A hybrid solution method for computational fluid dynamics software structure solvers

Abstract

This invention provides a hybrid solution method for computational fluid dynamics software structural solvers. The method first determines whether some model equations employ a loosely coupled approach, specifying that the number of conserved quantities equals the number of tightly coupled conserved quantities. Then, it iteratively solves the problem block by block, successively solving both tightly coupled and loosely coupled equation sets. Finally, it merges the loosely coupled and tightly coupled conserved quantities until all grid blocks have been iteratively solved. This method simultaneously implements both tightly coupled and loosely coupled architectures within a single data structure, enabling the software to simultaneously achieve high solution accuracy, high efficiency, and scalability.

Description

Technical Field

[0001] This invention belongs to the field of computational fluid dynamics technology, and particularly relates to a hybrid solution method for computational fluid dynamics software structure solvers. Background Technology

[0002] The design and development of aircraft places comprehensive demands on computational fluid dynamics (CFD) software. Aircraft undergo a range of aerodynamic physical challenges, from ground takeoff to high-altitude, high-speed operation, including laminar flow, turbulence, transition, engine combustion, high-temperature gas effects, dynamic characteristics, and multibody separation. The core of CFD software is to solve the Navier-Stokes equations, given boundary and initial field conditions, to obtain the fluid flow development pattern. To address the aerodynamic physical problems involved in aircraft development, it is necessary to add various physical models to the existing Navier-Stokes equations, such as turbulence models, transition models, chemical nonequilibrium models, thermal nonequilibrium models, and models with overlapping mesh processing.

[0003] Currently, in CFD software, the Navier-Stokes equations can be solved using either a tightly coupled or loosely coupled solution architecture with other model equations. Taking the turbulence model as an example, if a tightly coupled solution strategy is used, the turbulence model equations and the Navier-Stokes equations are solved together in each iteration step. If a loosely coupled solution strategy is used, the turbulence model equations and the Navier-Stokes master equations are solved separately. After the turbulence model equations are solved, the turbulent viscosity coefficient is transferred to the master equations before the master equations are solved. After the master equations are solved, the conserved flow field information is transferred to the turbulence model for the next iteration step. The tightly coupled strategy offers high accuracy and efficiency, but the software is difficult to extend, has poor maintainability, and poor robustness. The loosely coupled strategy makes the software easier to extend and maintain, but the solution accuracy and efficiency are correspondingly reduced. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0005] According to one aspect of the present invention, a hybrid solution method for computational fluid dynamics software structure solvers is provided, the method comprising:

[0006] (1) Input condition parameters;

[0007] (2) Allocate variable storage space according to the solution requirements of all physical model equations and the mainstream NS equations;

[0008] (3) Enter the solution stage and determine whether some model equations adopt the loose coupling treatment method. If so, specify the number of conserved quantities equal to the tightly coupled conserved quantities and execute step (4); otherwise, execute step (4).

[0009] (4) Solve for the time step;

[0010] (5) Select a grid block;

[0011] (6) Calculate the inviscid flux of the main NS equation; calculate the inviscid flux of the tightly coupled physical model equation; calculate the Reynolds stress based on the viscosity coefficient of the previous time step; calculate the viscous flux of the main NS equation; calculate the viscous flux of the tightly coupled physical model; solve the source terms of the tightly coupled physical model.

[0012] (7) Conduct time-progression solution;

[0013] (8) Complete the tightly coupled equation system and obtain the results of the tightly coupled conservation quantities;

[0014] (9) Determine whether some model equations adopt the loose coupling treatment method. If so, start solving each loosely coupled physical model equation separately, obtain the loosely coupled conservation results, and execute step (10); if not, execute step (10).

[0015] (10) Merge loosely coupled and tightly coupled conserved quantities, place the values ​​of the loosely coupled conserved quantity array into the tightly coupled conserved quantity array space, and obtain the conserved quantities of all equations; update the conserved quantities of all equations in the entire flow field; restore the number of conserved quantities to the sum of the loosely coupled and tightly coupled conserved quantities;

[0016] (11) Determine whether all mesh blocks have been solved. If yes, proceed to step (12); otherwise, select the next mesh block in sequence and return to step (6) until all mesh blocks have been solved.

[0017] (12) Unify the updating of gas variables and boundary information;

[0018] (13) Determine whether convergence has occurred. If not, proceed to the next time step and return to step (4). If yes, output the calculation result and complete the solution.

[0019] According to another aspect of the present invention, a computer program product is provided, comprising a computer program or instructions that, when executed by a processor, implement the steps of the hybrid solution method as described above.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the hybrid solution method as described above.

[0021] This invention provides a hybrid solution method for computational fluid dynamics software structural solvers. The method first determines whether some model equations employ loose coupling, specifies that the number of conserved quantities equals the number of tightly coupled conserved quantities, and then iteratively solves each mesh block, successively solving both tightly coupled and loosely coupled equation sets. Finally, it merges the loosely coupled and tightly coupled conserved quantities until all mesh blocks are solved iteratively. This invention simultaneously implements both tightly coupled and loosely coupled architectures within a single data structure, enabling the software to simultaneously achieve high solution accuracy, high efficiency, and scalability. Attached Figure Description

[0022] The accompanying drawings, which form part of this specification, are provided to further illustrate embodiments of the invention and, together with the textual description, explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0023] Figure 1 A flowchart illustrating a hybrid solution method for a computational fluid dynamics software structure solver, according to a specific embodiment of the present invention, is shown. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0026] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0027] like Figure 1 As shown in the figure, a hybrid solution method for computational fluid dynamics software structure solvers is provided according to a specific embodiment of the present invention. The method includes:

[0028] (1) Input condition parameters;

[0029] (2) Allocate variable storage space according to the solution requirements of all physical model equations and the mainstream NS equations;

[0030] (3) Enter the solution stage and determine whether some model equations adopt the loose coupling treatment method. If so, specify the number of conserved quantities equal to the tightly coupled conserved quantities and execute step (4); otherwise, execute step (4).

[0031] (4) Solve for the time step;

[0032] (5) Select a grid block;

[0033] (6) Calculate the inviscid flux of the main NS equation; calculate the inviscid flux of the tightly coupled physical model equation; calculate the Reynolds stress based on the viscosity coefficient of the previous time step; calculate the viscous flux of the main NS equation; calculate the viscous flux of the tightly coupled physical model; solve the source terms of the tightly coupled physical model.

[0034] (7) Conduct time-progression solution;

[0035] (8) Complete the tightly coupled equation system and obtain the results of the tightly coupled conservation quantities;

[0036] (9) Determine whether some model equations adopt the loose coupling treatment method. If so, start solving each loosely coupled physical model equation separately, obtain the loosely coupled conservation results, and execute step (10); if not, execute step (10).

[0037] (10) Merge loosely coupled and tightly coupled conserved quantities, place the values ​​of the loosely coupled conserved quantity array into the tightly coupled conserved quantity array space, and obtain the conserved quantities of all equations; update the conserved quantities of all equations in the entire flow field; restore the number of conserved quantities to the sum of the loosely coupled and tightly coupled conserved quantities;

[0038] (11) Determine whether all mesh blocks have been solved. If yes, proceed to step (12); otherwise, select the next mesh block in sequence and return to step (6) until all mesh blocks have been solved.

[0039] (12) Unify the updating of gas variables and boundary information;

[0040] (13) Determine whether convergence has occurred. If not, proceed to the next time step and return to step (4). If yes, output the calculation result and complete the solution.

[0041] This configuration provides a hybrid solution method for computational fluid dynamics software structural solvers. The method first determines if any model equations employ loose coupling, assigns a number of conserved quantities equal to the number of tightly coupled conserved quantities, and then iteratively solves each mesh block, successively solving both tightly coupled and loosely coupled equation sets. Finally, it merges the loosely coupled and tightly coupled conserved quantities until all mesh blocks are solved iteratively. This invention achieves both tightly coupled and loosely coupled architectures within a single data structure, enabling the software to simultaneously maintain solution accuracy, efficiency, and scalability.

[0042] The hybrid architecture solution method combining tight and loose coupling has three operating modes in practice: fully tightly coupled mode, fully loosely coupled mode, and partially tightly coupled and partially loosely coupled mode. In the fully tightly coupled mode, the main NS equations are combined with other physical model equations into a large system of equations for solution. First, the inviscid flux, viscous flux, and source term of each equation are solved to obtain the right-hand side terms, and then the overall solution time is advanced iteratively. In the fully loosely coupled mode, the main NS equations are solved completely separately from the other physical model equations. After each iteration, when updating the conserved variables, the influence of the main NS equations on the other physical model equations is transferred. The partially tightly coupled and partially loosely coupled mode involves solving some physical models coupled to the NS equations and others decoupled from them.

[0043] One of the core challenges of hybrid solution methods is how to simultaneously implement tightly coupled and loosely coupled architectures under a single data structure.

[0044] If a loosely coupled architecture is adopted, the mainstream Navier-Stokes equations have 5 conserved quantities. The component equation has several conserved quantities, while the turbulence equation has one or two conserved quantities. Conserved quantities in the model equations of the interaction between turbulence and chemical reaction Conserved quantities in the transition model equations These can be set as different arrays, independent of each other, with the influence propagated after each solution step. The loosely coupled conserved quantity array form is shown below:

[0045]

[0046] Where ρ is density, u1, u2, u3 are velocity components, E is total energy, and Y is... n Re represents the mass fraction of the component, θ1 and θ2 are both turbulent transport quantities, and Re θ Both γ1 and γ2 are transition model variables, γ1 and γ2 are fluctuation quantities in the turbulence-chemical reaction interaction model, and J is the Jacobian matrix.

[0047] If a tightly coupled architecture is adopted, the five conserved quantities in the mainstream NS equations And multiple conserved quantities in the component equation, and 1 to 2 conserved quantities in the turbulence equation. Four conserved quantities in the transition model equation Conserved quantities in the model equations of the interaction between turbulence and chemical reaction They need to be combined into a single array. The tightly coupled array of conserved quantities is shown below:

[0048]

[0049] or

[0050]

[0051] In this invention, the core method of the hybrid architecture is as follows: firstly, it is assumed that all equations are solved in a tightly coupled form; when allocating the space for conserved quantities, the array space for conserved quantities is allocated in a fully tightly coupled architecture.

[0052] During the solution process, each grid block is looped, and the number of conserved quantities contained in the conserved quantity array q is dynamically adjusted during the solution process, but the actual space occupied by the array remains unchanged.

[0053] When solving tightly coupled equations, only the conserved quantities of the tightly coupled physical model equations contained in the array are processed. Among them, the conserved quantities ρθ1, ρθ2, ρRe in the loosely coupled physical model equations are... θ The value of ρr remains unchanged.

[0054] When solving the loosely coupled physical model equations, only the conserved quantities ρθ1, ρθ2, ρRe contained in the array q are processed. θAfter solving the equation, ρr combines the loosely coupled and tightly coupled conserved quantities, and the conserved quantity becomes again...

[0055] After completing the current iteration calculation, continue to update the boundary information and gas variables in a fully tightly coupled architecture.

[0056] By following the above process, computational efficiency and accuracy can be guaranteed, while greatly improving the scalability and maintainability of the software.

[0057] According to another aspect of the present invention, a computer program product is provided, comprising a computer program or instructions that, when executed by a processor, implement the steps of the hybrid solution method as described above.

[0058] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the hybrid solution method as described above.

[0059] To gain a further understanding of the present invention, the following description is provided in conjunction with... Figure 1 The hybrid solution method of the present invention for a computational fluid dynamics software structure solver is described in detail.

[0060] This embodiment provides a hybrid solution method for computational fluid dynamics software structure solvers, specifically including the following steps.

[0061] (1) Input the grid, gas model, incoming flow parameters, physical property parameters, chemical reaction mechanism and other conditions.

[0062] (2) Based on the requirement of tightly coupled solution of all physical model equations and mainstream NS equations, storage space is allocated for all conserved quantities, gas variables, geometric quantities, etc. Taking the transition calculation considering the high-temperature gas effect as an example, the conserved quantities... That is, the conserved quantities in the physical model equations and the conserved quantities in the mainstream Navier-Stokes equations are uniformly allocated to q_all(num_q), where num_q = num_q_all (the total number of all conserved quantities), and num_q_all = num_qj (the number of tightly coupled conserved quantities) + num_qs (the number of loosely coupled conserved quantities). If a fully tightly coupled architecture is adopted, num_qs = 0.

[0063] (3) Begin iterative solution, and solve the equation as follows: Where t is the time term, For inviscid flux, including F~ iz Mainstream Navier-Stokes equations for inviscid flux Inviscid flux in tightly coupled models Inviscid flux in loosely coupled model, ξ i For curve coordinates, For viscous flux, including D~ z Mainstream Navier-Stokes equations for viscous flux, Viscous flux in tightly coupled models Viscous flux in loosely coupled models For source terms, including Source terms of tightly coupled model Loosely coupled model source terms.

[0064] Determine whether some model equations adopt the loose coupling method, that is, determine whether num_qs is not equal to 0. If yes, specify the number of conserved quantities equal to the tightly coupled conserved quantities num_q = num_qj, and execute step (4); otherwise, execute step (4).

[0065] (4) Solve for the time step.

[0066] (5) Select a grid block.

[0067] (6) Calculate the inviscid flux F~ of the main Navier-Stokes equation. iz ; Calculate the inviscid flux F in the tightly coupled physical model equations. ij The Reynolds stress is calculated based on the viscosity coefficient at the previous time step; the viscous flux of the main Navier-Stokes equations is calculated. Calculate the viscous flux of a tightly coupled physical model Solving the source terms of a tightly coupled physical model

[0068] (7) Conduct time-based solution.

[0069] (8) Complete the tightly coupled equation system and obtain the tightly coupled conserved quantity result q_all(num_q).

[0070] (9) Determine whether some model equations adopt the loose coupling processing method, that is, determine whether num_qs is not equal to 0. If yes, the loosely coupled physical model equations are solved separately, including gradient calculation, viscous flux calculation, inviscous flux calculation, source term solution, and time-progression iteration, to obtain the loosely coupled conserved quantity result qs(num_qs), and execute step (10); if no, execute step (10).

[0071] (10) Combine loosely coupled and tightly coupled conserved quantities, place the values ​​of the qs array into the q array space, and obtain the conserved quantities q_all(num_q+num_qs) of all equations;

[0072] Update the conserved quantities in all equations of the entire flow field;

[0073] The number of conserved quantities is restored to num_q = num_qj + num_qs.

[0074] (11) Determine whether all grid blocks have been solved; if yes, proceed to step (12); if no, select the next grid block in sequence and return to step (6) until all grid blocks have been solved.

[0075] (12) Update gas variables and boundary information in a unified manner.

[0076] (13) Determine whether convergence has occurred. If not, proceed to the next time step and return to step (4). If yes, output the calculation result and complete the solution.

[0077] In summary, this invention provides a hybrid solution method for computational fluid dynamics software structural solvers. This method first determines whether some model equations employ a loosely coupled approach, specifying that the number of conserved quantities equals the number of tightly coupled conserved quantities. Then, it iteratively solves each mesh block sequentially, solving both tightly coupled and loosely coupled equation sets. Finally, it merges the loosely coupled and tightly coupled conserved quantities until all mesh blocks are solved iteratively. This invention achieves both tightly coupled and loosely coupled architectures under a single data structure, enabling the software to simultaneously balance solution accuracy, solution efficiency, and software scalability.

[0078] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A hybrid solution method for a computational fluid dynamics software structure solver, characterized in that, The hybrid solving method for the solver of the computational fluid dynamics software structure comprises: (1) inputting condition parameters; (2) allocating variable storage spaces according to solving requirements of all physical model equations and the main flow NS equation tight coupling; (3) entering a solving link, judging whether some model equations adopt a loose coupling processing method, if yes, specifying that the number of conservative quantities is equal to the tight coupling conservative quantity, and executing step (4); if no, executing step (4); (4) solving a time step; (5) selecting a grid block; (6) calculating a main flow NS equation inviscid flux; calculating a tight coupling physical model equation inviscid flux; calculating Reynolds stress by using a viscous coefficient of a previous time step; calculating a main flow NS equation viscous flux; calculating a tight coupling physical model viscous flux; solving a tight coupling physical model source term; (7) developing a time advancing solution; (8) completing a tight coupling equation set, and obtaining a tight coupling conservative quantity result; (9) judging whether some model equations adopt a loose coupling processing method, if yes, separately starting to solve each loose coupling physical model equation, obtaining a loose coupling conservative quantity result, and executing step (10); if no, executing step (10); (10) merging the loose coupling conservative quantity and the tight coupling conservative quantity, placing a loose coupling conservative quantity array value in a tight coupling conservative quantity array space, obtaining all equation conservative quantities, updating all equation conservative quantities of the whole flow field, and restoring the number of conservative quantities to the sum of the loose coupling conservative quantity and the tight coupling conservative quantity; (11) judging whether all grid block solving is completed, if yes, executing step (12); if no, selecting a next grid block in sequence, returning to step (6), and until all grid block solving is completed; (12) uniformly updating gas variables and boundary information; (13) judging whether convergence is achieved, if no, entering a next time step, and returning to step (4); if yes, outputting a calculation result, and completing solving.

2. The hybrid solution approach to a computational fluid dynamics software structure solver according to claim 1, wherein, The condition parameters in step (1) comprise: a grid, a gas model, an incoming flow parameter, a physical property parameter, and a chemical reaction mechanism.

3. The hybrid solution approach to computational fluid dynamics software structure solver according to claim 1, wherein, The tight coupling physical model viscous flux calculation in step (6) comprises calculating a turbulent flow model equation, an energy equation, and a component equation viscous flux.

4. The hybrid solution approach to a computational fluid dynamics software structure solver according to claim 3, wherein, The tight coupling physical model source term solving in step (6) comprises solving a turbulent flow model, a component equation, a turbulent flow chemical reaction interaction model, and an ignition model source term.

5. The hybrid solution approach to a computational fluid dynamics software structure solver according to claim 4, wherein, The separately starting to solve each loose coupling physical model equation in step (9) comprises gradient calculation, viscous flux calculation, inviscid flux calculation, source term solving, and time advancing iteration.

6. A computer program product comprising computer programs or instructions, characterized in that, The computer program or instruction is executed by a processor to realize steps of the hybrid solving method in any one of claims 1 to 5.

7. A computer readable storage medium characterized in that, The computer program is stored on the computer readable storage medium, and is executed by a processor to realize steps of the hybrid solving method in any one of claims 1 to 5.

Images