Flow thermosetting coupling calculation method based on physical neural network

A technology of neural network and calculation method, applied in the field of fluid-thermo-structure coupling calculation based on physical neural network, can solve problems such as time-consuming and computational resources, and no solution to fluid-thermo-structure coupling.

Pending Publication Date: 2022-04-12
BEIHANG UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problems that the existing methods need to consume a lot of time and computing resources, and there are still no solutions for complex coupl

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Flow thermosetting coupling calculation method based on physical neural network
  • Flow thermosetting coupling calculation method based on physical neural network
  • Flow thermosetting coupling calculation method based on physical neural network

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0035] Specific implementation mode 1. Combination Figure 1 to Figure 4 Description of this embodiment, a calculation method based on physical neural network-based fluid thermo-solid coupling, that is, using physical neural network coupling to solve N-S equations with definite solution conditions, heat conduction equations, and elastic mechanics equations, by combining fluid domains and solid domains The neural network is established separately, and the continuity condition of the interface is used for coupling calculation, which is referred to as the partition coupling calculation method.

[0036] The method realizes partition coupling calculation through the training point sampling module, the N-S equation group solution module, the heat conduction equation solution module, the elastic mechanics equation group solution module, and the judgment calculation convergence module. The specific steps are as follows:

[0037] 1. Training point sampling module;

[0038] Set the com...

specific Embodiment approach 2

[0050] Specific embodiment two, combine Figure 5 with Image 6 Description of this embodiment, a flow-heat-solid coupling calculation method based on a physical neural network, the method establishes a set of neural networks for the fluid domain and the solid domain to solve the whole field calculation scheme, called the whole field coupling calculation method; the specific steps are as follows :

[0051] A, the training point sampling module; the calculation geometric model file is set as the input of the training point sampling module, and can be randomized on solid domain, fluid domain and boundary surface by commercial software (MeshLab, open3D, ICEM, etc.) that can carry out geometric analysis. Sampling, the output of the module is the coordinates of the sampling points in the solid domain, the coordinates of the sampling points in the fluid domain, the coordinates of the sampling points on the boundary surface, and the external normal vector of the sampling points on t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a fluid-thermosetting coupling calculation method based on a physical neural network, relates to the technical field of fluid-thermosetting coupling mechanical application, and solves the problems that an existing method needs to consume a large amount of time and calculation resources, and there is no solution for complex coupling problems such as fluid-thermosetting coupling. One scheme is a scheme of separately establishing neural networks for a fluid domain and a solid domain and performing coupling calculation by adopting a continuity condition of an interface, and is called as a partition coupling calculation method; and the second scheme is a whole-field calculation scheme for establishing a set of neural network for a fluid domain and a solid domain to solve, and is called as a whole-field coupling calculation method. According to the method, discretization of a control equation is not needed, spatial discretization is carried out by replacing grid division with spatial sampling, and bidirectional coupling calculation of fluid mechanics, heat transfer theory and elastic mechanics can be realized. Therefore, more accurate physical field information of a solid domain and a fluid domain can be obtained.

Description

technical field [0001] The invention relates to a flow heat-solid coupling calculation method based on a physical neural network. Background technique [0002] Fluid thermosolid coupling mechanics involves multiple disciplines such as fluid mechanics, heat transfer, and elastic mechanics. The solid in the flow field will generate convective heat exchange with the fluid, which will affect the temperature field of the solid. The inhomogeneity of the temperature field will cause the deformation of the solid structure, which in turn will affect the flow field. The important feature of the flow-heat-solid multi-field coupling problem is that the flow field, temperature field, and deformation field interact with each other, which will affect the accurate solution of the physical field. In a simple system where the coupling effect is not strong, the method of separately solving the physical fields can simplify the calculation and ensure a certain degree of calculation accuracy. H...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F17/12G06F17/13G06F30/10G06F30/27G06F111/10G06F113/08G06F119/14
Inventor 朱剑琴王燕嘉陶智邱璐黄俊杰姚广宇
Owner BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap