Finite element simulation result data coordinate conversion cloud chart display method

Abstract

The application belongs to the technical field of finite element simulation result data processing of structure design, and particularly relates to a finite element simulation result data coordinate conversion cloud chart display method, which comprises the following steps: step one, reading a finite element simulation result data file by CPU, extracting required physical quantities including names, values and IDs from the file, and storing the physical quantities into a host memory; step two, determining data attributes of the physical quantities as node attributes or element attributes according to the names of the physical quantities by the CPU; step three, transmitting the physical quantities, data attributes of the physical quantities, a rotation matrix of a conversion coordinate system and a coordinate origin to GPU display memory by the CPU; step four, performing coordinate conversion on values of the physical quantities according to the data attributes of the physical quantities by the GPU; step five, transmitting the values of the physical quantities after coordinate conversion back to the host memory to update the values of the physical quantities by the GPU; and step six, transmitting the values of the physical quantities after coordinate conversion into a rendering pipeline to perform cloud chart display under the conversion coordinate system by the CPU.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of structural design finite element simulation result data processing, and particularly relates to a finite element simulation result data coordinate transformation cloud chart display method. BACKGROUND

[0002] When designing a structure, for example, designing a wing, a fuselage panel and the like, a finite element simulation software is usually used to analyze the stress and strain of the structure. During the analysis, the cloud chart of the simulation result data needs to be viewed in different coordinate systems.

[0003] Currently, when viewing the cloud chart of the simulation result data in different coordinate systems, the CPU is usually used to perform coordinate transformation on the data. However, the CPU has a slow calculation speed, and it takes a long time to perform coordinate transformation on the simulation result data of a fine finite element model with a large number of grids, which cannot meet the needs of rapid analysis of structural design. Therefore, the application is proposed. SUMMARY

[0004] The application aims to provide a finite element simulation result data coordinate transformation cloud chart display method, which can quickly perform coordinate transformation on the finite element simulation result data, so as to quickly display the cloud chart in different coordinate systems for a structural designer to view, and meet the needs of rapid analysis of structural design.

[0005] The technical scheme of the application is as follows:

[0006] A finite element simulation result data coordinate transformation cloud chart display method comprises the following steps:

[0007] Step 1: reading the finite element simulation result data file by using the CPU, extracting the required physical quantities including the name, value and ID from the file, and storing them in the host memory;

[0008] Step 2: determining the data attribute of the physical quantity as a node attribute or an element attribute according to the name of the physical quantity by using the CPU;

[0009] Step 3: transmitting the physical quantity, the data attribute of the physical quantity and the rotation matrix of the transformation coordinate system, the coordinate origin and the like to the GPU memory by using the CPU;

[0010] Step 4: performing coordinate transformation on the value of the physical quantity according to the data attribute of the physical quantity by using the GPU;

[0011] Step 5: transmitting the value of the physical quantity after coordinate transformation back to the host memory by using the GPU, and updating the value of the physical quantity;

[0012] ​​Step six, the CPU transmits the converted value of the physical quantity coordinate into the rendering pipeline, and displays the cloud chart under the converted coordinate system.

[0013] Optionally, in the finite element simulation result data coordinate conversion cloud chart display method, in step one, a file reading interface of C / C++ is used to read the finite element simulation result data file.

[0014] The name of the physical quantity is stored in the host memory in the form of char[], and the value of the physical quantity is stored in the host memory in the form of double[].

[0015] Optionally, in the finite element simulation result data coordinate conversion cloud chart display method, in step two, the data attribute of the physical quantity with the name of Displacement is determined as a node attribute, and the data attribute of the physical quantity with the name of Stress is determined as an element attribute.

[0016] For each physical quantity, an int variable is initialized in the host memory, if the data attribute of the physical quantity is a node attribute, the int variable is initialized as 0, and if the data attribute of the physical quantity is an element attribute, the int variable is initialized as 1.

[0017] Optionally, in the finite element simulation result data coordinate conversion cloud chart display method, in step four, according to the value of the int variable, the data attribute of the physical quantity is confirmed, and then the value of the physical quantity is converted.

[0018] For the physical quantity with the node attribute, the value of the physical quantity is converted as follows:

[0019] ;

[0020] For the physical quantity with the element attribute, the value of the physical quantity is converted as follows:

[0021] ;

[0022] Wherein,

[0023] is the value of the physical quantity before coordinate conversion;

[0024] is the value of the physical quantity after coordinate conversion.

[0025] Optionally, in the finite element simulation result data coordinate conversion cloud chart display method, in step five, during the process of transmitting the converted value of the physical quantity back to the host memory, the main thread is blocked at the CPU end.

[0026] The present application has at least the following technical effects:

[0027] Provided is a finite element simulation result data coordinate transformation cloud chart display method, a CPU-GPU heterogeneous system, relying on the advantages of CPU in logical control and complex data processing, realizing efficient data extraction and judgment, and at the same time fully utilizing the powerful parallel computing capability of GPU to complete large-scale matrix operation, so as to realize fast coordinate transformation of finite element simulation result data, and to quickly display a cloud chart under different coordinate systems for a structure design personnel to view, meeting the needs of structure design for fast analysis. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 FIG. 1 is a schematic diagram of a finite element simulation result data coordinate transformation cloud chart display method provided by the present application;

[0029] Figure 2 FIG. 2 is a schematic diagram of the efficiency comparison of coordinate transformation of finite element simulation result data using a CPU-GPU heterogeneous system and using a CPU system;

[0030] Figure 3 FIG. 3 is a display cloud chart of finite element simulation result data before and after coordinate transformation.

[0031] In order to better illustrate the present embodiment, some contents in the drawings may be omitted, enlarged or reduced, and are only used for exemplary description, and cannot be understood as a limitation on the present application. DETAILED DESCRIPTION

[0032] In order to make the technical solutions of the present application and its advantages clearer, the technical solutions of the present application will be further clearly and completely described below in conjunction with the drawings, and it can be understood that the specific embodiments described herein are only some embodiments of the present application, and are only used to explain the present application, but not to limit the present application. It should be noted that, in order to facilitate description, only parts related to the present application are shown in the drawings, and other related parts can be referred to the general design.

[0033] In addition, unless otherwise defined, the technical terms or scientific terms used in the present application description should be the general meaning understood by the general technical personnel in the field to which the present application belongs. In the present application description, "including" indicates that the concept appearing before the word covers the concept listed after the word and its equivalent, and does not exclude other associated concepts.

[0034] A finite element simulation result data coordinate transformation cloud chart display method is as shown in FIG. 1. Figure 1

[0035] ​Step one, CPU reads the finite element simulation result data file, extracts the required physical quantities, including name, value and ID, and stores them in the host memory.

[0036] Read the finite element simulation result data file using the C / C++ file reading interface.

[0037] Store the name of the physical quantity in char[] type in the host memory, and store the value of the physical quantity in double[] in the host memory.

[0038] Step two, CPU determines the data attribute of the physical quantity according to the name of the physical quantity, which is node attribute or element attribute.

[0039] The data attribute of the physical quantity with the name Displacement is determined as node attribute, and the data attribute of the physical quantity with the name Stress is determined as element attribute.

[0040] For each physical quantity, initialize an int type variable in the host memory, if the data attribute of the physical quantity is node attribute, initialize the int type variable to 0, if the data attribute of the physical quantity is element attribute, initialize the int type variable to 1.

[0041] Step three, CPU transmits the physical quantity and its data attribute, as well as the rotation matrix of the transformed coordinate system , the coordinate origin to the GPU memory.

[0042] The three coordinate axis directions of the transformed coordinate system relative to the global coordinate system of the finite element simulation result data are , and the coordinate origin is , then , , in the global coordinate system of the finite element simulation result data can be expressed as , , , is a 3x3 matrix.

[0043] Step four, GPU performs coordinate transformation on the value of the physical quantity according to the data attribute of the physical quantity.

[0044] The physical quantity with node attribute, such as displacement , is a one-dimensional vector related to the coordinate origin, and the physical quantity with element attribute, such as stress , is a second-order tensor, which is a pure rotation related quantity and is independent of the coordinate origin translation. Therefore, the value coordinate transformation of the physical quantity with node attribute and element attribute adopts different methods.

[0045] According to the numerical value of the int type variable, the data attribute of the physical quantity is confirmed, and the coordinate transformation of the numerical value of the physical quantity is performed.

[0046] For the physical quantity with the data attribute as the node attribute, the coordinate transformation of the numerical value of the physical quantity is:

[0047]

[0048] For the physical quantity with the data attribute as the unit attribute, the coordinate transformation of the numerical value of the physical quantity is:

[0049]

[0050] wherein,

[0051] is the numerical value before the coordinate transformation of the physical quantity;

[0052] is the numerical value after the coordinate transformation of the physical quantity.

[0053] Step five, the numerical value after the coordinate transformation of the physical quantity is transmitted back to the host memory by the GPU, and the numerical value of the physical quantity is updated.

[0054] Considering that the GPU is executed asynchronously, the main thread is blocked at the CPU end, and after the GPU is executed, the subsequent related operations are performed by the CPU.

[0055] Step six, the numerical value after the coordinate transformation of the physical quantity is transmitted into the rendering pipeline by the CPU, and the cloud chart is displayed under the transformed coordinate system.

[0056] The above embodiment discloses a finite element simulation result data coordinate transformation cloud chart display method, which is based on a CPU-GPU heterogeneous system, relies on the advantages of the CPU in logical control and complex data processing, realizes efficient data extraction and judgment, fully utilizes the powerful parallel computing capability of the GPU, completes large-scale matrix operation, realizes fast coordinate transformation of the finite element simulation result data, and can quickly display the cloud chart under different coordinate systems, so that the structural design personnel can check and meet the needs of the structural design for rapid analysis.

[0057] In practical applications, the efficiency of the coordinate transformation of the finite element simulation result data by using the CPU-GPU heterogeneous system is compared with that of the coordinate transformation of the finite element simulation result data by using the CPU system, such as Figure 2 ​​As shown in the figure, the horizontal coordinate is the number of operation matrix, and the vertical coordinate is the logarithm of operation time, from which it can be known that, compared with the matrix operation using the CPU system, the matrix operation using the CPU-GPU heterogeneous system has higher efficiency, and the efficiency benefit will become more and more significant with the increase of the number of operation matrix.

[0058] In a specific example, the cloud map display method for coordinate transformation of finite element simulation result data disclosed in the above embodiment is based on the CPU-GPU heterogeneous system, and the cloud map display comparison before and after the coordinate transformation of the finite element simulation result data is as shown in Figure 3 As shown in the figure, the display effect is good.

[0059] So far, the technical solution of the application has been described in combination with the preferred embodiments shown in the drawings, and those skilled in the art should understand that the protection scope of the application is obviously not limited to these specific embodiments, and those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the application, and the technical solutions after the changes or replacements will fall within the protection scope of the application.

Claims

1. A method for displaying coordinate transformation cloud maps of finite element simulation results, characterized in that, include: Step 1: Use the CPU to read the finite element simulation result data file, extract the required physical quantities, including names, values ​​and IDs, and store them in the host memory; Step 2: The CPU determines the data attributes of the physical quantity based on its name, whether it is a node attribute or a unit attribute. Step 3: The CPU processes the physical quantities and their data attributes, as well as the rotation matrix for transforming the coordinate system. Origin of coordinates The data is transferred to the GPU memory. Step 4: The GPU performs coordinate transformation on the physical quantity values ​​based on the data attributes of the physical quantity. Step 5: The physical quantity coordinates are converted by the GPU and the resulting values ​​are transferred back to the host memory to update the physical quantity values. Step 6: The CPU transmits the converted physical quantity coordinates to the rendering pipeline, and displays the cloud map in the converted coordinate system. In step two, the data attribute of the physical quantity named Displacement is determined as a node attribute, and the data attribute of the physical quantity named Stress is determined as a unit attribute. For each physical quantity, initialize an int type variable in the host memory. If the data attribute of the physical quantity is a node attribute, initialize the int type variable to 0. If the data attribute of the physical quantity is a unit attribute, initialize the int type variable to 1. In step four, the data attributes of the physical quantity are confirmed based on the value of the int type variable, and then the coordinate transformation of the physical quantity value is performed. For physical quantities whose data attributes are node attributes, the numerical values ​​of the physical quantities are transformed into coordinates as follows: ； For physical quantities whose data attributes are unit attributes, the numerical values ​​of the physical quantities are transformed into coordinates as follows: ； in, The values ​​of the physical quantities before coordinate transformation; The values ​​are the physical quantities after coordinate transformation.

2. The method for displaying coordinate transformation cloud maps of finite element simulation results according to claim 1, characterized in that, In step one, the C / C++ file reading interface is used to read the finite element simulation result data file; The names of physical quantities are stored in host memory as char[] type, and the values ​​of physical quantities are stored in host memory as double[].

3. The method for displaying coordinate transformation cloud maps of finite element simulation results according to claim 2, characterized in that, In step five, during the process of transmitting the converted physical quantity coordinates back to the host memory, the main thread is blocked on the CPU.

Images