An Adaptive Flux Calculation Method for High-Resolution Simulation of Aircraft Flow
By using an adaptive flux calculation method, regions with different physical properties in the flow field are dynamically identified, and the most economical calculation path is selected. This solves the problem of high computational cost in high-resolution simulation of aircraft flow, and achieves more efficient flow simulation and more accurate aerodynamic and thermal prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CALCULATION AERODYNAMICS INST CHINA AERODYNAMICS RES & DEV CENT
- Filing Date
- 2026-06-30
- Publication Date
- 2026-07-31
AI Technical Summary
Existing high-resolution simulations of aircraft flow are costly, especially in smooth regions where computational resources are wasted, hindering their widespread application in practical engineering design.
An adaptive flux calculation method is adopted. By obtaining two sets of nonlinear reconstruction weights on the computing grid, calculating their difference, and selecting different calculation modes according to the difference: a simplified mode is used in smooth regions, and a high-precision mode is used in critical regions, so as to realize the on-demand allocation of computing resources.
With the same computing resources, it improves the efficiency of high-resolution simulation of aircraft flow, obtains aerodynamic and aerothermal data with higher confidence, and helps in the refined design of aircraft.
Smart Images
Figure CN122490954A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of high-performance computing in computational fluid dynamics, and more specifically, to an adaptive flux calculation method for high-resolution simulation of aircraft flow. Background Technology
[0002] High-precision computational fluid dynamics (CFD) is a core tool in modern aircraft aerodynamic design. To accurately predict aircraft drag, lift, and surface thermal loads, detailed simulations of key physical processes in the flow field are essential. These include boundary layer transition, turbulence development, shock wave-boundary layer interaction, and separated flows. These processes involve a wide range of spatiotemporal scales, especially the analytical aspects of turbulent structures, requiring the use of high-resolution numerical schemes (such as WENO and TENO) and large-scale calculations on very fine grids (such as Large Eddy Simulation (LES) or Direct Numerical Simulation (DNS)). However, the computational cost of such simulations is extremely high, becoming a major bottleneck restricting their widespread application in practical engineering design.
[0003] In existing technologies, methods such as dimensionality reduction feature projection have emerged to improve computational efficiency, reducing computational load by lowering the dimensionality of the feature projection. However, these methods still employ a uniform flux synthesis formula with high computational cost for each computational unit of the flow field. The inventors recognized that in aircraft flow fields, strong nonlinear effects (such as shock waves and strong shear) are usually concentrated only in local regions (such as the nose shock wave region, wing leading edge, and shock wave / boundary layer interference region), while most of the flow field is relatively smooth. In these smooth regions, some higher-order correction terms in the flux synthesis formula contribute very little, but the calculation is still forced, resulting in a huge waste of computational resources. Summary of the Invention
[0004] To address the efficiency bottleneck of high-resolution flow simulation in the refined design of aircraft, the present invention aims to improve the adaptive flux calculation efficiency of high-resolution flow simulation of aircraft, thereby improving the computational efficiency of full-field simulation of aircraft.
[0005] To achieve the above-mentioned objective, this invention provides an adaptive flux calculation method for high-resolution simulation of aircraft flow, the method comprising:
[0006] Step S1: On the computational grid describing the local configuration of the aircraft surface, two sets of nonlinear reconstruction weights are obtained at the cell interfaces in the computational grid through a flux reconstruction process, namely the first set of weights and the second set of weights.
[0007] Step S2: Calculate the difference between the first set of weights and the second set of weights;
[0008] Step S3: If the difference is less than or equal to a preset threshold, the first calculation mode is used to calculate the numerical flux at the unit interface; if the difference is greater than the preset threshold, the second calculation mode is used to calculate the numerical flux at the unit interface; wherein, the computational complexity of the second calculation mode is higher than that of the first calculation mode.
[0009] Step S4: Based on the numerical flux calculated in step S3, advance the flow field time integration to realize the evolution simulation of the aircraft flow, and output the flow field parameters for analyzing aerodynamics and aerothermal properties.
[0010] The core of this method lies in a simple adaptive judgment mechanism. In the flux reconstruction process based on feature projection, two sets of reconstruction weights corresponding to different flow modes are first obtained. Then, the difference Δω between these weights is calculated and compared with a preset threshold ε0. Based on the comparison result, different numerical flux calculation paths are adaptively selected: if Δω≤ε0, it is determined to be a smooth or weak shear region, and a simplified mode with lower computational cost is adopted. The weak shear and smooth flow regions include, but are not limited to: uniform inflow regions, the far-field region of the aircraft, the boundary layer external potential flow region, and a large-scale smooth pressure region on the aircraft surface. If Δω>ε0, it is determined to be a critical region containing shock waves, strong gradients, or turbulent fluctuations, and a complete mode including acoustic mode correction terms, which has higher computational cost but greater accuracy, is adopted. The aforementioned method addresses the trade-off between turbulence resolution and computational cost in aircraft flow simulations. It employs a low-cost decision step to maintain high analytical accuracy in highly nonlinear regions crucial for aerodynamic / thermal predictions (such as boundary layer transition zones, shock wave / boundary layer interference zones, and near-wall turbulent structures), while significantly reducing computational load in vast, smooth flow fields. This enables more refined numerical simulations of aircraft with the same computational resources, yielding higher-confidence aerodynamic and aerothermal data. This facilitates more sophisticated aircraft design, improves the adaptive flux efficiency of high-resolution aircraft flow simulations, and ultimately enhances the computational efficiency of full-field aircraft simulations.
[0011] Preferably, the first group of weights is ω k,1 The weight of the second group is ω k,2 , where k is the sub-template index.
[0012] The preferred method for calculating the difference is as follows:
[0013] Δω = max(|ω k,1 - ω k,2 |);
[0014] Wherein, Δω represents the difference quantity. Calculating the difference quantity by taking the maximum value allows for more sensitive capture of any local, isolated strong discontinuities or singularities in the flow field, ensuring the robustness of the numerical scheme under extreme conditions such as shock waves. Calculating the difference quantity by taking the average value provides a more stable reflection of the overall nonlinear intensity of a region, is less affected by numerical oscillations at individual grid points, and is suitable for large eddy simulations (LES) that are sensitive to noise. This invention provides these two calculation methods, allowing users to flexibly choose the most suitable criterion based on specific simulation objectives (such as robust capture of shock waves or high-fidelity analytical turbulence).
[0015] The preferred method for calculating the difference is as follows:
[0016] Δω = average(|ω k,1 - ω k,2 |);
[0017] Where Δω is the difference.
[0018] Preferably, the numerical flux calculation method corresponding to the first calculation mode is as follows:
[0019] F = Σ(ω k,1 · q k );
[0020] Where F is the numerical flux, q k This is the interpolation of physical quantities at the unit interface obtained based on the reconstruction of the k-th sub-template.
[0021] Preferably, the numerical flux calculation method corresponding to the second calculation mode is as follows:
[0022] F = Σ(ω k,1 · q k ) + C;
[0023] Where F is the numerical flux, q k C is the interpolation of physical quantities at the unit interface obtained based on the reconstruction of the k-th sub-template, and C is the interpolation based on the weight difference ω. k,2 - ω k,1 The calculation correction term. This correction term is activated only when a critical physical region is identified, and is used to compensate for the acoustic modal effects neglected in the simplified calculation, so as to maintain the ability to capture shock waves and turbulent fluctuations at high resolution.
[0024] Preferably, the flow region corresponding to the numerical flux at the unit interface is calculated using the first calculation mode in the following regions: a uniform inflow region, a far-field region of the aircraft, a boundary layer external potential flow region, or a smooth pressure region on the aircraft surface. In regions such as uniform inflow, far-field region of the aircraft, boundary layer external potential flow, or smooth pressure region, the gradient of flow field variables is small, the physical process is stable, and higher-order nonlinear effects are weak. Using the first calculation mode in these regions allows for the neglect of higher-order correction terms with negligible impact on accuracy, thereby significantly reducing the number of floating-point operations and memory access overhead, achieving computational acceleration without sacrificing the accuracy required for engineering. This is precisely the key to this method's ability to allocate computational resources on demand and improve overall simulation efficiency.
[0025] Preferably, the flow region corresponding to the numerical flux at the unit interface is calculated using the second calculation mode as: the boundary layer transition region, the shock wave interference region, the boundary layer interference region, or the near-wall turbulent structure region.
[0026] The first and second calculation modes of this invention are not simple formula substitutions, but rather independently designed based on a profound understanding of the physical meaning of nonlinear weights in flux reconstruction. The core idea is to use the difference between two sets of weights, sensitive to shear and acoustic modes respectively, as a benchmark to measure the local flow nonlinear complexity. When the difference is small, it means that the shear and acoustic modes are decoupled, and the flow can be approximated as a shear-dominated smooth flow, allowing the simplified first mode to be used. When the difference is large, it means that the shear and acoustic modes are strongly coupled, with shock waves or strong compression / expansion effects, requiring the use of the second mode, which includes an acoustic correction term C, to ensure physical accuracy. This design precisely matches computational overhead with the flow's physical complexity, achieving an optimal balance between computational efficiency and simulation fidelity.
[0027] Preferably, the preset threshold is ε0, and the value of ε0 ranges from 0.2 to 0.3. The preset threshold ε0 is a key parameter of this method, and its value directly determines the sensitivity of adaptive switching. Extensive numerical experiments on typical supersonic vehicle flow fields (such as shock wave-boundary layer interference, compression corner flow, etc.) have revealed that if ε0 is too small (e.g., <0.1), too many mesh regions will be misclassified as critical regions, leading to the activation of the computationally expensive second mode and a loss of computational efficiency. If ε0 is too large (e.g., >0.4), some regions containing weak shock waves or strong shear may be misclassified as smooth regions, resulting in a decrease in the analytical accuracy of key flow features. Setting the value of ε0 to 0.2 to 0.3 is the optimal balance range determined after extensive testing, which can maximize the improvement of computational efficiency while ensuring high-fidelity capture of key structures such as shock waves and separation bubbles.
[0028] Preferably, the flux reconstruction process in step S1 is a flux reconstruction process based on dimensionality-reduced feature projection. The purpose of this flux reconstruction process in step S1 is to project the original high-dimensional conserved variable system into a two-dimensional feature subspace spanned by shear modes and acoustic modes. Performing nonlinear weight calculations within this subspace not only significantly reduces computational and storage overhead but also more clearly separates the contributions of different physical modes, thus making the adaptive criterion based on the difference between the two sets of weights more physically relevant and accurate. This design is the physical basis for achieving both high efficiency and high accuracy in this method.
[0029] One or more technical solutions provided by this invention have at least the following technical effects or advantages:
[0030] To address the trade-off between turbulence resolution and computational cost in aircraft flow simulations, a low-cost decision step is employed to maintain high analytical accuracy in strongly nonlinear regions crucial for aerodynamic / thermal predictions (such as boundary layer transition regions, shock wave / boundary layer interference regions, and near-wall turbulent structures), while significantly reducing computational load in vast smooth flow fields. This enables more refined numerical simulations of aircraft with the same computational resources, yielding higher-confidence aerodynamic and aerothermal data, thus facilitating more sophisticated aircraft design. Attached Figure Description
[0031] The accompanying drawings, which are provided to further illustrate embodiments of the invention and constitute a part of this invention, are not intended to limit the scope of the invention.
[0032] Figure 1 This is a flowchart illustrating an adaptive flux calculation method for high-resolution simulation of aircraft flow.
[0033] Figure 2 The shock wave reflection density cloud map is calculated using three formats: WENO, TENO, and WENN, for the complete model. Detailed Implementation
[0034] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, where there is no conflict, the embodiments of the present invention and the features thereof can be combined with each other.
[0035] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.
[0036] Please refer to Figure 1 , Figure 1 This invention provides an adaptive flux calculation method for high-resolution simulation of aircraft flow, applicable to the description of turbulent structures around an aircraft. The method includes:
[0037] Step S1: On the computational grid describing the local configuration of the aircraft surface, two sets of nonlinear reconstruction weights are obtained at the cell interfaces in the computational grid through a flux reconstruction process, namely the first set of weights and the second set of weights.
[0038] Step S2: Calculate the difference between the first set of weights and the second set of weights;
[0039] Step S3: If the difference is less than or equal to a preset threshold, the first calculation mode is used to calculate the numerical flux at the unit interface; if the difference is greater than the preset threshold, the second calculation mode is used to calculate the numerical flux at the unit interface; wherein, the computational complexity of the second calculation mode is higher than that of the first calculation mode.
[0040] Step S4: Based on the numerical flux calculated in step S3, advance the flow field time integration to realize the evolution simulation of the aircraft flow, and output the flow field parameters for analyzing aerodynamics and aerothermal properties.
[0041] To address the efficiency bottleneck of high-resolution flow simulation in existing aircraft fine-grained design, this invention provides an adaptive flux calculation method. Its core objective is not to create a new reconstruction format, but rather to introduce a general, lightweight decision layer based on dimensionality reduction reconstruction. This decision layer can distinguish regions with different physical properties in the flow field based on intermediate calculation results and adaptively select the most economical calculation path for these regions. This achieves high accuracy in critical regions (such as near-wall turbulence and shock wave regions) and high efficiency in smooth regions such as uniform inflow and potential flow, ultimately significantly improving the computational efficiency of full-field aircraft simulation.
[0042] This method can dynamically identify critical regions in the flow field that require high-precision analysis of physical processes and smooth regions where simplified calculations are acceptable. This maximizes overall computational efficiency while ensuring the accuracy of simulations of key physical processes, making large-scale, high-resolution simulations of complex flows such as turbulence on aircraft surfaces possible in engineering practice. The adaptive flux calculation for the shock wave capture format aims to efficiently and at high resolution simulate complex physical phenomena such as turbulent structures and shock wave / boundary layer disturbances on aircraft surfaces, in order to accurately predict aerodynamic forces and aerothermal activity. The core workflow of this method includes:
[0043] Obtaining the weights: On the computational grid, the flux is projected into a two-dimensional feature subspace. Within this subspace, two sets of weights ω are obtained through nonlinear reconstruction processes such as WENO, TENO, or WENN. k,1 ω k,2 .
[0044] Calculate and determine the difference: Calculate the degree of difference Δω between these weights. This difference is used as an indirect indicator of the strength of local flow nonlinearity. Compare Δω with an empirical threshold ε0.
[0045] Adaptive computation:
[0046] Smooth / weak shear region (Δω≤ε0): This region is characterized by smooth flow with minimal turbulent fluctuations or shock wave effects. In this case, a simplified mode is used, employing the flux synthesis formula with minimal computational cost and ignoring higher-order correction terms that have little impact on accuracy. This corresponds to the mainstream flow region and uniform inflow region far from the wall in the aircraft flow field.
[0047] Key physical regions (Δω>ε0): These are identified as areas with shock waves, strong gradients, or significant turbulent fluctuations. At this point, a complete set of flux synthesis formulas, including all necessary corrections, is employed to ensure accurate resolution of turbulent structures within the boundary layer, shock wave locations, separation bubbles, and other physical phenomena crucial for aerodynamic / thermal prediction. This corresponds to aircraft nose shock waves, wing leading edges, shock wave / boundary layer interference zones, and near-wall turbulent regions.
[0048] The construction of the computational grid includes: obtaining a three-dimensional geometric model of the aircraft, determining the range and boundaries of the fluid region, and obtaining the corresponding computational grid by inputting relevant parameters of the aircraft based on ANSYS series software.
[0049] In this embodiment, the specific implementation steps for obtaining two sets of nonlinear reconstruction weights through the flux reconstruction process in step S1 are as follows:
[0050] First, at any cell interface of the computational grid, construct the Jacobian matrix at that interface based on the average value of the conserved variables on both sides of the interface, and solve for the left and right eigenvector matrices of this matrix. Using the right eigenvector matrix, project the conserved variables or eigenvalues to be reconstructed into the eigenspace.
[0051] Secondly, perform dimensionality reduction feature projection: based on the physical meaning of the eigenvalues, select specific feature directions corresponding to shear waves and acoustic waves to construct a two-dimensional dimensionality reduction feature subspace. The feature space variables obtained in the previous step are then projected onto this subspace, thereby achieving preliminary separation of shear modes and acoustic modes.
[0052] Next, independent nonlinear reconstruction is performed: within the reduced-dimensional subspace, a high-resolution reconstruction scheme (such as WENO, TENO, or WENN scheme) is applied to each of the two separated modal components. For each modal component, the scheme generates a set of nonlinear weights for several candidate sub-templates based on the smoothness of the local flow field. Thus, a set of nonlinear weights is obtained for the shear modal component, and another set of nonlinear weights is obtained for the acoustic modal component.
[0053] In this embodiment of the invention, the first set of weights is ω. k,1 The weight of the second group is ω k,2 , where k is the sub-template index.
[0054] In this embodiment of the invention, the first method for calculating the difference is to take |ω| on all sub-templates k. k,1 -ω k,2 The maximum value of | is:
[0055] Δω = max(|ω k,1 - ω k,2 |);
[0056] Where Δω is the difference.
[0057] In this embodiment of the invention, the second method for calculating the difference is to take |ω| on all sub-templates k. k,1 -ω k,2 The average value of |, that is:
[0058] Δω = average(|ω k,1 - ω k,2 |);
[0059] Where Δω is the difference.
[0060] In this embodiment of the invention, the calculation method for the numerical flux corresponding to the first calculation mode is as follows:
[0061] F = Σ(ω k,1 · q k );
[0062] Where F is the numerical flux, q k This is the interpolation of physical quantities at the unit interface obtained based on the reconstruction of the k-th sub-template.
[0063] In this embodiment of the invention, the calculation method for the numerical flux corresponding to the second calculation mode is as follows:
[0064] F = Σ(ω k,1 · q k ) + C;
[0065] Where F is the numerical flux, q k C is the interpolation of physical quantities at the unit interface obtained based on the reconstruction of the k-th sub-template, and C is the interpolation based on the weight difference ω. k,2 - ω k,1 The correction term for the calculation.
[0066] The specific calculation method for this correction term is as follows:
[0067] In the formula, A relaxation factor between 0.5 and 1.5 is used to control the strength of the correction. This is a function related to the sub-template k and the flow field variable φ, used to characterize the contribution of the acoustic modes to the numerical flux. It should be noted that the difference Δω is a scalar index calculated based on the difference between the two sets of weights to determine the flow state, while the correction term C is a physical quantity correction value constructed using the specific distribution of the difference between the two sets of weights to correct the numerical flux.
[0068] In this embodiment of the invention, the flow region corresponding to the numerical flux at the unit interface is calculated using a first calculation mode as: a uniform inflow region, a far-field region of the aircraft, a boundary layer external potential flow region, or a smooth pressure region on the surface of the aircraft.
[0069] In this embodiment of the invention, the flow region corresponding to the numerical flux at the unit interface is calculated using a second calculation mode as: a boundary layer transition region, a shock wave interference region, a boundary layer interference region, or a near-wall turbulent structure region.
[0070] In this embodiment of the invention, the preset threshold is ε0, and the value of ε0 ranges from 0.2 to 0.3.
[0071] In this embodiment of the invention, the flux reconstruction process in step S1 is a flux reconstruction process based on dimensionality reduction feature projection. Specifically, in this embodiment, the process of obtaining the two sets of nonlinear reconstruction weights in step S1 is as follows:
[0072] 1. Feature Projection: At the unit interface, first calculate the Jacobian matrix and its left and right eigenvector matrices of the interface based on the average values of the conserved variables on the left and right sides. Using the right eigenvector matrix, project the physical quantities to be reconstructed (such as original variables or characteristic variables) into the feature space.
[0073] 2. Dimensionality Reduction: Based on the physical meaning of the eigenvalues, specific feature directions corresponding to shear waves and sound waves are selected to construct a two-dimensional reduced-dimensional feature subspace. The variables after feature projection are further projected into this subspace to separate different physical modes.
[0074] 3. Independent Reconstruction: Within the reduced-dimensional subspace, for two selected modal components, classical high-resolution reconstruction schemes (such as WENO, TENO, or WENN schemes) are applied respectively. For each modal component, the scheme generates a set of nonlinear weights for its candidate reconstruction template based on the smoothness of the local flow field. Thus, we obtain a set of nonlinear weights ω for the main response shear modes. k,1 and a set of nonlinear weights ω for the main response acoustic modes k,2 .
[0075] Please refer to Figure 2 , Figure 2 The shock wave reflection density contour map is calculated using three formats: WENO, TENO, and WENN, for the entire model. Figure 2 (a) and (b) are results in WENO format; (c) and (d) are results in TENO format; (e) and (f) are results in WENN format.
[0076] Please refer to Table 1, which is a comparison of the computation time of the method in this invention and the traditional unified algorithm used across the entire domain. As can be seen from Table 1, the method of this invention has a significant improvement in efficiency.
[0077] Table 1
[0078] conventional methods 748s 2132s 1012s Method of the present invention 582s 1044s 545s Improve efficiency 22.2% 51.0% 46.1%
[0079] This invention directly addresses the core pain point of excessively high costs in high-resolution turbulence simulation in aircraft design, providing an effective efficiency optimization solution. It achieves on-demand allocation of computing resources, conserving computational power in the predominantly smooth regions and concentrating resources on key local areas that play a decisive role in aerodynamic performance prediction. This makes it possible to perform LES / DNS-type simulations with higher grid resolution and longer time scales given computing resources, providing technical support for obtaining high-confidence aerodynamic and thermodynamic data, directly serving the refined and optimized design of aircraft. The adaptive judgment logic itself has extremely low computational overhead and can be widely applied to various high-order formats, making it an efficient and practical acceleration technology.
[0080] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.
[0081] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. An adaptive flux calculation method for high-resolution simulation of aircraft flow, characterized in that, The method includes: Step S1: On the computational grid describing the local configuration of the aircraft surface, two sets of nonlinear reconstruction weights are obtained at the cell interfaces in the computational grid through a flux reconstruction process, namely the first set of weights and the second set of weights. Step S2: Calculate the difference between the first set of weights and the second set of weights; Step S3: If the difference is less than or equal to a preset threshold, the first calculation mode is used to calculate the numerical flux at the unit interface; if the difference is greater than the preset threshold, the second calculation mode is used to calculate the numerical flux at the unit interface; wherein, the computational complexity of the second calculation mode is higher than that of the first calculation mode. Step S4: Based on the numerical flux calculated in step S3, advance the flow field time integration to realize the evolution simulation of the aircraft flow and output the flow field parameters for analyzing aerodynamics and aerothermal properties.
2. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The first set of weights is ω k,1 and the second set of weights is ω k,2 where k is a sub-template index.
3. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 2, characterized in that, The method for calculating the difference is as follows: Δω = max(|ω k,1 - ω k,2 |) Where Δω is the difference.
4. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 2, characterized in that, The method for calculating the difference is as follows: Do = average(|ω) k,1 - oh k,2 |); Where Δω is the difference.
5. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The numerical flux calculation method corresponding to the first calculation mode is as follows: F = Σ(ω k,1 · q k ); Where F is the numerical flux, q k This is the interpolation of physical quantities at the unit interface obtained based on the reconstruction of the k-th sub-template.
6. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The calculation method for numerical flux corresponding to the second calculation mode is as follows: F = Σ(ωk,1 · q k ) + C; Where F is the numerical flux, q k C is the interpolation of physical quantities at the unit interface obtained based on the reconstruction of the k-th sub-template, and C is the interpolation based on the weight difference ω. k,2 - ω k,1 The correction term for the calculation.
7. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The first calculation mode is used to calculate the flow region corresponding to the numerical flux at the unit interface, which is: a uniform inflow region, a far-field region of the aircraft, a boundary layer external potential flow region, or a smooth pressure region on the surface of the aircraft.
8. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The second calculation mode is used to calculate the flow region corresponding to the numerical flux at the unit interface, which is: the boundary layer transition region, the shock wave interference region, the boundary layer interference region, or the near-wall turbulent structure region.
9. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The preset threshold is ε0, and the value of ε0 ranges from 0.2 to 0.
3.
10. The adaptive flux calculation method for high-resolution simulation of aircraft flow according to claim 1, characterized in that, The flux reconstruction process in step S1 is a flux reconstruction process based on dimensionality reduction feature projection.