CN115017773A - Dimension reduction method of three-dimensional grid model, electronic equipment and medium - Google Patents
Dimension reduction method of three-dimensional grid model, electronic equipment and medium Download PDFInfo
- Publication number
- CN115017773A CN115017773A CN202210692935.5A CN202210692935A CN115017773A CN 115017773 A CN115017773 A CN 115017773A CN 202210692935 A CN202210692935 A CN 202210692935A CN 115017773 A CN115017773 A CN 115017773A
- Authority
- CN
- China
- Prior art keywords
- node
- dimension reduction
- matrix
- model
- reduction method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computational Linguistics (AREA)
- Computer Graphics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a dimension reduction method of a three-dimensional grid model, electronic equipment and a medium, wherein the dimension reduction method extracts simulation result data of the three-dimensional grid model into a matrix, then maps the simulation result data into an image, screens out a dimension reduction area and a non-dimension reduction area through pixel adjustment and calculation data volume, and effectively avoids the problems of difficult solution and long time consumption caused by huge dimension in the calculation of grid node data; the dimension reduction method realizes the dimension reduction of the grid model of the dimension reduction area and the node information prediction of the non-dimension reduction area through a neural network algorithm, realizes the high-precision dimension reduction of the three-dimensional grid model, obtains the complete grid node information, achieves the rapid simulation and realizes the real-time diagnosis of the running state of the entity equipment.
Description
Technical Field
The invention relates to the technical field of agricultural machinery, in particular to a dimension reduction method of a three-dimensional grid model, electronic equipment and a medium.
Background
The simulation technology can help designers to provide design basis and guidance, simulation models are mostly grid models and comprise the fields of structures, fluids, electromagnetism and the like, the existing grid model calculation mainly depends on finite element simulation software such as ABAQUS, ANSYS, FLUENT, STARCCM + and the like to carry out solution calculation, for some complex models, dozens of or even millions of grids are often required to be divided in order to guarantee calculation accuracy, when the grid model is solved, a multi-core parallel simulation mode is often adopted to solve, a large amount of CPU resources are consumed, and the consumed time is long.
The grid models are of three types, namely one-dimensional, two-dimensional and three-dimensional. The one-dimensional and two-dimensional grid models are simple, the calculation speed is high, and dimension reduction is not needed. In a conventional simulation calculation, the number of grids of a three-dimensional grid model can be usually in the order of one hundred thousand, even millions and tens of millions, and more grids can effectively improve the calculation accuracy of the model, but a large amount of calculation resources are consumed. The conventional calculation mode of the grid model depends on finite element analysis software, a large amount of time needs to be consumed, and meanwhile, the data of the grid model has the characteristics of less input and huge output dimension. When the traditional machine learning algorithm is used for carrying out dimensionality reduction on the agent model, dimensionality disasters occur, and the agent model is difficult to construct.
In the prior art, a digital twin model construction method adopts a principal component analysis method to realize data dimension reduction, and has the defects that on one hand, the method has no specific physical significance, and on the other hand, data loss occurs during data recovery, and 100% data recovery cannot be realized; another aspect is that such algorithms are suitable for input data to be correlated, and will not be suitable when the input data are completely orthogonal or uncorrelated. The existing model dimension reduction calculation method adopts a parallel simulation mode to improve the simulation efficiency, and has the defect that solution calculation can be completed only by consuming a large amount of calculation resources. The quick re-analysis method excludes the computational grid in the model, so that the simulation solving range can be reduced, the initial matrix dimension of the model is huge, if the dimension for processing data is not limited, a large amount of running memory still needs to be consumed for simple data processing, and the solving speed of the grid model with the reduced range still does not respond to the requirement of rapid evaluation after the scheme is changed by designers.
In view of the above, there is an urgent need for a dimension reduction method for a three-dimensional grid model that better responds to design requirement changes and evaluates related schemes in time to solve the problems in the prior art.
Disclosure of Invention
The invention aims to provide a dimension reduction method of a three-dimensional grid model, electronic equipment and a medium, and the specific technical scheme is as follows:
a dimension reduction method of a three-dimensional grid model comprises the following specific steps:
step S1: performing three-dimensional grid model simulation, specifically, performing finite element analysis on a three-dimensional grid model to obtain a space coordinate and node information of each grid node, performing multiple simulation on the three-dimensional grid model based on the space coordinate and the node information to obtain simulation result data, wherein the simulation result data is stored as a matrix A in a matrix form, and comprises a node maximum value, node space coordinate information, a node stress value and a node strain value;
step S2: dividing a dimensionality reduction area, specifically, giving colors to nodes of the matrix A in the step S1 and establishing a node three-dimensional graph, wherein the larger the stress value in simulation result data is, the darker the color is, otherwise, the lighter the stress value is, converting the node three-dimensional graph into a red-green-blue numerical table, dividing an area in the node three-dimensional graph where color difference changes into dimensionality reduction areas based on color difference changes, reconstructing the nodes in the dimensionality reduction areas into a matrix B, reconstructing the nodes outside the dimensionality reduction areas into a matrix C;
step S3: and D, performing dimensionality reduction processing to obtain a dimensionality reduction model, specifically, establishing a training network through a neural network algorithm, taking the simulation boundary condition as an input parameter, taking the node space coordinate information, the node stress value and the node strain value of the matrix B in the step S2 as output parameters, and performing dimensionality reduction processing on the matrix B to obtain the dimensionality reduction model.
Specifically, in step S1, the node information includes a stress condition range, and the three-dimensional mesh model selects at least two stress conditions within the stress condition range for simulation calculation, where the stress conditions include a maximum stress condition and a minimum stress condition.
Specifically, in step S2, the specific steps of assigning colors to the nodes in the matrix a are as follows:
establishing a color interval, equally dividing the interval into N small intervals, endowing different intervals with different colors, and setting the value of N to be 3-50; and normalizing the node stress values of the nodes, and mapping the nodes and the color intervals one by one according to the node stress values to obtain a three-dimensional node graph.
Specifically, the range of the color interval is [0, 1 ].
Specifically, in step S2, the color values are all represented by 0 to 256, and when any one of the color values changes over K, the color difference is considered to change, and the value range of K is 0 to 50.
Specifically, in step S3, the neural network algorithm structure includes an input layer, an output layer, a hidden layer, and a derived layer, where the input layer is used to write input parameters, the output layer is used to render output parameters, the hidden layer is used to calculate the dimension reduction model, and the derived layer is used to store the dimension reduction model.
Specifically, in step S3, the parameters of the hidden layer are set as follows: the number of the hidden layers is set to be 5-100, the activation function is a sigmoid function or a tanh function, 5-30% of the total data amount is selected for verification data and test data, and the iteration turns are 100-10000.
In addition, the present invention also provides an electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the dimension reduction method as described above when executing the computer program.
In addition, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the dimension reduction method as described above.
The technical scheme of the invention has the following beneficial effects:
the dimension reduction method can perform data matrixing processing on any grid model, is suitable for multi-field grid models, and reduces the file capacity of simulation results.
The dimension reduction method provided by the invention divides the dimension reduction area through image analysis, and utilizes the pixels to adjust the maximum calculated amount, thereby greatly reducing the data operation amount.
The dimension reduction method realizes the node information prediction of the grid model dimension reduction area and the non-dimension reduction area of the dimension reduction area through a neural network algorithm, and simultaneously can realize the quick calculation of a simulation result and the quick simulation technology.
In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of the steps of a dimension reduction method;
FIG. 2 is a schematic diagram of a three-dimensional mesh model;
FIG. 3 is a schematic representation of a coordinate system of a three-dimensional graph of nodes;
FIG. 4 is a schematic drawing of color map (illustrated by a gray scale map);
FIG. 5 is a schematic diagram of a coordinate system of the culling of the color difference free variation area of FIG. 3;
fig. 6 is a schematic diagram of the neural network algorithm structure.
Detailed Description
Embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways, which are defined and covered by the claims.
Example 1:
in order to solve the technical problems that a simulation model is slow in operation speed, a result file is large and is not easy to store, and a design change evaluation requirement cannot be responded quickly, as shown in fig. 1, the embodiment discloses a dimension reduction method for a three-dimensional mesh model, the preferable three-dimensional mesh model in the embodiment is shown in fig. 2, it needs to be explained that the dimension reduction model is a black box model, and the dimension reduction method specifically includes the following steps:
step S1: three-dimensional grid model simulation, specifically, carrying out finite element analysis on the three-dimensional grid model to obtain the space coordinates and node information of each grid node, the node information comprises a stress working condition range, the three-dimensional grid model selects at least two stress working conditions in the stress working condition range for simulation calculation, the stress working conditions comprise maximum stress working conditions and minimum stress working conditions, the three-dimensional grid model is simulated for multiple times based on the space coordinates and the stress working conditions to obtain simulation result data, the simulation result data is stored as a matrix A in a matrix form, the rows of the matrix A represent the serial numbers of the nodes, the preferred serial number sequence of the nodes is consistent with the serial number sequence of the nodes in the finite element software, the simulation result data are placed in the columns of the matrix A in sequence, matrixing processing is adopted, the method is suitable for multi-field grid models, and the file capacity of the simulation result is reduced;
further, the simulation result data includes a node maximum value, node spatial coordinate information, a node stress value, and a node strain value, in this embodiment, a first column of the matrix a stores a node number, a second column, a third column, and a fourth column of the matrix a sequentially stores node spatial coordinate X, Y, Z information, a fifth column of the matrix a stores a node stress value, and a sixth column of the matrix a stores a node strain value;
it should be noted that the node numbering sequence may also be arranged according to a principle of going from top to bottom or going from left to right, and the node numbering does not affect the dimension reduction effect of this embodiment.
Step S2: dividing a dimensionality reduction area, specifically, giving colors to nodes of the matrix A in the step S1 and establishing a node three-dimensional graph, wherein the larger the stress value in simulation result data is, the darker the color is, otherwise, the lighter the stress value is, converting the node three-dimensional graph into a red-green-blue numerical table (RGB numerical table), dividing an area with changed color difference in the node three-dimensional graph into dimensionality reduction areas based on the color difference change, reconstructing the nodes in the dimensionality reduction areas into a matrix B, and reconstructing the nodes outside the dimensionality reduction areas into a matrix C;
the preferred specific steps of assigning colors to the nodes in the matrix a in this embodiment are as follows:
establishing a color interval with the range of [0, 1], endowing each node with an initial color, equally dividing the interval into seven small intervals, respectively defining the small intervals as red, orange, yellow, green, cyan, blue and purple intervals, normalizing the node stress values of the nodes, and mapping the nodes and the color intervals one by one according to the node stress values to obtain a three-dimensional node graph; as shown in the three-dimensional node graph of fig. 3, in fig. 3, gray contrast is used as color distinction, the darker the node is, the node is located in the purple section or close to the purple section, and the lighter the node is, the node is located in the red section or close to the red section;
further, as shown in fig. 4, the middle cube region is a node reconstruction region, which is perpendicular to the YOZ plane, two 1920 × 1080 pixel node reconstruction color maps are respectively selected from the t1 direction and the t2 direction, the node reconstruction color maps are converted into an RGB numerical table, the region where the color difference has a change space is removed, so as to obtain the node three-dimensional map (the color difference is represented by gray scale contrast) with the color difference-free change region as shown in fig. 5, the dimension reduction region is divided by image analysis, the maximum calculation amount is adjusted by using the pixels, and the data calculation amount is greatly reduced;
further, the preferable red-green-blue values in this embodiment are all represented by 0 to 256, where the color difference change is that when any one of the three red-green-blue values changes over K, the color difference is considered to change, the value range of K is 0 to 50, and the preferable value of K in this embodiment is 10.
Further, in the preferred embodiment, the simulation result data stored in the matrix C is replaced by a variable X, and specifically, the node stress values and the node strain values stored in the matrix C are respectively assigned to X1 and X2.
Step S3: performing dimensionality reduction processing to obtain a dimensionality reduction model, specifically, establishing a training network through a neural network algorithm, taking a simulation boundary condition as an input parameter, taking node space coordinate information, a node stress value and a node strain value of the matrix B in the step S2 as output parameters, performing dimensionality reduction processing on the matrix B to obtain the dimensionality reduction model, realizing high-precision dimensionality reduction and non-dimensionality reduction area node information prediction of a grid model of a dimensionality reduction area through the neural network algorithm, and simultaneously realizing quick calculation of a simulation result;
as shown in fig. 6, the neural network algorithm structure includes an input layer, an output layer, a hidden layer, and a derived layer, where the input layer is used to write input parameters, the output layer is used to render output parameters, the hidden layer is used to calculate a dimension reduction model, and the derived layer is used to store the dimension reduction model.
Further, the parameters of the hidden layer in this embodiment are set as follows: the number of layers of the hidden layer is set to 25, the activation function is a sigmoid function, 15% of total data quantity is selected for verification data and test data, and the iteration turn is set to 1000. The correlation coefficient between the dimensionality reduction model and the original model after the iteration is finished is not lower than 99%, and the dimensionality reduction processing of the three-dimensional grid model is realized.
Specifically, the correlation coefficient is 100% of the pearson correlation coefficient, and the expression of the pearson correlation coefficient is as follows:
wherein cov (x, y) is covariance, σ is standard deviation, x represents original model, and y represents dimension reduction model.
Further, in order to better illustrate the advantages and purposes of the embodiment, the embodiment also discloses a rapid simulation method based on the dimension reduction model, which includes the following specific steps:
and clearing the fifth and sixth columns of data of the matrixes A and B, and reserving the node numbers and the coordinate information. And inputting boundary conditions into the dimensionality reduction model, and taking node space coordinate information, node stress values and node strain values in the rapid output matrix B as output parameters by the model. And assigning the node stress value and the node strain value in the matrix B to the corresponding node in the matrix A (based on the correspondence of the node space coordinate information and the node number) according to the node space coordinate information and the node number in the matrix B, and setting the node information as the node stress value and the node strain value by other nodes which are not assigned in the matrix A. And (4) rapidly completing the three-dimensional grid model solution according to the output result of the dimension reduction model, wherein the model calculation precision depends on the correlation coefficient of the dimension reduction model in the third step. And the calculation precision of the three-dimensional grid model is ensured by ensuring that the correlation coefficient of the model is more than 99%.
In addition, this embodiment also discloses an electronic device, including:
a memory for storing a computer program;
a processor for implementing the dimension reduction method as described above when executing the computer program.
In addition, the embodiment also discloses a computer readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the dimension reduction method is implemented as described above.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A dimension reduction method of a three-dimensional grid model is characterized by comprising the following specific steps:
step S1: performing three-dimensional grid model simulation, specifically, performing finite element analysis on a three-dimensional grid model to obtain a space coordinate and node information of each grid node, performing multiple simulation on the three-dimensional grid model based on the space coordinate and the node information to obtain simulation result data, wherein the simulation result data is stored as a matrix A in a matrix form and comprises a maximum node value, node space coordinate information, a node stress value and a node strain value;
step S2: dividing a dimensionality reduction area, specifically, giving colors to nodes of the matrix A in the step S1, establishing a node three-dimensional graph, dividing an area with changed color difference in the node three-dimensional graph into a dimensionality reduction area and a non-dimensionality reduction area based on color difference change, reconstructing the nodes in the dimensionality reduction area into a matrix B, and reconstructing the nodes in the non-dimensionality reduction area into a matrix C;
step S3: and D, performing dimensionality reduction processing to obtain a dimensionality reduction model, specifically, establishing a training network through a neural network algorithm, taking the simulation boundary condition as an input parameter, taking the node space coordinate information, the node stress value and the node strain value of the matrix B in the step S2 as output parameters, and performing dimensionality reduction processing on the matrix B to obtain the dimensionality reduction model.
2. The dimension reduction method according to claim 1, wherein in step S1, the node information includes a stress condition range, and the three-dimensional mesh model selects at least two stress conditions within the stress condition range for simulation calculation, where the stress conditions include a maximum stress condition and a minimum stress condition.
3. The dimension reduction method according to claim 1, wherein in step S2, the specific steps of assigning colors to the nodes in the matrix a are as follows:
establishing a color interval, endowing each node with an initial color, equally dividing the interval into N small intervals, endowing different intervals with different colors, and enabling the value of N to be 3-50; and normalizing the node stress values of the nodes, and mapping the nodes and the color intervals one by one according to the node stress values to obtain a three-dimensional node graph.
4. A dimension reduction method according to claim 3, wherein the color interval is in the range of [0, 1 ].
5. The dimension reduction method according to claim 1, wherein in step S2, the numerical value of the color is represented by 0 to 256; and when any value changes and exceeds K, the color difference is considered to be changed, and the value range of the K is 0 to 50.
6. The dimension reduction method according to claim 1, wherein in step S3, the neural network algorithm structure comprises an input layer, an output layer, a hidden layer and a derived layer, wherein the input layer is used for writing input parameters, the output layer is used for rendering output parameters, the hidden layer is used for calculating the dimension reduction model, and the derived layer is used for storing the dimension reduction model.
7. The dimension reduction method according to claim 6, wherein in step S3, the parameters of the hidden layer are set as follows: the number of the hidden layers is set to be 5-100, the activation function is a sigmoid function or a tanh function, 5-30% of the total data amount is selected for verification data and test data, and the iteration turns are 100-10000.
8. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the dimension reduction method according to any one of claims 1 to 7 when executing the computer program.
9. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, implements the dimension reduction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210692935.5A CN115017773A (en) | 2022-06-17 | 2022-06-17 | Dimension reduction method of three-dimensional grid model, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210692935.5A CN115017773A (en) | 2022-06-17 | 2022-06-17 | Dimension reduction method of three-dimensional grid model, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115017773A true CN115017773A (en) | 2022-09-06 |
Family
ID=83075734
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210692935.5A Pending CN115017773A (en) | 2022-06-17 | 2022-06-17 | Dimension reduction method of three-dimensional grid model, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115017773A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116051786A (en) * | 2023-03-29 | 2023-05-02 | 核工业航测遥感中心 | Quick display method for standard grid three-dimensional model |
| CN117709128A (en) * | 2024-02-05 | 2024-03-15 | 国家超级计算天津中心 | Super-computing-oriented multi-dimensional parallel simulation method, device, equipment and storage medium |
-
2022
- 2022-06-17 CN CN202210692935.5A patent/CN115017773A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116051786A (en) * | 2023-03-29 | 2023-05-02 | 核工业航测遥感中心 | Quick display method for standard grid three-dimensional model |
| CN117709128A (en) * | 2024-02-05 | 2024-03-15 | 国家超级计算天津中心 | Super-computing-oriented multi-dimensional parallel simulation method, device, equipment and storage medium |
| CN117709128B (en) * | 2024-02-05 | 2024-05-14 | 国家超级计算天津中心 | Super-computing-oriented multi-dimensional parallel simulation method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111369681B (en) | Three-dimensional model reconstruction method, device, equipment and storage medium | |
| Gyulassy et al. | Efficient computation of Morse-Smale complexes for three-dimensional scalar functions | |
| CN111784602B (en) | Method for generating countermeasure network for image restoration | |
| CN106778682B (en) | A kind of training method and its equipment of convolutional neural networks model | |
| CN115017773A (en) | Dimension reduction method of three-dimensional grid model, electronic equipment and medium | |
| CN112270406B (en) | Nerve information visualization method of brain-like computer operating system | |
| CN111523546B (en) | Image semantic segmentation method, system and computer storage medium | |
| CN103838829B (en) | Raster vectorization system based on hierarchical boundary-topology search model | |
| CN112489164B (en) | Image coloring method based on improved depth separable convolutional neural network | |
| CN113688545B (en) | Visualization method and system for finite element post-processing result and data processing terminal | |
| CN109447897B (en) | Real scene image synthesis method and system | |
| CN107516131A (en) | Acceleration method and device, electronic equipment and the storage medium of convolutional calculation | |
| CN110334450B (en) | Method for repairing object plane projection error in multi-block structure grid generation | |
| Al’boschiy et al. | Automated balancing method of vector Illustration and its software implementation | |
| CN103871086B (en) | FPGA (Field Programmable Gata Array) construction-based layered raster-to-vector processing method | |
| CN111640296A (en) | Traffic flow prediction method, system, storage medium and terminal | |
| CN114359269A (en) | Virtual food box defect generation method and system based on neural network | |
| CN115953330B (en) | Texture optimization method, device, equipment and storage medium for virtual scene image | |
| CN111402422B (en) | Three-dimensional surface reconstruction method and device and electronic equipment | |
| CN112258652A (en) | Automatic mesh generation method for casting process modeling and simulation application | |
| CN115859011B (en) | Matrix operation method, device, unit and electronic equipment | |
| CN111667052A (en) | Standard and nonstandard volume consistency transformation method for special neural network accelerator | |
| CN113592021B (en) | Stereo matching method based on deformable and depth separable convolution | |
| CN111047654A (en) | High-definition high-speed video background modeling method based on color information | |
| CN112836872B (en) | Multi-GPU-based high-performance numerical solution method for pollutant convection diffusion equation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |