Data transfer method based on compactly supported radial basis function based on geometric space principal feature extraction

Abstract

The invention discloses a compactly supported radial basis function data transfer method based on geometric space principal feature extraction, comprising steps: step 1, extracting the structural calculation grid and calculating the temperature field of the calculated aerodynamic heat generated by the overall shape of the aircraft or the shape of the local components The original grid node coordinates of the unstructured computational grid form the original coordinate matrix; step 2, perform principal component analysis on the original coordinate matrix formed in step 1, and obtain the eigenvector matrix; step 3, use the eigenvector matrix Carrying out coordinate transformation on the original coordinate matrix; step 4, performing geometric scale normalization on the node coordinates after the coordinate transformation in step 3; step 5, performing interpolation on the normalized node coordinates based on the compact radial basis function, etc.; The invention realizes finer prediction of the temperature of the aerodynamic thermal environment and the structural field of the aircraft, realizes higher-precision data transmission; improves the interpolation efficiency at the grid interface, and further improves the data transmission efficiency and the like.

Description

technical field [0001] The present invention relates to the field of data transfer methods, and more specifically, relates to a compactly supported radial basis function data transfer method based on geometric space principal feature extraction. Background technique [0002] The loose coupling method is currently the main method for solving multi-physics coupling problems of aircraft aerodynamics / thermal / structure. Existing CFD, CSD solution methods and program modules. Due to the different characteristics of the well-developed solvers for the flow field and the structure, the grids used in the calculation will also be very different, which causes the mismatch between the two sets of grids on the interface, which leads to In the multi-field coupling analysis, the problem of interpolation calculation and processing of data transmission at the grid interface must be solved. In recent years, the interpolation calculation processing method using radial basis function (radial b...

AI Technical Summary

Problems solved by technology

[0003] The traditional compactly supported radial basis function is isotropic, and considering the distribution of physical quantities of hypersonic vehicles with complex shapes is highly correlated with local flow characteristics and the gradient changes sharply, such as the aircraft head, leading edge, wing rudder interference area and machine The distribution of physical quantities such as pressure and heat flow in a large area of ​​the body varies greatly; in addition to the above factors, since the division of complex shape meshes will be encrypted in areas with sharp changes in shape curvature or physical quantities, the distribution of grid points is also anisotropic, artificially introduced The anisotropy factors of different grid point distributions are eliminated, so there are defects in the use of isotropic basis functions for the calculation and processing of highly anisotropic physical quantities, resulting in a bottleneck in the improvement of accuracy and conservation

Images