A parallel computation method for parafoil fluid-structure coupling

Abstract

This invention discloses a parallel computation method for paraglider fluid-structure interaction (FSI) problems, comprising: flow field computation nodes reading their initial local mesh information according to the original order of the mesh points of the spatial background mesh; using ParMETIS for parallel partitioning of the flow field mesh; using MPI for data transfer so that each computation node obtains the actual mesh partitioning information; using METIS for mesh partitioning within each computation node; and using OpenMP to establish the geometric information required for flow field simulation in parallel. Solid preprocessing adopts a similar approach. The flow field computation uses the immersion boundary method to solve the compressible Navier-Stokes equations, utilizing MPI / OpenMP hybrid parallel technology, where MPI is used for data transfer between nodes and OpenMP is used for data transfer between threads within a node. This invention provides important support for the efficient computation of paraglider FSI problems.

Description

Technical Field

[0001] The invention relates to a fluid-solid coupling numerical simulation technology, in particular to a parafoil fluid-solid coupling parallel calculation method. Background Art

[0002] At present, parafoils, as a form of advanced parachute, have high lift-to-drag ratio and good handling characteristics, and have become an important aerodynamic deceleration device in the aerospace field, playing an indispensable role in precision airborne airdrops, etc. With the continuous improvement of aerospace technology requirements, higher requirements are also put forward for the performance of parafoils, so it is necessary to further develop advanced design, analysis and calculation methods of parafoils.

[0003] Similar to traditional parachutes, parafoils are made of flexible fabric materials. They are initially deployed in a folded state, then inflated and expanded in the air, undergoing large unsteady deformations, and their aerodynamic characteristics also undergo significant changes. After inflation is complete, their attitude and speed will gradually reach a balanced state, entering a stable gliding state, and finally achieving a precise and safe landing through deceleration.

[0004] During the operation of the parafoil, the parafoil undergoes huge structural deformation, which not only affects the surrounding flow field but is also affected by the flow field, that is, there is a very complex fluid-solid coupling phenomenon. For this type of problem, airdrop tests can accurately provide the true performance of the parafoil, but they are expensive and time-consuming. Therefore, numerical simulation methods based on computational fluid dynamics and structural dynamics are increasingly used as a research method for parafoils and play an increasingly important role.

[0005] Although the rapid development of computer technology has provided a hardware foundation for the numerical simulation of parafoil fluid-solid coupling, the computational complexity of fluid-solid coupling simulation is still a huge challenge. Therefore, it is necessary to combine the characteristics of parafoil "zero thickness" and large deformation, and conduct in-depth research on appropriate fluid-solid coupling parallel methods to achieve efficient and accurate parafoil fluid-solid coupling calculation and analysis. Summary of the invention

[0006] In order to efficiently numerically simulate the large deformation unsteady fluid-solid coupling process of a parafoil, the present invention provides a parallel calculation method for fluid-solid coupling of a parafoil, which fully utilizes the hardware environment of the current multi-core CPU cluster in an MPI+OpenMP hybrid distributed parallel manner, and provides an effective method for efficiently and accurately solving the fluid-solid coupling problem of the parafoil.

[0007] In order to achieve the above object, the technical solution provided by the present invention is: a parafoil fluid-solid coupling parallel calculation method, comprising the following steps:

[0008] Step 1, determining the number of flow field calculation nodes, the number of threads inside the flow field calculation nodes, the number of solid calculation nodes, and the number of threads inside the solid calculation nodes according to requirements;

[0009] Step 2: All flow field calculation nodes and solid calculation nodes read the flow field space background grid file simultaneously, and a process of each flow field calculation node and solid calculation node reads the initial grid information of each node according to the grid point sequence;

[0010] Step 3: Each flow field calculation node and solid calculation node reads the relevant spatial background grid unit information according to the initial local grid points, generates the required topological information, and calls ParMETIS for distributed parallel grid partitioning;

[0011] Step 4: According to the grid partitioning results, each flow field calculation node and solid calculation node uses MPI to transmit grid information and establish corresponding geometric and topological relationships;

[0012] Step 5: Each flow field calculation node and solid calculation node calls METIS to perform serial grid partitioning within the node;

[0013] Step 6: Each thread in the flow field calculation node and the solid calculation node uses OpenMP to establish the geometric relationship of each partition in parallel, including calculating the volume and normal vector geometric information of the dualmesh, reordering the mesh, and performing geometric verification;

[0014] Step 7, respectively establishing data transmission relationships between flow field calculation nodes and solid calculation nodes, wherein the data transmission relationships are used for MPI communication during flow field calculation;

[0015] Step 8, establishing the data transmission relationship required for fluid-solid coupling;

[0016] Step nine, perform fluid-solid coupling calculation; wherein, in the process of fluid-solid coupling calculation, the immersed boundary method is used to solve the compressible Navier-Stokes equations, the local one-dimensional half-Riemann problem is used to calculate the wall flux of the parafoil wall, MPI is used for data transmission between flow field calculation nodes and between solid calculation nodes, and OpenMP is used for coarse-grained parallelism within the flow field calculation nodes and solid calculation nodes.

[0017] As a preferred technical solution, the number of flow field calculation nodes, the number of solid calculation nodes, and the number of threads inside the flow field calculation nodes and solid calculation nodes in step one are positive integers greater than or equal to 1.

[0018] As a preferred technical solution, in step 2, each flow field calculation node and solid calculation node only reads its own initial grid information, and the master node does not need to process the global grid information.

[0019] As a preferred technical solution, the parallel grid partitioning in step three adopts the graph partitioning method of ParMETIS to perform distributed parallel grid partitioning, and ensures the internal continuity of each partition.

[0020] As a preferred technical solution, in the step five, each flow field calculation node and solid calculation node has its grid partitioned within it. Each flow field calculation node and solid calculation node uses METIS graph partitioning to perform serial grid partitioning according to the number of threads within the node, and ensures continuity within each partition.

[0021] As a preferred technical solution, the solid calculation in step eight adopts an independent computing node different from the flow field calculation node, the parafoil adopts finite element surface mesh discretization, ParMETIS is used for distributed parallel partitioning between the computing nodes, METIS is used for serial partitioning within the solid computing node, and all threads calculate local geometric information.

[0022] As a preferred technical solution, the MPI data transmission in step nine adopts non-blocking communication for overlapping communication and calculation.

[0023] As a preferred technical solution, MPI is used for data transmission between the flow field computing nodes and the solid computing nodes.

[0024] As a preferred technical solution, the flow field and solid computing nodes respectively use separate programs, and coupling is achieved by transferring data between different programs through MPI.

[0025] The beneficial effects of the present invention relative to the prior art are:

[0026] (1) The present invention fully utilizes the hardware environment of a multi-core CPU cluster in an MPI+OpenMP hybrid parallel mode, and can efficiently solve the large deformation unsteady fluid-solid coupling problem of a parafoil;

[0027] (2) The present invention has no direct restrictions on the number of nodes and the number of threads within a node, and is only limited by the computing hardware environment and the operating system;

[0028] (3) The present invention has no restrictions on the parafoil structure and spatial flow field and has wide applicability;

[0029] (4) In the present invention, flow field calculation and solid calculation use separate programs respectively, and data coupling exchange is performed through MPI, which is more efficient than data exchange in the file reading and writing mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flow chart of a parafoil fluid-solid coupling parallel calculation method provided by the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] Reference Figure 1 This embodiment provides a parallel calculation method for parafoil fluid-structure coupling, which specifically includes the following steps:

[0033] Step 1: Specify the number of flow field computing nodes, the number of threads inside the flow field computing nodes, the number of solid computing nodes, and the number of threads inside the solid computing nodes. There is no direct limit on the number of computing nodes and the number of threads inside the computing nodes. Generally, the number of flow field computing nodes is greater than the number of solid computing nodes. The number of threads inside the node is set to the number of physical CPU cores in the node.

[0034] Step 2: In the pre-processing stage, all flow field calculation nodes read the flow field space background grid file at the same time. A process of each calculation node reads the initial grid information of each node according to the original order of the space background grid, that is, the space background grid is evenly distributed to all flow field calculation nodes according to the original order of the grid points. The space background grid adopts unstructured tetrahedron.

[0035] Step 3: Each computing node reads the required spatial background grid cell information according to the read grid point data, generates the required topological information, and calls ParMETIS to perform distributed parallel grid partitioning in a graph partitioning manner. The number of grid partitions is equal to the number of flow field computing nodes.

[0036] Step 4: Based on the mesh partitioning results, MPI is used to transmit data between nodes. Each node obtains the actual mesh partitioning information and establishes the corresponding geometric and topological relationships.

[0037] Step 5: Each node calls METIS to partition the grid within the node in the form of graph partitioning. The number of grid partitions within the node is equal to the number of threads within the node.

[0038] Step 6: Each thread in the node uses OpenMP to establish the geometric relationship of each partition in parallel according to the mesh partition situation in the node, including calculating the volume, normal vector and other geometric information of the dualmesh, reordering the mesh, and performing geometric verification;

[0039] Step 7: Establishing data transmission relationship between nodes for MPI communication during flow field calculation;

[0040] Step 8: The solid calculation of the parafoil uses an independent calculation node different from the flow field calculation node. The parafoil uses finite element surface mesh discretization. Similar to the flow field partitioning step, ParMETIS is used for distributed parallel partitioning between solid calculation nodes, and METIS is used for serial partitioning inside the solid calculation node. All threads calculate local geometric information and establish the data transmission relationship required for fluid-solid coupling.

[0041] Step nine, in the process of fluid-solid coupling calculation, the immersed boundary method is used to solve the compressible Navier-Stokes equations, and the local one-dimensional half-Riemann problem is used to calculate the wall flux on the parafoil wall. MPI is used for data transmission between the computing nodes of the flow field, and OpenMP is used for coarse-grained parallelism among the threads within the nodes. MPI is also used for data transmission between the computing nodes of the solid, and OpenMP is used for coarse-grained parallelism within the nodes.

[0042] It should be noted that in this embodiment, MPI is used for data transmission between the flow field calculation nodes and the solid calculation nodes. Separate programs are used for flow field calculation and solid calculation respectively, and coupling is achieved by transferring data between different programs through MPI.

[0043] In addition, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium may store a program, and when the program is executed, the program includes part or all of the steps of a parafoil fluid-solid coupling parallel calculation method recorded in the above method embodiment.

[0044] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0045] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0046] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0047] An exemplary flow chart for calculating the parallelism of the parafoil fluid-solid coupling according to an embodiment of the present invention is described above with reference to the accompanying drawings. It should be noted that the large amount of details included in the above description are only exemplary illustrations of the present invention, rather than limitations of the present invention. In other embodiments of the present invention, the method may have more, fewer or different steps, and the order, inclusion, function, etc. of the steps may be different from those described and illustrated.

Claims

1. A parallel calculation method for parafoil fluid-solid coupling, characterized in that: The steps include: Step 1, determining the number of flow field calculation nodes, the number of threads inside the flow field calculation nodes, the number of solid calculation nodes, and the number of threads inside the solid calculation nodes according to requirements; Step 2: All flow field calculation nodes and solid calculation nodes read the flow field space background grid file simultaneously, and a process of each flow field calculation node and solid calculation node reads the initial grid information of each node according to the grid point sequence; Step 3: Each flow field calculation node and solid calculation node reads the relevant spatial background grid unit information according to the initial local grid points, generates the required topological information, and calls ParMETIS for distributed parallel grid partitioning; Step 4: According to the grid partitioning results, each flow field calculation node and solid calculation node uses MPI to transmit grid information and establish corresponding geometric and topological relationships; Step 5: Each flow field calculation node and solid calculation node calls METIS to perform serial grid partitioning within the node; Step 6: Each thread in the flow field calculation node and the solid calculation node uses OpenMP to establish the geometric relationship of each partition in parallel, including calculating the volume and normal vector geometric information of the dualmesh, reordering the mesh, and performing geometric verification; Step 7, respectively establishing data transmission relationships between flow field calculation nodes and solid calculation nodes, wherein the data transmission relationships are used for MPI communication during flow field calculation; Step 8, establishing the data transmission relationship required for fluid-solid coupling; Step nine, perform fluid-solid coupling calculation; wherein, in the process of fluid-solid coupling calculation, the immersed boundary method is used to solve the compressible Navier-Stokes equations, the local one-dimensional half-Riemann problem is used to calculate the wall flux of the parafoil wall, MPI is used for data transmission between flow field calculation nodes and between solid calculation nodes, and OpenMP is used for coarse-grained parallelism within the flow field calculation nodes and solid calculation nodes.

2. The parallel calculation method for parafoil fluid-structure coupling according to claim 1 is characterized in that: In the step 1, the number of flow field calculation nodes, the number of solid calculation nodes, and the number of threads inside the flow field calculation nodes and the solid calculation nodes are positive integers greater than or equal to 1.

3. The parallel calculation method for parafoil fluid-structure coupling according to claim 1 is characterized in that: In the step 2, each flow field calculation node and solid calculation node only reads its own initial grid information, and there is no need for the master node to process the global grid information.

4. The parallel calculation method for parafoil fluid-structure coupling according to claim 1 is characterized in that: The parallel grid partitioning in step 3 adopts the graph partitioning method of ParMETIS to perform distributed parallel grid partitioning, and ensures the internal continuity of each partition.

5. The parallel calculation method for parafoil fluid-structure coupling according to claim 1, characterized in that: In the step 5, each flow field calculation node and solid calculation node is grid partitioned. Each flow field calculation node and solid calculation node is serially grid partitioned using METIS graph partitioning according to the number of threads within the node, and continuity within each partition is ensured.

6. The parallel calculation method for parafoil fluid-structure coupling according to claim 1, characterized in that: The solid calculation in step eight uses an independent computing node different from the flow field calculation node. The parafoil uses finite element surface mesh discretization. ParMETIS is used for distributed parallel partitioning between the computing nodes. METIS is used for serial partitioning within the solid computing node. All threads calculate local geometric information.

7. The parallel calculation method for parafoil fluid-structure coupling according to claim 1, characterized in that: The MPI data transmission in step nine adopts non-blocking communication for overlapping communication and calculation.

8. The parallel calculation method for fluid-structure coupling of a parafoil according to any one of claims 1 to 7, characterized in that: The flow field computing nodes and the solid computing nodes use MPI for data transmission.

9. The parallel calculation method for parafoil fluid-structure coupling according to claim 8, characterized in that: The flow field and solid computing nodes use separate programs respectively, and coupling is achieved by transferring data between different programs through MPI.

10. A computer-readable storage medium, characterized in that: include: The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of a parallel calculation method for fluid-structure coupling of a parafoil according to any one of claims 1 to 9.

Images