Implicit solution based on adjoint residual sorting

Abstract

The invention discloses an implicit solution based on adjoint residual sorting. The implicit solution comprises the following steps: (1) establishing a flow field control equation for an object to bemeasured; (2) setting a boundary condition and an initial flow field condition to calculate a residual error of each cell; (3) calculating a residual influence quantity estimated value according to anadjoint equation of the flow field control equation; and (4) performing sorting calculation on the influence of the residual influence quantity estimation value on the calculation result. The methodcan ensure that all calculations are effective calculations, and can ensure that a calculation area always covers a signal propagation path along with a signal propagation process, so that the redundant calculation amount is reduced to the maximum extent; the computational domain is always driven by propagation of disturbance signals, and an effective region and an ineffective region are automatically distinguished; since disturbance is not propagated to the upstream, invalid calculation of a downstream area which does not influence a calculation result can be effectively reduced; and the precision of a final calculation result can be obtained by utilizing the adjoint matrix, and the result is corrected, so that relatively high calculation precision can be obtained with relatively small calculation amount.

Description

Technical field [0001] The present invention relates to the technical field of computational fluid dynamics, in particular to an implicit solution method based on adjoint residual ordering. Background technique [0002] In the field of computational fluid dynamics (CFD), most calculation models can be solved by the following matrix equations: [0003] Ax+b=0 (1) [0004] The matrix A is the matrix representing the control equation, b is the boundary condition, and x is the vector to be sought. [0005] Usually the method to solve the flow field control equation (1) is to adopt the time correlation method, and set an initial value x for the vector x to be sought 0 , Go through a course that changes over time [0006] [0007] Finally, the solution of formula (1) is obtained, where R is the residual. [0008] The goal of steady-state calculation is to make the residuals tend to 0 and obtain the target solution of the flow field. The size of R is the main sign of the flow field convergence...

AI Technical Summary

Problems solved by technology

[0011] However, these methods have limitations: the calculation domain is limited, and one solution cannot be used to achieve the highest efficiency for all calculation states; PNS changes the equation, which affects the accuracy, and can only be applied to the supersonic region; ADI also has a certain blindness There is no accurate tracking of the propagation path of the signal, and there is a large degree of computational redundancy

However, the residuals in different regions of the flow field have different effects on the final calculation results, especially in the supersonic flow field, because the influence of the disturbance cannot be propagated upstream, and the downstream residuals have no influence on the calculation results at all. The calculation is necessary, and the residual sorting cannot completely avoid the double calculation of this part

Images