A numerical simulation method for load out-of-cabin process based on dynamic nested grid motion

Abstract

The application discloses a load out-of-cabin process numerical simulation method based on dynamic nested grid motion, and self-adaptive control of grid node motion is realized through a UDF program for defining grid motion.The application has the following advantages: by adopting the dynamic nested grid motion control technology, the flow characteristics in the load out-of-cabin process, especially the flow field characteristics of the boundary layer, can be simulated more accurately; by self-adaptive control of the grid node motion, the isolated cell problem caused by the 'hole digging' treatment is effectively solved, the grid quality and the numerical solution accuracy are significantly improved, the numerical simulation of the load out-of-cabin process is ensured to maintain the grid quality, and the simulation result accuracy and reliability are improved, and the solution distortion problem caused by the isolated cell of the traditional nested grid technology at the boundary layer is solved.

Description

Technical Field

[0001] This invention relates to a numerical simulation method for the payload exit process based on dynamic nested grid motion, belonging to the field of aerial separation technology. Background Technology

[0002] Payload release systems are a type of in-flight multibody separation system, specifically including payload release from UAVs, small satellite release from spacecraft, and multi-stage separation from rockets. This process involves complex coupling of aerodynamics and kinematics, where dynamic changes in the relative positions and velocities of the multiple bodies can trigger various complex flow phenomena, such as shock wave interference, shock wave-boundary layer interaction, boundary layer separation, and vortex flow. These phenomena result in highly nonlinear and unsteady flow field parameters, significantly interfering with the aerodynamic and kinematic characteristics of the spacecraft and its separated components, making the separation process a critical and high-risk aspect of the flight mission. Although the separation process is short, it determines the flight attitude and velocity of the launch vehicle and payload after separation, significantly impacting the success of subsequent operations. To ensure the safety of the flight mission, a comprehensive and in-depth analysis of multibody separation and kinematic issues is crucial. By meticulously studying the flow field characteristics during the separation process, we can more accurately grasp the kinematic characteristics of the separated components and gain a deeper understanding of the mechanisms and influences of key factors.

[0003] For aerial multibody separation problems with moving boundaries, existing CFD numerical simulation methods are mainly based on dynamic mesh technology. Traditional methods for handling moving boundary problems generally employ DMS (Dynamic Modeling and Simulation), MDM (Moving / Deforming Mesh), and nested mesh techniques.

[0004] The DMS method performs steady-state aerodynamic calculations by selecting representative positions and attitudes during the separation process. The calculated discrete aerodynamic data is then fitted into a continuous aerodynamic data curve using correction coefficients and other methods, and input into the dynamic model for rigid body motion simulation. However, although the DMS method has low computational resource requirements, it does not consider the unsteady effects caused by multibody motion during the separation process. Furthermore, the aerodynamic database of the DMS method is fitted from discrete steady-state data, while the actual process of separation in the air is a continuous transient process. The computational accuracy of this method is insufficient for quantitative research on the entire separation process.

[0005] The MDM method requires frequent mesh adjustments to match changing geometries during boundary movement. This can lead to mesh quality degradation, such as mesh distortion or overstretching, and may also introduce negative volume elements, potentially causing numerical solution failure. Furthermore, the implementation of the MDM method is relatively complex, involving dynamic mesh updates, data transfer, and real-time adjustment of boundary conditions, all of which increase computational complexity and resource requirements. The dynamic changes in the mesh may also introduce additional numerical instabilities.

[0006] The concept of nested meshing is as follows: each part of the model generates its own mesh independently. The mesh can undergo rigid body motion with the part, and the meshes of different parts overlap, solving the problem by interpolation. Nested meshes offer strong independence, high quality, and fast generation speed, but the key to the nested meshing method is to automatically and efficiently establish the interpolation relationship between the nested meshes.

[0007] To address the motion boundary issues during payload exit, the complexity of the separation phase often necessitates detailed simulations of flow characteristics. Existing patent CN117436305A, a numerical simulation method for payload exit based on dynamic nested mesh technology, utilizes CFD-6DOF to address the shortcomings of the DMS method, which fails to consider the unsteady effects of multibody motion during separation. It also addresses the issue that the DMS method's aerodynamic database is fitted from discrete steady-state data, while the actual aerial separation process is a continuous transient process, making it difficult to achieve the required computational accuracy for quantitative research on the overall separation process. While offering greater flexibility and stability, this method also presents the following problems: During separation, due to the close proximity of component boundaries and the presence of a shear layer, the nested mesh undergoes "hole-cutting" (minimizing overlapping areas), resulting in interpolation relationships between components at the boundary layer. To accurately solve the flow field characteristics of the boundary layer, the cabin component mesh has a high resolution, leading to significant differences in element resolution between the two component meshes at the boundary layer. Consequently, the recipient component struggles to find suitable donor elements for interpolation at the boundary layer, resulting in isolated elements. Isolated elements in the boundary layer have a significant negative impact on flow field solutions, mainly manifested in reduced numerical stability, distortion of flow field characteristics, difficulty in correctly applying boundary conditions, and alterations in local flow characteristics. Since the boundary layer plays a decisive role in the fluid's viscous drag and heat exchange, the presence of isolated elements can lead to misjudgments of overall flow characteristics, especially when local effects are amplified. Furthermore, the limitations of numerical methods in handling boundary layers and mesh dependency issues further exacerbate the difficulty of solving the problem due to the presence of isolated elements. Summary of the Invention

[0008] This invention addresses the problem of solution distortion caused by isolated elements at the boundary layer in traditional nested mesh techniques used in fluent commercial software simulations of load exiting the chamber during engineering applications. It provides a numerical simulation method for load exiting the chamber based on dynamic nested mesh motion, which adaptively controls the motion of mesh nodes through the define_grid_motion UDF program, thus overcoming the shortcomings of the nested mesh method in solving multibody separation.

[0009] This invention is achieved through the following scheme: a numerical simulation method for the load exit process based on dynamic nested mesh motion, which adaptively controls the motion of mesh nodes by defining a UDF program for mesh motion.

[0010] It includes the following steps:

[0011] S1. Define a set of numbers before the simulation begins;

[0012] S2. After S1, follow the simulation time and determine whether the calculation time step is the first time step. If yes, proceed to S3; otherwise, proceed directly to S5.

[0013] S3. Traverse the wall cells of the moving block and record the cell resolution;

[0014] S4. After S3, traverse the nodes and update the array to form nested grid node coordinate twins;

[0015] S5. Update nodes based on basic motion and update twin array synchronously;

[0016] S6. After S5, calculate the projected distance from the node to the rear end face of the cabin in the direction of motion.

[0017] S7. After S6, calculate the relative coordinate difference between the twin and the static block boundary, and update the node based on the coordinate difference.

[0018] S1 defines an array used to store the coordinate information of nested grid nodes.

[0019] S1 defines a set of arrays, implemented by a UDF program, based on C++, and defines a variable array with fixed dimensions using basic syntax, the dimensions of which are determined by the number of unit nodes.

[0020] If the relative coordinate difference between the twin and the static block boundary of S7 is 0, then the twin coordinates are directly assigned to the node to complete the node update.

[0021] When the relative coordinate difference between the twin and the boundary of the static block in S7 is 0, the nested mesh node is located on the back end face of the static block.

[0022] If the absolute value of the relative coordinate difference between the twin and the static block boundary of S7 is greater than 0 and less than and equal to 1 / 2 unit resolution, then the updated node matches the static block boundary and the node update is completed.

[0023] If the relative coordinate difference between the twin and the static block boundary of S7 is greater than 1 / 2 and not greater than 1 unit resolution, or not less than -1 and less than -1 / 2 unit resolution, then the updated node matches the unit node of the inner layer of the static block boundary.

[0024] If the relative coordinate difference between the twin and the static block boundary of S7 is greater than 1 and not greater than 3 / 2 of the unit resolution, or not less than -3 / 2 and less than -1 of the unit resolution, then the updated node matches the unit node of the outer layer of the static block boundary.

[0025] If the absolute value of the relative coordinate difference between the twin and the static block boundary of S7 is greater than 3 / 2 of the unit resolution, then the twin coordinates are directly assigned to the node to complete the node update.

[0026] The beneficial effects of this invention are as follows:

[0027] 1. The present invention provides a numerical simulation method for load exit process based on dynamic nested grid motion. By adopting dynamic nested grid motion control technology, it can more accurately simulate the flow characteristics during load exit process, especially the flow field characteristics of the boundary layer.

[0028] 2. The present invention provides a numerical simulation method for the load exit process based on dynamic nested mesh motion. By adaptively controlling the movement of mesh nodes, this method effectively solves the problem of isolated elements caused by the "hole-cutting" process. This improvement significantly enhances the quality of the mesh and the accuracy of the numerical solution, ensuring that the numerical simulation of the load exit process maintains the quality of the mesh while also improving the accuracy and reliability of the simulation results. It also solves the problem of solution distortion caused by isolated elements at the boundary layer in traditional nested mesh technology. Attached Figure Description

[0029] Figure 1 This is a flowchart of the process steps of the present invention.

[0030] Figure 2 This is a schematic diagram of the two-dimensional grid division of the carrier in an embodiment of the present invention.

[0031] Figure 3 This is a schematic diagram of the two-dimensional grid division of the payload compartment in an embodiment of the present invention.

[0032] Figure 4 This is a flow field pressure cloud map at 0.1s after the load separation from the chamber in this embodiment of the invention.

[0033] Figure 5 This is a flow field pressure cloud map at 0.3s after the load separation from the chamber in this embodiment of the invention.

[0034] Figure 6 This is a flow field pressure cloud map at 0.5s after the load separation from the chamber in this embodiment of the invention. Detailed Implementation

[0035] The following is combined Figure 1-6 The present invention will be further described, but the scope of protection of the present invention is not limited to the contents described herein.

[0036] For clarity, not all features of the actual embodiments will be described. In the following description, well-known functions and structures will not be described in detail, as they would confuse the invention with unnecessary details. It should be understood that in the development of any actual embodiment, a great deal of implementation detail must be made to achieve the developer’s specific goals, such as changing one embodiment to another according to the limitations of the system or business. In addition, it should be understood that such development work may be complex and time-consuming, but is merely routine work for those skilled in the art.

[0037] This invention addresses the solution distortion problem caused by isolated elements at the boundary layer in traditional nested mesh techniques used in Fluent commercial software simulations of load ejection processes in engineering applications. It proposes a numerical simulation method for the load ejection process based on dynamic nested mesh motion control technology. This method adaptively controls the motion of mesh nodes by defining a UDF program for mesh motion, optimizing the CFD-6DOF coupled solution for the load ejection process. Figure 1 , Figure 2 and Figure 3 As shown, the method of the present invention includes a background mesh, a hull nested mesh, and a UDF program, and includes the following steps:

[0038] Step 1: Create two-dimensional geometric models of the launch vehicle and the payload compartment using DesignModeler. The launch vehicle is 8m long and 2m wide, and the payload compartment is 1.5m long and 1m wide.

[0039] Step 2: The two-dimensional model is meshed using the nested mesh method. Based on the motion of the separation system, the separation system is divided into two mesh layers. The mesh containing the carrier is the background mesh, and the mesh containing the payload compartment is the nested mesh. Since the background mesh has a lower priority in the computational model, it is responsible for providing the receiver unit to receive the data transmitted from the motion mesh interpolation. The center area of ​​the mesh is the motion trajectory area of ​​the separation system, and local densification is performed to ensure the accuracy of the model. The mesh size smoothly transitions from the wall to the far field and gradually becomes sparser.

[0040] Step 3: Based on the three-dimensional compressible unsteady Navier-Stokes equations and the SST K-Omega turbulence model, establish a physical model of airflow during the separation process; this includes the following steps:

[0041] a: Establish a three-dimensional compressible unsteady Navier-Stokes equation model, whose governing equations are in integral form as follows:

[0042]

[0043] Where W is a conserved variable, F, F v Let Ω be the flow vector term and the dissipation flow vector term, respectively; dS be the area of ​​the infinitesimal element; Re be the Reynolds number; and n be the normal vector outside the control body.

[0044] b: The flow control equations are discretized using the finite volume method. The computational grid is spatially discretized using a second-order upwind scheme, and in the time direction, it is discretized using a k-order backward differential scheme.

[0045] Step 4: Import the mesh files for the pressure far field and the load chamber into Fluent, and perform steady-state calculations by making the following settings in Fluent:

[0046]

[0047] Step 5: After 800 calculations, determine whether the entire flow field has converged based on whether the residuals and aerodynamic coefficients have converged.

[0048] Step Six: After the steady-state calculations are completed, unsteady-state calculations need to be performed based on the stable flow field from the steady-state calculations. Simultaneously, the prepared UDF file is imported for unsteady state settings. Alternatively, the required motion patterns of the moving bodies can be defined by adding code to the UDF program. In this embodiment, the payload compartment is defined to separate at a speed of 0.1 m / s relative to the launch vehicle at the start of the unsteady calculations. Figure 1 As shown, the method for writing a UDF file is as follows:

[0049] S1. Before the simulation begins, define a set of numbers to store the coordinate information of the nested mesh nodes;

[0050] S2. After S1, follow the simulation time and determine whether the calculation time step is the first time step. If yes, proceed to S3; otherwise, proceed directly to S5.

[0051] S3. Traverse the wall cells of the moving block and record the cell resolution;

[0052] S4. After S3, traverse the nodes and update the array to form nested grid node coordinate twins;

[0053] S5. Update nodes based on basic motion and update twin array synchronously;

[0054] S6. After S5, calculate the projected distance from the node to the rear end face of the cabin in the direction of motion.

[0055] S7. After S6, calculate the relative coordinate difference between the twin and the static block boundary, and update the node based on the coordinate difference.

[0056] If the relative coordinate difference is 0, the twin coordinates are directly assigned to the node to complete the node update.

[0057] When the relative coordinate difference is 0, the nested grid node is located on the back face of the static block.

[0058] If the absolute value of the relative coordinate difference is greater than 0 and less than or equal to 1 / 2 unit resolution, then the updated node matches the boundary of the quiet block and the node update is completed.

[0059] If the relative coordinate difference is greater than 1 / 2 and not greater than 1 unit resolution, or not less than -1 and less than -1 / 2 unit resolution, then the updated node matches the unit node one layer inside the boundary of the quiet block.

[0060] If the relative coordinate difference is greater than 1 and not greater than 3 / 2 of the unit resolution, or not less than -3 / 2 and less than -1 of the unit resolution, then the updated node matches the unit node one layer outside the boundary of the quiet block.

[0061] If the absolute value of the relative coordinate difference is greater than 3 / 2 unit resolution, the twin coordinates are directly assigned to the node to complete the node update.

[0062] Step 7: Set the calculation time step to 0.01 and the calculation time step to 60 steps.

[0063] Step 8: After the calculation is completed, check the flow field pressure change characteristics during the exit process.

[0064] Figure 4 , 5 Figures 6 and 7 show the flow field pressure contour maps at 0.1s, 0.3s, and 0.5s after separation. In the early stages of separation, the launch vehicle and payload compartment are under the same pressure field. Due to air compression at the launch vehicle's head, a high-pressure zone forms, while the air density decreases on both sides of the launch vehicle, causing a pressure drop and forming a shock wave. As time progresses, the shock wave rapidly spreads, extending the low-pressure zone and affecting the exit process.

[0065] The results show that the mesh control technology based on the UDF program can effectively simulate the load exit process; during the simulation, no isolated elements are generated in the boundary layer, and the results are in good agreement with the actual situation.

[0066] Although the technical solutions of the present invention have been described and enumerated in detail, it should be understood that modifications to the above embodiments or the adoption of equivalent alternatives are obvious to those skilled in the art. Such modifications or improvements made without departing from the spirit of the present invention are all within the scope of protection claimed by the present invention.

Claims

1. A method for numerical simulation of load offloading process based on dynamic nested mesh motion, characterized in that, The UDF program for defining the grid motion is used for self-adaptive control of the grid node motion, including the following steps: S1, defining a set of numbers before the simulation starts; S2, after S1, following the simulation time advancement, judging whether the calculation time step is the first time step, if yes, entering S3, if not, directly entering S5; S3, traversing the moving block wall surface unit and recording the unit resolution; S4, after S3, traversing the node, updating the array to form the nested grid node coordinate twin; S5, updating the node according to the basic motion and synchronously updating the twin array; S6, after S5, calculating the projection distance of the node to the rear end surface of the cabin in the motion direction; S7, after S6, calculating the relative coordinate difference between the twin and the static block boundary, and completing the node update according to the coordinate difference.

2. The method of claim 1, wherein, The S1 defines a set of arrays for storing the nested grid node coordinate information.

3. The method of claim 1, wherein, The S1 defines a set of arrays, which is realized by the UDF program, the base is c++, and the dimension-fixed variable array is defined by the basic syntax, and the dimension is determined according to the number of unit nodes.

4. The method of claim 1, wherein, When the relative coordinate difference between the twin and the static block boundary in S7 is 0, the twin coordinate is directly assigned to the node to complete the node update.

5. The method of claim 1, wherein, When the relative coordinate difference between the twin and the static block boundary in S7 is 0, the nested grid node is on the rear end surface of the static block.

6. The method of claim 1, wherein, When the absolute value of the relative coordinate difference between the twin and the static block boundary in S7 is greater than 0 and less than or equal to 1 / 2 unit resolution, the node is updated to match the static block boundary to complete the node update.

7. The method of claim 1, wherein, When the relative coordinate difference between the twin and the static block boundary in S7 is greater than 1 / 2 and not greater than 1 unit resolution or not less than -1 and less than -1 / 2 unit resolution, the node is updated to match the inside one layer of unit nodes of the static block boundary.

8. The method of claim 1, wherein, When the relative coordinate difference between the twin and the static block boundary in S7 is greater than 1 and not greater than 3 / 2 unit resolution or not less than -3 / 2 and less than -1 unit resolution, the node is updated to match the outside one layer of unit nodes of the static block boundary.

9. The method of claim 1, wherein, When the absolute value of the relative coordinate difference between the twin and the static block boundary in S7 is greater than 3 / 2 unit resolution, the twin coordinate is directly assigned to the node to complete the node update.

Images