Implicit solution based on residual magnitude sorting

Abstract

The invention discloses an implicit solution based on residual magnitude sorting, which relates to the field of numerical calculation. The solution comprises the following steps: establishing a flow field matrix equation for an object to be measured; setting a boundary condition and an initial flow field condition, and calculating a residual error of each cell; defining a residual modulus, and establishing a plurality of solving queues according to the magnitude of the residual modulus; taking out a cell from the non-empty queue with the maximum residual magnitude to carry out control equationiteration until the modulus of the residual drops by one magnitude; calculating and updating the residual error of the cell with the residual error change influenced by the current cell change, and putting the current cell and the cell with the residual error change influenced by the current cell change into a corresponding queue according to the magnitude of a residual error mode; and obtaininga final flow field calculation convergence result until all residual queues greater than the convergence criterion magnitude are empty. The solution only controls the computational domain, can be applied to any algorithm, does not sacrifice the precision or limit the algorithm, can effectively reduce the computing resources required by sorting, and improves the computing efficiency.

Description

technical field [0001] The invention relates to the field of numerical calculation, and in particular to an implicit solution method based on residual magnitude ranking. Background technique [0002] The invention is an accelerated solution method of implicit equations based on residual sorting, which is mainly used for solving fluid mechanics equations, and can also be used for solving other physical field equations or wider linear or nonlinear equations, thereby improving the solution efficiency. [0003] In the field of computational fluid dynamics (CFD), most computational models can be solved with the following matrix equations: [0004] Ax+b=0. (1) [0005] Among them, matrix A is the matrix representing the governing equation, b represents the boundary condition, and x is the vector to be obtained. [0006] The present invention focuses on reducing the amount of repeated calculations and improving the efficiency of solving equations. Specifically in the field of C...

AI Technical Summary

Problems solved by technology

[0015] However, the sorting of residuals also requires resource consumption, and frequent sorting operations may partially offset the effect of improving computational efficiency

In fact, units with the same magnitude of residuals have the same impact on the calculation results, and the order of their calculations does not greatly improve the calculation efficiency.

Images