Parallel meshfree method and system for solving strong heat fluid-structure interaction problem

Abstract

The invention relates to a parallel meshless method and system for solving a strong heat fluid-structure interaction problem. The parallel meshless method is characterized in that time discretizationis carried out based on an optimal transportation theory, spatial discretization is carried out through adopting a material point method, and a local maximum entropy interpolation function is adoptedas a shape function. According to the parallel meshless method and the system, the mass conservation and the momentum conservation are strictly met in the aspect of time discretization, the Kroneckerattribute is met in the aspect of boundary, and extremely high precision and stability are shown in solving the problems of large deformation, fluid-structure interaction and the like of materials.

Description

technical field [0001] The invention relates to the fields of computational mechanics and numerical simulation, in particular to a parallel gridless method and system for solving strong heat-fluid-solid coupling problems. Background technique [0002] In most industrial applications, including emerging industries such as manufacturing and new energy industries, materials usually suffer from strong heat-fluid-solid coupling problems, especially the recent rapid development of technologies such as additive manufacturing and 3D printing have made strong heat-fluid-solid coupling problems It has become one of the focuses of global research. These strong heat-fluid-solid coupling problems usually include: extremely large deformation problems, large temperature gradient problems, high heating or cooling rate problems, high-speed impact and geometric distortion problems, material fission problems, metal material forming problems, multi-phase transformation problems, etc. physical ...

AI Technical Summary

Problems solved by technology

[0008] First, the Kronecker property is not satisfied, and the calculation accuracy is low. At the same time, because the value of the node shape function on the boundary of the discrete domain is not zero, it is very difficult to apply the essential boundary conditions, and complex special processing is required.

[0009] Second, the equivalent numerical integration in the "weak" form is difficult. At present, several integration schemes mainly used in the meshless method, such as: nodal integration scheme, stress point integration scheme, background grid integration scheme, they all have different degrees Faced with difficulties in dealing with tensile stress instability, numerical integration errors, etc.

[0010] Third, the amount of calculation is large and the efficiency of parallel computing is low

The gridless method uses a high-order interpolation function, which improves the continuity of the interpolation, but also greatly increases the amount of calculation. To solve the interpolation coefficient of each point in the domain, it is necessary to invert the corresponding node coupling matrix, and the smoother the The higher the order of the high-coupling matrix, the more computationally expensive the corresponding inversion

[0011] Fourth, there is a lack of strong thermal-fluid-solid coupling algorithms that can simultaneously predict the temperature field and material phase transition under large deformations

However, at present, the meshless method is still mainly used to solve mechanical problems, and there are few studies on the application of the meshless method to solve strongly coupled thermodynamic problems, and no one has proposed to solve the problem including the energy dissipation mechanism based on the meshless method. Mathematical framework for coupled non-equilibrium thermodynamic problems

Images