Lattice data compression method and system for additive manufacturing and medium

Abstract

The invention discloses a lattice data compression method and system for additive manufacturing and a medium, belongs to the technical field of additive manufacturing, and aims at solving the technical problem that an efficient compression method for lattice data is urgently needed at present to improve the efficiency and precision of lattice structure modeling at the same time. The method comprises the following steps: constructing a unit cell volume field data set based on a predefined unit cell type; constructing a neural network model based on an auto-encoder architecture; training the neural network model through the unit cell volume field data set to obtain a unit cell volume field prediction model; extracting a three-dimensional grid structure of target unit cell volume field data through the unit cell volume field prediction model; and performing accuracy verification on the extracted three-dimensional grid structure, and performing optimization training on the unit cell volume field prediction model according to a verification result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of additive manufacturing, in particular to a lattice data compression method, system and medium for additive manufacturing. BACKGROUND

[0002] With the rapid development of additive manufacturing technology (such as 3D printing technology), the lattice structure in it has been widely used in the fields of aerospace, medical treatment, industry, etc. due to its lightweight, energy absorption, surface area optimization and other characteristics. The digital modeling of the lattice structure is a key link in the additive manufacturing process. For cloud-based computer-aided design software, such as CAD design software, its lattice modeling function directly affects the design efficiency and innovation ability of engineers.

[0003] However, the existing cloud CAD lattice modeling function has the following shortcomings: first, the generation efficiency is slow, and the traditional method has large calculation amount when generating complex lattice structure, resulting in long modeling time and unable to meet the rapid design requirement. Second, the data transmission amount is large, and the lattice volume field data contains a large amount of three-dimensional discrete point data, which occupies a large amount of resources in the storage and transmission process, especially in the cloud collaboration or remote manufacturing scene, the transmission delay significantly affects the efficiency.

[0004] In the prior art, part of the solution modeling algorithm relies on CPU serial calculation, which is difficult to parallel processing, resulting in limited generation speed of complex lattice. The volume field data is directly stored and transmitted in the original format without effective compression, causing waste of storage space and network transmission delay. The generation and transmission time is long, which affects the design feedback cycle, and the designers are difficult to quickly verify the combination of different lattice parameters, which restricts the innovation efficiency. Although some schemes use simplified models or low-precision approximation algorithms, the precision and design flexibility of the lattice structure are sacrificed, and it is difficult to balance efficiency and high fidelity. The traditional general compression algorithm does not mine the periodicity, topological correlation and other characteristics of the lattice data, and the compression rate is low and the key physical information is easy to lose, which cannot meet the strict requirement of precision of additive manufacturing.

[0005] In summary, there is an urgent need for an efficient compression method for lattice data to improve the efficiency and precision of lattice structure modeling. SUMMARY

[0006] The embodiments of the present application provide a lattice data compression method, system and medium for additive manufacturing, which is used to solve the technical problem that there is an urgent need for an efficient compression method for lattice data to improve the efficiency and precision of lattice structure modeling.

[0007] The embodiments of the present application adopt the following technical solutions: In one aspect, the embodiment of the present application provides a lattice data compression method for additive manufacturing, which comprises the following steps: constructing a neural network model based on a self-encoder architecture; training the neural network model through the unit cell volume field data set to obtain a unit cell volume field prediction model; extracting a three-dimensional grid structure of target unit cell volume field data through the unit cell volume field prediction model; verifying the accuracy of the extracted three-dimensional grid structure, and optimizing and training the unit cell volume field prediction model according to the verification result.

[0008] In one possible implementation, the unit cell volume field data set is constructed based on a predefined unit cell type, specifically comprising the following steps: constructing a unit cell basic topology based on the predefined unit cell type; calculating a signed distance field (SDF) of each unit cell based on the unit cell basic topology, and converting the SDF into a volume field representing a structure proportion; performing boundary reflection filling on the volume field of each unit cell to obtain a complete unit cell volume field; randomly sampling data points in the complete unit cell volume field, and uniformly adjusting the coordinate ranges of the data points to generate the unit cell volume field data set.

[0009] In one possible implementation, after the unit cell volume field data set is constructed based on the predefined unit cell type, the method further comprises the following steps: visualizing and verifying the unit cell volume field data in the unit cell volume field data set through an existing lattice structure digital display method, and determining whether the unit cell volume field data meet verification standards; if not, optimizing and adjusting the unqualified items until the verification standards are met; if yes, saving the unit cell volume field data set in a file form in a local device.

[0010] In one possible implementation, the neural network model based on the self-encoder architecture is constructed, specifically comprising the following steps: constructing an encoder submodule and a decoder submodule, wherein the output end of the encoder submodule is connected to the input end of the decoder submodule to obtain a self-encoder architecture; the encoder submodule is used to map input data to a hidden space to obtain a hidden code vector; and the decoder submodule is used to map the hidden code vector back to an original data space; embedding the self-encoder architecture into a neural network structure to obtain the neural network model based on the self-encoder architecture; The neural network model is deployed in parallel threads of a GPU to realize parallel compression and reconstruction of the unit cell volume field data.

[0011] In an available embodiment, the neural network model is trained by the unit cell volume field data set to obtain a unit cell volume field prediction model, specifically including: The unit cell volume field data set is input into the neural network model, the input unit cell volume field data is converted into an implicit encoding vector by the encoder submodule, and the implicit encoding vector is reconstructed into unit cell volume field data by the decoder submodule to obtain predicted volume field data, forming a complete end-to-end training process; In the training process, the weight parameters and bias parameters of the network are updated by the Adam optimizer; The loss value of the output predicted volume field data and the target value is calculated by the mean square error loss function, and the neural network model is optimized according to the loss value until the network converges, and the unit cell volume field prediction model is obtained.

[0012] In an available embodiment, the three-dimensional grid structure of the target unit cell volume field data is extracted by the unit cell volume field prediction model, specifically including: The pre-trained unit cell volume field prediction model and its model parameters are loaded; Select the target unit cell type and preset the volume field threshold as the judgment standard for grid extraction; The target unit cell volume field data is input into the unit cell volume field prediction model for volume field reconstruction to obtain target unit cell volume field reconstruction data; According to the unit cell type and the preset volume field threshold, the corresponding three-dimensional grid structure is extracted in the target unit cell volume field reconstruction data by the double contour extraction algorithm ODC.

[0013] In an available embodiment, after the three-dimensional grid structure of the target unit cell volume field data is extracted by the unit cell volume field prediction model, the method further includes: The extracted three-dimensional grid structure is uploaded to the visualization software for visualization to obtain a three-dimensional unit cell grid structure diagram corresponding to the target unit cell volume field data; Based on the three-dimensional unit cell grid structure diagram, the target object is manufactured by additive manufacturing.

[0014] In an available embodiment, the extracted three-dimensional grid structure is verified for accuracy, and the unit cell volume field prediction model is optimized and trained according to the verification result, specifically including: In the visualization software, the three-dimensional unit cell grid structure diagram is verified for accuracy and integrity, to obtain construction precision and construction integrity; Based on the comparison result of the construction precision and the preset precision threshold, and the comparison result of the construction integrity and the preset integrity threshold, the model parameters of the unit cell volume field prediction model are adjusted and optimized.

[0015] On the other hand, the embodiment of the present application also provides a lattice data compression system for additive manufacturing, which comprises: A data set construction module is configured to construct a unit cell volume field data set based on a predefined unit cell type; A model training module is configured to construct a neural network model based on an autoencoder architecture, and train the neural network model based on the unit cell volume field data set to obtain a unit cell volume field prediction model; A grid extraction module is configured to extract a three-dimensional grid structure of a target unit cell volume field data through the unit cell volume field prediction model, and perform accuracy verification on the extracted three-dimensional grid structure, and optimize the unit cell volume field prediction model according to the verification result.

[0016] Finally, the embodiment of the present application also provides a storage medium, which is a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores at least one program, each of which includes instructions, which, when executed by a terminal, causes the terminal to execute the lattice data compression method for additive manufacturing.

[0017] Compared with the prior art, the lattice data compression method, system and medium for additive manufacturing provided by the embodiment of the present application have the following beneficial effects: Firstly, compared with the traditional method, the present application realizes efficient compression of unit cell volume field data through an autoencoder architecture, realizes compact representation of lattice data, and greatly reduces data storage and transmission costs. At the same time, the autoencoder output result is combined with the ODC algorithm for grid extraction, realizing the organic combination of unit cell volume field reconstruction and grid extraction algorithm. And data preprocessing, model training, inference and grid extraction are all included in the GPU acceleration framework, realizing end-to-end parallel efficiency optimization. It can reduce the time-consuming of cloud CAD lattice modeling from minutes to seconds, fully utilize the server computing power, and adapt to large-scale lattice structure processing demand.

[0018] Secondly, the present application ensures that the reconstructed unit cell structure is highly consistent with the original structure through a deep learning model, and guarantees the geometric precision; the grid extraction based on the ODC algorithm ensures the topological correctness and geometric accuracy of the reconstructed lattice, and meets the precision requirements of additive manufacturing.

[0019] Finally, the application supports 34 common cell structure types, wide application range; through the extensible neural network architecture, different types of lattice design requirements can be adapted to, providing technical support for various additive manufacturing application scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art without creative labor. In the drawings: Figure 1 A lattice data compression method for additive manufacturing provided by the embodiment of the present application is shown in the flow chart; Figure 2 A lattice data compression method for additive manufacturing provided by the embodiment of the present application is shown in the flow chart; Figure 3 A lattice data compression method for additive manufacturing provided by the embodiment of the present application is shown in the flow chart; Figure 4 A body-centered cubic cell structure visualization schematic diagram provided by the embodiment of the present application is shown in the flow chart; Figure 5 A lattice data compression method for additive manufacturing provided by the embodiment of the present application is shown in the flow chart. DETAILED DESCRIPTION

[0021] In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0022] The embodiment of the present application provides a lattice data compression method for additive manufacturing, as shown in Figure 1 The lattice data compression method for additive manufacturing specifically includes steps S101-S105: S101, based on the pre-defined cell type, constructing a cell volume field data set.

[0023] Specifically, a unit cell basic topology is constructed based on a predefined unit cell type; a signed distance field (SDF) of each unit cell is calculated based on the unit cell basic topology, and the SDF is converted into a volume field representing a structure proportion; a complete unit cell volume field is obtained by performing boundary reflection padding on the volume field of each unit cell; data points are randomly sampled in the complete unit cell volume field, and the coordinate ranges of the data points are unified to generate a unit cell volume field dataset.

[0024] In data set preprocessing, especially for image, signal, time series data and other data with "spatial / sequence continuity", the core significance of boundary reflection padding (also called "mirror padding") is to maximize the retention of the characteristic continuity of the original data when supplementing the "virtual area" outside the data boundary, and to avoid information loss, distortion or artifacts caused by boundary truncation or operations (such as convolution). Further, the unit cell volume field data in the unit cell volume field dataset is visualized and verified by an existing lattice structure digital display method, and it is judged whether it meets the verification standard. If not, the unqualified items are optimized and adjusted until the verification standard is met; if so, the unit cell volume field dataset is saved in the local in the form of a file.

[0025] As a feasible implementation manner, Figure 2 A lattice data compression method for additive manufacturing provided by the embodiment of the present application has the specific flow chart as shown in the figure, Figure 2 The present application realizes efficient generation and compression of lattice data by constructing a unit cell volume field auto-encoder based on deep learning and combining GPU acceleration technology. The core process includes: a) Unit cell volume field dataset generation: through data processing operations such as unit cell topology construction, volume field calculation, boundary padding, and random sampling, a unit cell volume field dataset is constructed.

[0026] b) Unit cell volume field training: a deep neural network based on an auto-encoder architecture is designed to realize efficient compression and high-precision reconstruction of unit cell volume field data.

[0027] c) Grid extraction: based on the ODC algorithm, a three-dimensional grid is extracted from the reconstructed volume field.

[0028] d) GPU acceleration: the parallel computing capability of GPU is used to optimize the model training and inference process, and to speed up the grid extraction efficiency.

[0029] And the process of generating the unit cell volume field dataset includes: 1. Cell topology construction: generate a basic topology based on predefined cell types. Cell types include at least 34 common cell structure types, such as Diamond, Fluorite, Octet, Truncated Cube, Truncated Octahedron, Kelvin Cell, IsoTruss, Re-entrant, Weaire-Phelan, Triangular Honeycomb, Triangular Honeycomb Rotated, Hexagonal Honeycomb, Re-entrant Honeycomb, Square Honeycomb Rotated, Square Honeycomb, Face Centered Cubic Foam, Body Centered Cubic Foam, Simple Cubic Foam, Gyroid, Schwarz, Schwarz Diamond, Lidinoid, SplitP, Neovius, etc.

[0030] 2. Volume field calculation: based on the cell basic topology structure, calculate the signed distance field (SDF) of each cell, and convert the signed distance field (SDF) to a volume field representing the structure proportion. 3. Boundary filling: reasonably fill the cell boundary to ensure data continuity and integrity.

[0031] 4. Random sampling: randomly sample points from the complete cell to generate a training data set.

[0032] 5. Coordinate normalization: unify the coordinate range to eliminate training bias caused by scale differences.

[0033] 6. Visualization verification: visualize the cell volume field data in the cell volume field data set through existing lattice structure digital display methods, and determine whether it meets the verification standard.

[0034] S102, construct a neural network model based on an autoencoder architecture.

[0035] Specifically, an encoder submodule and a decoder submodule are constructed, the output end of the encoder submodule is connected to the input end of the decoder submodule, and an autoencoder architecture is obtained; wherein the encoder submodule is used to map the input data to a hidden space to obtain a hidden code vector; the decoder submodule is used to map the hidden code vector back to the original data space.

[0036] Further, the autoencoder architecture is embedded in the neural network structure to obtain a neural network model based on the autoencoder architecture. The neural network model is deployed in the parallel threads of the GPU to realize parallel compression and reconstruction of the unit cell volume field data.

[0037] S103, training the neural network model through the unit cell volume field data set to obtain a unit cell volume field prediction model.

[0038] Specifically, the unit cell volume field data set is input into the neural network model, the input unit cell volume field data is converted into an implicit encoding vector through the encoder submodule, and then the implicit encoding vector is reconstructed into the unit cell volume field data through the decoder submodule to obtain the predicted volume field data, forming a complete end-to-end training process.

[0039] During the training process, the weight parameters and bias parameters of the network are updated by the Adam optimizer. The loss value of the output predicted volume field data and the target value is calculated by the mean square error loss function, and the neural network model is optimized according to the loss value until the network converges, obtaining the unit cell volume field prediction model.

[0040] As a feasible implementation, Figure 3 A unit cell volume field prediction model training flowchart is provided for the embodiment of the application, as shown in Figure 3 The neural network based on the autoencoder architecture is trained in combination with the GPU acceleration to realize efficient compression and high-precision reconstruction of the unit cell volume field data. The module converts the original unit cell volume field data into 128-dimensional implicit encoding through the encoder, and then reconstructs the implicit encoding into the original volume field data through the decoder, forming a complete end-to-end training process.

[0041] The autoencoder architecture includes two submodules of the encoder and the decoder. The encoder maps the input data to the implicit space, and the mathematical expression of the encoder function is: ; wherein P is the point set data, V is the volume field data, and z is the implicit encoding.

[0042] The decoder maps the implicit encoding back to the original data space, and its mathematical expression is: ; wherein P is the point set data, z is the implicit encoding, and V is the reconstructed volume field data.

[0043] Optimizer: Adam optimizer is used to update the weight and bias of the network during the training process.

[0044] Loss function: Mean Squared Error (MSE) function is used to calculate the squared difference between the output and the target value; Grid extraction module: Extract high-fidelity three-dimensional grid from implicit function Occupancy Function, and use GPU parallelization to achieve fast calculation.

[0045] S104, extract the three-dimensional grid structure of the target unit cell volume field data through the unit cell volume field prediction model.

[0046] Specifically, load the pre-trained unit cell volume field prediction model and its model parameters. Select the target unit cell type and pre-set the volume field threshold as the judgment standard for grid extraction.

[0047] Further, input the target unit cell volume field data into the unit cell volume field prediction model to reconstruct the volume field and obtain the target unit cell volume field reconstruction data.

[0048] Further, according to the unit cell type and the pre-set volume field threshold, extract the corresponding three-dimensional grid structure in the target unit cell volume field reconstruction data through the double contour extraction algorithm ODC.

[0049] Further, upload the extracted three-dimensional grid structure to the visualization software for visualization and display to obtain the three-dimensional unit cell grid structure diagram corresponding to the target unit cell volume field data. Based on the three-dimensional unit cell grid structure diagram, the target object is subjected to additive manufacturing.

[0050] As a feasible implementation, 34 different types of unit cell structure data sets are used, each containing a large number of three-dimensional point coordinates and their corresponding volume field values, to construct a comprehensive training data set covering a variety of lattice morphologies. The batch size is set to 2, the optimizer is Adam, the initial learning rate is 0.0005, and the training round is 100000 times to ensure model convergence and reconstruction accuracy.

[0051] Compression effect: The original volume field data of 3N dimensions (N is the number of points, 100,000 points) is successfully compressed to 128-dimensional implicit encoding vectors. The original volume field data file size is 106251 KB, and the compressed file size is 3510 KB, significantly reducing data storage and transmission costs.

[0052] Load the pre-trained autoencoder model, including the encoder and decoder network parameters. Select the unit cell type and parameters: select the target unit cell type from 34 unit cell structures and set the volume field threshold to 0.5 as the judgment standard for grid extraction. Use the decoder network to reconstruct the 128-dimensional implicit encoding vector into the original dimension volume field data. Use the ODC algorithm to extract the three-dimensional grid from the reconstructed volume field.

[0053] Finally, the extracted three-dimensional grid is visualized to verify the geometric accuracy and integrity of the unit cell structure, as shown in Figure 4 ​

[0054] S105. Verify the accuracy of the extracted three-dimensional mesh structure, and optimize and train the cell volume field prediction model based on the verification results.

[0055] Specifically, the accuracy and integrity of the three-dimensional unit cell mesh structure diagram are verified in the visualization software to obtain the construction accuracy and construction integrity.

[0056] Furthermore, based on the comparison results of the construction accuracy and the preset accuracy threshold, as well as the comparison results of the construction completeness and the preset completeness threshold, the model parameters of the cell volume field prediction model are adjusted and optimized training is performed.

[0057] In addition, embodiments of the present invention also provide a lattice data compression system for additive manufacturing, such as... Figure 5 As shown, the lattice data compression system 500 for additive manufacturing specifically includes: Dataset construction module 510 is used to construct a unit cell volume field dataset based on a predefined unit cell type; The model training module 520 is used to construct a neural network model based on an autoencoder architecture; the neural network model is trained using the cell volume field dataset to obtain a cell volume field prediction model. The mesh extraction module 530 is used to extract the three-dimensional mesh structure of the target cell volume field data through the cell volume field prediction model; to verify the accuracy of the extracted three-dimensional mesh structure; and to optimize and train the cell volume field prediction model based on the verification results.

[0058] Finally, this embodiment of the invention also provides a storage medium, which is a non-volatile computer-readable storage medium storing at least one program, each program including instructions that, when executed by a terminal, cause the terminal to perform: Construct a unit cell volume field dataset based on predefined unit cell types; Construct a neural network model based on an autoencoder architecture; The neural network model is trained using the cell volume field dataset to obtain a cell volume field prediction model. The three-dimensional mesh structure of the target unit cell volume field data is extracted using the unit cell volume field prediction model. The accuracy of the extracted three-dimensional mesh structure is verified, and the cell volume field prediction model is optimized and trained based on the verification results.

[0059] Various embodiments of the present application are described in progression, for example, as a person of ordinary skill in the art would, upon understanding the disclosure, readily conceive the various embodiments as built-up from one of progressive nature. The same elements have the same reference numerals to the extent possible between the various drawings like reference numerals designate corresponding parts throughout the several views. Each embodiment is focused on the differences from other embodiments. In particular, the device, apparatus, and non-transitory computer storage medium embodiments are described more simply as they are substantially similar to the method embodiments.

[0060] The foregoing description of specific embodiments of the application has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise forms disclosed. Many modifications and variations are possible in light of the above teaching. It is intended that the scope of the application be limited not with this detailed description, but rather by the claims appended hereto.

[0061] The embodiments of the application described hereinabove are intended to be merely exemplary and those skilled in the art will recognize variations from these specific embodiments that fall within the broad scope of the application. Those variations are intended to be included within the broad scope of the application that is defined by the claims that follow.

Claims

1. A lattice data compression method for additive manufacturing, characterized by, The method comprises: constructing a unit cell volume field dataset based on a predefined unit cell type; constructing a neural network model based on an autoencoder architecture; training the neural network model based on the unit cell volume field dataset to obtain a unit cell volume field prediction model; extracting a three-dimensional grid structure of target unit cell volume field data through the unit cell volume field prediction model; verifying the accuracy of the extracted three-dimensional grid structure and optimizing the training of the unit cell volume field prediction model according to the verification result.

2. A lattice data compression method for additive manufacturing according to claim 1, characterized in that, Based on the predefined unit cell type, the unit cell volume field dataset is constructed, specifically comprising: constructing a unit cell basic topology based on the predefined unit cell type; calculating the signed distance field (SDF) of each unit cell based on the unit cell basic topology, and converting the signed distance field (SDF) into a volume field representing the structure proportion; boundary reflection filling the volume field of each unit cell to obtain a complete unit cell volume field; randomly sampling data points in the complete unit cell volume field and unifying the coordinate range of each data point to generate the unit cell volume field dataset.

3. A lattice data compression method for additive manufacturing according to claim 2, characterized in that, After constructing the unit cell volume field dataset based on the predefined unit cell type, the method further comprises: visualizing and verifying the unit cell volume field data in the unit cell volume field dataset through an existing lattice structure digital display method, and determining whether it meets the verification standard; if not, optimizing and adjusting the unqualified items until the verification standard is met; if so, saving the unit cell volume field dataset in file form locally.

4. The lattice data compression method for additive manufacturing of claim 1, wherein, Constructing a neural network model based on an autoencoder architecture, specifically comprising: constructing an encoder submodule and a decoder submodule, the output of the encoder submodule is connected to the input of the decoder submodule to obtain an autoencoder architecture; wherein the encoder submodule is used to map the input data to an implicit space to obtain an implicit encoding vector; the decoder submodule is used to map the implicit encoding vector back to the original data space; embedding the autoencoder architecture into a neural network structure to obtain the neural network model based on the autoencoder architecture; deploying the neural network model in the parallel threads of GPU to realize the parallel compression and reconstruction of unit cell volume field data.

5. A lattice data compression method for additive manufacturing according to claim 4, characterized in that, Training the neural network model based on the unit cell volume field dataset to obtain a unit cell volume field prediction model, specifically comprising: inputting the unit cell volume field dataset into the neural network model, converting the input unit cell volume field data into an implicit encoding vector through the encoder submodule, and reconstructing the implicit encoding vector into unit cell volume field data through the decoder submodule to obtain prediction volume field data, forming a complete end-to-end training process; updating the weight parameters and bias parameters of the network through the Adam optimizer during the training process; calculating the loss value of the output prediction volume field data and the target value through the mean square error loss function, and optimizing the parameters of the neural network model according to the loss value until the network converges to obtain the unit cell volume field prediction model.

6. The lattice data compression method for additive manufacturing of claim 1, wherein, The method comprises the following steps: loading a pre-trained crystal cell volume field prediction model and model parameters thereof; selecting a target crystal cell type and presetting a volume field threshold as a judgment criterion for grid extraction; inputting the target crystal cell volume field data into the crystal cell volume field prediction model to perform volume field reconstruction and obtain target crystal cell volume field reconstruction data; extracting a corresponding three-dimensional grid structure from the target crystal cell volume field reconstruction data according to the crystal cell type and the preset volume field threshold through an ODC double-contour extraction algorithm.

7. The lattice data compression method for additive manufacturing of claim 1, wherein, After extracting the three-dimensional grid structure of the target crystal cell volume field data through the crystal cell volume field prediction model, the method further comprises the following steps: uploading the extracted three-dimensional grid structure to a visualization software to perform visualization and obtain a three-dimensional crystal cell grid structure diagram corresponding to the target crystal cell volume field data; performing additive manufacturing on a target object based on the three-dimensional crystal cell grid structure diagram.

8. A lattice data compression method for additive manufacturing according to claim 7, characterized in that, verifying the accuracy of the extracted three-dimensional grid structure and optimizing the training of the crystal cell volume field prediction model according to the verification result, specifically comprising the following steps: verifying the accuracy and completeness of the three-dimensional crystal cell grid structure diagram in the visualization software to obtain construction accuracy and construction completeness; adjusting the model parameters of the crystal cell volume field prediction model based on the comparison results of the construction accuracy and a preset accuracy threshold and the comparison results of the construction completeness and a preset completeness threshold, and performing optimization training.

9. A lattice data compression system for additive manufacturing, characterized by, The system comprises: a data set construction module configured to construct a crystal cell volume field data set based on a predefined crystal cell type; a model training module configured to construct a neural network model based on an autoencoder architecture, train the neural network model based on the crystal cell volume field data set, and obtain a crystal cell volume field prediction model; a grid extraction module configured to extract a three-dimensional grid structure of target crystal cell volume field data through the crystal cell volume field prediction model, verify the accuracy of the extracted three-dimensional grid structure, and optimize the training of the crystal cell volume field prediction model according to the verification result.

10. A storage medium, characterized by The storage medium is a non-volatile computer-readable storage medium, which stores at least one program, each of which includes instructions that, when executed by a terminal, cause the terminal to perform a crystal lattice data compression method for additive manufacturing according to any one of claims 1-8.

Images