matrix material, the number of unit matrices thereof must be consistent with the number of unit matrices of the anisotropic matrix material, and the unit matrices representing material properties in different directions have the same mechanical parameter values.

[0023] In some embodiments, mapping the substrate bitmap pixels to the material indexes in the anisotropic mechanical parameter library to obtain a material index record matrix includes:

[0024] The base material corresponding to the position area of ​​the substrate bitmap pixel is filled with the material index in the anisotropic mechanical parameter library to obtain a material index record matrix containing the material index.

[0025] In some embodiments, filling the substrate bitmap pixels with the base materials corresponding to the position areas as material indexes in the anisotropic mechanical parameter library includes:

[0026] Matrix logic operation is used to extract the corresponding logic matrix for each type of base material in the substrate bitmap. The number of logic matrices is the same as the number of materials in the substrate bitmap, and the size of the rows and columns of the logic matrix is ​​consistent with the pixels of the substrate bitmap. If the value of the logic matrix element is "true", it means that the material of the corresponding pixel in the substrate bitmap is the material represented by the logic matrix. If the value of the logic matrix element is "false", it means that the material is not represented by the logic matrix.

[0027] Create a new material index record matrix. If the value of the element in the logic matrix is ​​"true", fill in the material index corresponding to the logic matrix at the same coordinate position in the material index record matrix.

[0028] In some embodiments, each mechanical parameter channel of the substrate bitmap is represented by a two-dimensional matrix, and the value of each element in the two-dimensional matrix is ​​assigned to the value of the unit matrix element of the corresponding material according to the type and direction of the mechanical parameter represented by the channel.

[0029] In some embodiments, the method for assigning values ​​to each element in the two-dimensional matrix includes:

[0030] The material index record matrix and the material index are subjected to matrix logic operation to obtain a material index mask matrix, and then the material index mask matrix is ​​multiplied by the unit matrix corresponding to the material index in the anisotropic mechanical parameter library according to the corresponding elements.

[0031] In some embodiments, a three-dimensional tensor is used to integrate the mechanical parameter channels of the substrate bitmap; wherein the dimensions of the three-dimensional tensor represent the height direction pixels, width direction pixels, and channel number of the substrate bitmap respectively.

[0032] In some embodiments, a four-dimensional tensor is used to integrate the mechanical parameter channels of the substrate bitmap; wherein, a three-dimensional tensor is first used to integrate the mechanical parameter channels of the substrate bitmap, and then the three-dimensional tensors of multiple substrate bitmaps are organized, and an additional fourth dimension is added as the number of substrate bitmaps, which is used to index and distinguish each substrate bitmap.

[0033] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0034] 1. The present invention encodes the mechanical parameters of different types and directions of anisotropic substrates into a channel of the image, so that one pixel point contains multiple mechanical parameter information at the same time, and then constructs a multi-channel tensor or multi-channel image data that integrates the anisotropic material parameters of the substrate. It realizes the embedding of information on multiple mechanical parameters of the material using pixels at the same position, facilitates deep convolutional neural networks to learn and extract equivalent material properties under the joint action of different mechanical parameter information, and improves the generalization ability of equivalent mechanical properties of anisotropic materials.

[0035] 2. Based on the present invention, composite substrates of different materials and anisotropic parameters of materials can be represented and fused through normalized format data, and by expanding or reducing the number of channels, the fusion and adaptation of characteristic parameter information of different materials can be achieved, and isotropic substrate materials can be compatible, thereby achieving the effect of modeling multiple material agents through the same model. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flowchart of a multi-channel fusion method for anisotropic substrate mechanical property parameters provided by a preferred embodiment of the present invention.

[0037] Figure 2 Schematic diagram of the anisotropic mechanical parameter library of the matrix material in a preferred embodiment of the present invention.

[0038] Figure 3 Schematic diagram of substrate bitmap pixels and corresponding material indexes in a preferred embodiment of the present invention.

[0039] Figure 4 Schematic diagram of various mechanical parameter channels in a preferred embodiment of the present invention.

[0040] Figure 5 Schematic diagram of a three-dimensional tensor of anisotropic mechanical parameters of a fusion material and a four-dimensional tensor of multiple substrates in a preferred embodiment of the present invention.

[0041] Figure 6 The present invention provides a schematic structural diagram of an electronic device according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0043] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are intended to fall within the scope of protection of the present invention.

[0044] Example

[0045] like Figure 1 As shown, this embodiment proposes a multi-channel fusion method for anisotropic substrate mechanical property parameters, including:

[0046] S1: Construct anisotropic mechanical parameter library of matrix materials.

[0047] For commonly used typical matrix materials, multiple unit matrices corresponding to each material are constructed. The unit matrices are used to store various anisotropic mechanical parameters corresponding to each material, such as the elastic modulus Ex, Ey, Ez in the x / y / z directions, the shear modulus Gxy, Gxz, Gyz, and the thermal expansion coefficient CTEx, CTEy, CTEz;

[0048] All element matrices of a material are grouped together and the corresponding material index is established to construct the anisotropic mechanical parameter library of the matrix material, such as Figure 2 As shown, it is convenient for subsequent calls;

[0049] The number of rows and columns of each unit matrix is ​​the same as the number of pixels in height and width of the substrate bitmap used, and is usually an N×N square matrix;

[0050] Among them, the number of unit matrices corresponding to each matrix material is the same as the number of anisotropic mechanical parameters;

[0051] Among them, the element assignment of the unit matrix is ​​the result Pn of linear normalization of the minimum value Pmin and the maximum value Pmax of the mechanical parameter in the common material based on the corresponding mechanical parameter P. The normalization formula is as follows:

[0052] Pn=(P-Pmin) / (Pmax-Pmin)

[0053] In particular, if the anisotropic mechanical parameter library in this embodiment contains an isotropic matrix material (such as copper Cu), the number of its unit matrices must be consistent with the number of unit matrices of the anisotropic matrix material (such as resin ABF12, AF41), and the unit matrices representing the material properties in different directions have the same mechanical parameter values, such as the same elastic modulus values ​​in the x / y / z directions.

[0054] S2: Mapping the substrate bitmap pixels with the material indexes in the anisotropic mechanical parameter library to obtain a material index record matrix.

[0055] Specifically, the substrate bitmap pixels (pixels in the substrate layout in bitmap format) are filled with the material index in the anisotropic mechanical parameter library according to the base material corresponding to the position area, and a material index record matrix containing the material index is obtained. Figure 3 As shown, the base material corresponding to the pixel in the D1 area is copper Cu, so the corresponding pixel is mapped as Figure 2 Similarly, the pixels corresponding to the D2 area are mapped to material index 2, and the pixels corresponding to the D3 area are mapped to material index 3.

[0056] Preferably, to achieve efficient processing of substrate bitmaps, matrix logic operations are used to extract the corresponding logical matrix for each type of base material in the substrate bitmap. The number of logical matrices is the same as the number of base materials, and the row and column size of the logical matrix is ​​consistent with the pixels in the substrate bitmap. A "true" value in the logical matrix element indicates that the material of the corresponding pixel in the substrate bitmap is the material represented by the logical matrix, while a "false" value indicates that the material is not represented by the logical matrix. Furthermore, a new material index record matrix is ​​created. If the value of the element in the logical matrix is ​​"true", the material index corresponding to the logical matrix is ​​filled in at the same coordinate position in the material index record matrix. This process uses matrix broadcast multiplication in the computer code implementation, that is, multiplying the logical matrix by the material index to achieve faster speed.

[0057] S3: combining the material index record matrix and the anisotropic mechanical parameter library, obtaining channel layers corresponding to the mechanical parameters of the substrate one by one as the mechanical parameter channels of the substrate bitmap.

[0058] Each mechanical parameter channel of the substrate bitmap is used to represent different mechanical parameter information of the substrate, and the channel can be named using the mechanical parameter. Wherein, each mechanical parameter channel of the substrate bitmap is represented by a two-dimensional matrix, and the value of each element in the two-dimensional matrix is ​​assigned to the value of the unit matrix element of the corresponding material according to the type and direction of the mechanical parameter represented by the channel. Preferably, in order to realize the rapid assignment of the channel layer, the material index record matrix and the material index are subjected to matrix logic operation to obtain a material index mask matrix, and then the material index mask matrix is ​​multiplied by the corresponding element with the unit matrix of the corresponding material index in the anisotropic mechanical parameter library.

[0059] like Figure 4 As shown, if the pixel value of the D1 area in the Ex channel is multiplied by the corresponding elements mentioned above, it corresponds to the normalized value of the elastic modulus in the x direction of copper, the pixel value of the D2 area corresponds to the normalized value of the elastic modulus in the x direction of resin ABF12, and the pixel value of the D3 area corresponds to the normalized value of the elastic modulus in the x direction of resin ABF41.

[0060] S4: Integrate the mechanical parameter channels of the substrate bitmap to generate substrate multi-channel bitmap tensor data.

[0061] In some embodiments, a three-dimensional tensor is used to integrate the mechanical parameter channels of the substrate bitmap; wherein the dimensions of the three-dimensional tensor represent the height direction pixels, width direction pixels, and number of channels of the substrate bitmap, respectively.

[0062] In some schemes, a four-dimensional tensor is used to integrate the various mechanical parameter channels of the substrate bitmap; wherein, a three-dimensional tensor is first used to integrate the various mechanical parameter channels of the substrate bitmap, and then the three-dimensional tensors of multiple substrate bitmaps are organized, and an additional fourth dimension is added as the number of substrate bitmaps, which is used to index and distinguish each substrate bitmap.

[0063] like Figure 5 As shown in the figure, the size of the three-dimensional tensor X is N×N×C, where C is the number of channels, which is the same as the number of various isotropic mechanical parameters that the material should encode. To facilitate the use of deep convolutional neural networks, multiple substrate bitmaps are usually organized into a four-dimensional tensor T with a size of N×N×C×S, where S is the number of substrate bitmaps.

[0064] Based on the same technical concept, an embodiment of the present invention also provides an electronic device that can implement the multi-channel fusion method process of anisotropic substrate mechanical characteristic parameters provided in the above embodiment of the present invention. In one embodiment, the electronic device can be a server, or a terminal device or other electronic device. Figure 6 As shown, the electronic device may include:

[0065] At least one processor, and a memory connected to the at least one processor. The embodiment of the present invention does not limit the specific connection medium between the processor and the memory. Figure 6 The example in this article is that the processor and memory are connected via a bus. Figure 6 The connections between the other components are shown in bold lines, which are only for illustration and not intended to be limiting. The bus can be divided into address bus, data bus, control bus, etc. Figure 6 The processor is represented by a single thick line, but this does not mean that there is only one bus or only one type of bus. Alternatively, the processor can also be called a controller, without any limitation on the name.

[0066] In an embodiment of the present invention, the memory stores instructions that can be executed by at least one processor. The at least one processor can execute the multi-channel fusion method of anisotropic substrate mechanical characteristic parameters discussed above by executing the instructions stored in the memory. The processor can implement Figure 6 The functions of each module in the device shown.

[0067] Among them, the processor is the control center of the device, which can use various interfaces and lines to connect the various parts of the entire control device, and monitor the device as a whole by running or executing instructions stored in the memory and calling data stored in the memory, the various functions of the device and processing data.

[0068] In an optional design, the processor may include one or more processing units, and the processor may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, and the modem processor primarily processes wireless communications. It is understood that the modem processor may not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip, or in some embodiments, they may be implemented on separate chips.

[0069] The processor can be a general-purpose processor, such as a CPU, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the multi-channel fusion method for anisotropic substrate mechanical property parameters disclosed in the embodiments of the present invention can be directly implemented as a hardware processor, or can be implemented using a combination of hardware and software modules in the processor.

[0070] As a non-volatile computer-readable storage medium, memory can be used to store non-volatile software programs, non-volatile computer executable programs and modules. Memory can include at least one type of storage medium, for example, can include flash memory, hard disk, multimedia card, card-type memory, random access memory (Random Access Memory, RAM), static random access memory (Static Random Access Memory, SRAM), programmable read-only memory (Programmable Read Only Memory, PROM), read-only memory (Read Only Memory, ROM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), magnetic memory, disk, optical disk, etc. Memory is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the embodiment of the present invention can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0071] By designing and programming a processor, the code corresponding to the multi-channel fusion method for anisotropic substrate mechanical property parameters described in the aforementioned embodiment can be embedded in the chip, enabling the chip to execute the steps of the method described in the aforementioned embodiment during operation. Designing and programming a processor is well known to those skilled in the art and will not be further described here.

[0072] Based on the same inventive concept, an embodiment of the present invention further provides a storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer executes a multi-channel fusion method for anisotropic substrate mechanical characteristic parameters discussed above.

[0073] In some optional embodiments, the present invention also provides various aspects of a multi-channel fusion method for anisotropic substrate mechanical property parameters, which can also be implemented in the form of a program product, which includes program code. When the program product is run on an apparatus, the program code is used to enable the control device to execute the steps of a multi-channel fusion method for anisotropic substrate mechanical property parameters according to various exemplary embodiments of the present invention described above in this specification.

[0074] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of a unit described above can be further divided into multiple units to be embodied. In addition, although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, certain steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0075] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a server, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0077] Program code for performing the operations of the present invention may be written using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0078] Where a remote computing device is involved, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0079] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0081] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A multi-channel fusion method for anisotropic substrate mechanical property parameters, characterized in that: include: Construct anisotropic mechanical parameter library of matrix materials; Mapping substrate bitmap pixels to material indices in the anisotropic mechanical parameter library to obtain a material index record matrix; Combining the material index record matrix and the anisotropic mechanical parameter library, obtaining channel layers corresponding to the mechanical parameters of the substrate one by one as the mechanical parameter channels of the substrate bitmap; The mechanical parameter channels of the substrate bitmap are integrated to generate substrate multi-channel bitmap tensor data.

2. The multi-channel fusion method for anisotropic substrate mechanical property parameters according to claim 1, characterized in that: The anisotropic mechanical parameter library of the constructed matrix material includes: Constructing multiple unit matrices corresponding to each material, where the unit matrices are used to store various anisotropic mechanical parameters corresponding to each material; All element matrices of a material are grouped together and the corresponding material index is established to construct the anisotropic mechanical parameter library of the matrix material.

3. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 2, characterized in that: In the anisotropic mechanical parameter library of the matrix material: The number of rows and columns of each unit matrix is ​​the same as the number of pixels in height and width of the substrate bitmap used; The number of unit matrices corresponding to each matrix material is the same as the number of anisotropic mechanical parameters; The element assignment of the unit matrix is ​​based on the corresponding mechanical parameters. The minimum and maximum values ​​of the mechanical parameters in the material are often linearly normalized.

4. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 1, characterized in that: If the anisotropic mechanical parameter library contains an isotropic matrix material, the number of its unit matrices must be consistent with the number of unit matrices of the anisotropic matrix material, and the unit matrices representing material properties in different directions have the same mechanical parameter values.

5. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 1, characterized in that: Mapping the substrate bitmap pixels with the material indexes in the anisotropic mechanical parameter library to obtain a material index record matrix includes: The base material corresponding to the position area of ​​the substrate bitmap pixel is filled with the material index in the anisotropic mechanical parameter library to obtain a material index record matrix containing the material index.

6. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 5, characterized in that: The step of filling the substrate bitmap pixels with the base materials corresponding to the position areas as the material indexes in the anisotropic mechanical parameter library includes: Matrix logic operations are used to extract the corresponding logic matrix for each type of base material in the substrate bitmap. The number of logic matrices is the same as the number of materials in the substrate bitmap, and the size of the logical matrix rows and columns is consistent with the pixels in the substrate bitmap. If the value of the logical matrix element is "true", it means that the material of the corresponding pixel in the substrate bitmap is the material represented by the logical matrix. If the value of the logical matrix element is "false", it means that the material is not represented by the logical matrix. Create a new material index record matrix. If the value of the element in the logical matrix is ​​"true", fill in the material index corresponding to the logical matrix at the same coordinate position in the material index record matrix.

7. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 2, characterized in that: Each mechanical parameter channel of the substrate bitmap is represented by a two-dimensional matrix, and the value of each element in the two-dimensional matrix is ​​assigned to the value of the unit matrix element of the corresponding material according to the type and direction of the mechanical parameter represented by the channel.

8. The multi-channel fusion method for anisotropic substrate mechanical property parameters according to claim 2, characterized in that: The method for assigning values ​​to each element in the two-dimensional matrix includes: The material index record matrix and the material index are subjected to matrix logic operation to obtain a material index mask matrix, and then the material index mask matrix is ​​multiplied by the unit matrix corresponding to the material index in the anisotropic mechanical parameter library according to the corresponding elements.

9. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 1, characterized in that: A three-dimensional tensor is used to integrate the mechanical parameter channels of the substrate bitmap; wherein each dimension of the three-dimensional tensor represents the height direction pixels, width direction pixels, and channel number of the substrate bitmap respectively.

10. The multi-channel fusion method for anisotropic substrate mechanical characteristic parameters according to claim 1, characterized in that: A four-dimensional tensor is used to integrate the mechanical parameter channels of the substrate bitmap; wherein, a three-dimensional tensor is first used to integrate the mechanical parameter channels of the substrate bitmap, and then the three-dimensional tensors of multiple substrate bitmaps are organized, and an additional fourth dimension is added as the number of substrate bitmaps for indexing and distinguishing each substrate bitmap.

Images