A multi-body separation impact simulation method based on master-slave surface detection and impulse response

Abstract

The application discloses a multi-body separation collision simulation method based on master-slave surface detection and impulse response, relates to the field of air separation, and comprises the following steps: reading all sub-grid data; preprocessing the sub-grid data; assembling a dynamic grid system for the preprocessed data, iteratively solving a flow field control equation of a current physical time step, and obtaining a flow field result; performing integral solving based on the flow field result, and obtaining motion parameters acting on each object; calculating the position and attitude parameters of the object at a next physical time step by solving a rigid body six-degree-of-freedom motion equation or according to a preset motion rule; performing collision detection and response, updating the motion parameters when geometric intersection occurs between the objects, and otherwise directly outputting the motion parameters; obtaining a current calculation result according to the output motion parameters; and ending the simulation process when the calculation result meets the requirements. The application can realize continuous simulation of the interaction between the motion state and the fluid after the collision event occurs.

Description

Technical Field

[0001] This application relates to the field of aerial separation technology, and in particular to a multi-body separation collision simulation method based on master-slave surface detection and impulse response. Background Technology

[0002] Currently, in the study of multibody separation problems, flight testing, wind tunnel testing, and numerical simulation all have limitations in handling separation scenarios involving collisions with accompanying objects. Flight testing is costly and difficult to simulate collision conditions while ensuring safety; data acquisition is challenging, and introducing collision testing introduces significant risks. While wind tunnel testing is relatively controllable, it is also costly and time-consuming, and suffers from problems such as support interference, incomplete measurements, aerodynamic unsteadiness, and limited range of motion. The uncertainty of the collision process further affects data accuracy, easily leading to distorted results or even failure. Numerical simulation, as an effective supplement, can significantly reduce costs, shorten the cycle, and improve design flexibility. The coupled solution of fluid control equations and rigid body motion equations based on the moving mesh method can achieve numerical prediction of the motion safety of the multibody separation process. However, in multibody separation simulations, once a collision occurs, the calculation is forcibly interrupted, making it impossible to continue simulating the post-collision motion and fluid interaction, and thus determine that the separation condition has safety hazards.

[0003] When the relative velocity of the objects to be collided is too high, the collision will significantly affect the projectile's flight trajectory, attitude stability, and even structural integrity, easily leading to serious consequences such as structural damage and attitude instability. Such situations must be assessed as safety accidents. However, in some practical engineering scenarios, if the distance between the two objects is too close and the relative velocity is small at the time of collision, the impact of the collision on the projectile's attitude angle is limited and usually does not endanger system safety. Such minor collisions are acceptable or even unavoidable in engineering practice. For example, during the separation of submunitions from a cluster munition, there are significant aerodynamic interference effects between the submunitions and between the submunitions and the parent munition, resulting in a flow field exhibiting strong unsteady and nonlinear characteristics. This unsteady aerodynamic force directly affects the spatial attitude evolution of the submunitions, making collisions between projectiles difficult to avoid. In addition, the design of modern airdrop systems often involves the cluster scattering of a large number of unguided missiles. During this process, the strong aerodynamic interference between the mother aircraft and the projectiles will significantly change their trajectories, making collisions possible.

[0004] For multi-body separation problems with moving boundaries in the air, existing numerical simulation methods mainly employ dynamic meshing techniques and a computational fluid dynamics-coupled rigid body six-degree-of-freedom approach. However, during fluid simulation, when objects experience contact penetration through the membrane, floating-point overflow errors occur, forcibly halting the calculation. Summary of the Invention

[0005] The purpose of this application is to provide a multibody separation collision simulation method based on master-slave surface detection and impulse response, which can realize continuous simulation of motion state and fluid interaction after a collision event, and avoid the inability to fully evaluate the separation condition due to the termination of simulation caused by the collision.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] This application provides a multi-body separation collision simulation method based on master-slave surface detection and impulse response, which includes the following steps.

[0008] Read all sub-mesh data; the sub-mesh data includes the mesh cells and nodes of several objects to be collided with.

[0009] The subgrid data is preprocessed to obtain preprocessed data.

[0010] The preprocessed data is then assembled using a dynamic mesh system to obtain the assembled mesh.

[0011] Based on the assembled mesh, the flow field control equations for the current physical time step are solved iteratively to obtain the flow field results.

[0012] Based on the flow field results, the motion parameters acting on each object are obtained by integral solution; the motion parameters include aerodynamic force and torque parameters.

[0013] Based on the motion parameters, the position and attitude parameters of the object at the next physical time step are calculated by solving the rigid body six-degree-of-freedom motion equations or by following a preset motion law.

[0014] Based on the position and attitude parameters of the object in the next physical time step, collision detection and response are performed to determine whether geometric intersection occurs between the objects. If so, the motion parameters are updated, and the system returns "Based on the motion parameters, the position and attitude parameters of the object in the next physical time step are calculated by solving the rigid body six-degree-of-freedom motion equation or according to the preset motion law"; otherwise, the motion parameters are directly output.

[0015] Based on the output motion parameters, the current calculation results are obtained; the current calculation results include: repositioning the spatial positions of each object and adjusting the spatial coordinates and topological relationships of the mesh accordingly.

[0016] Determine whether the current calculation result meets the set convergence or termination conditions. If not, return "Perform dynamic mesh system assembly on the preprocessed data to obtain the assembled mesh"; if yes, end the simulation process.

[0017] Based on the specific embodiments provided in this application, the following technical effects are disclosed.

[0018] This application provides a multi-body separation and collision simulation method based on master-slave surface detection and impulse response. The method includes: reading all sub-mesh data; the sub-mesh data includes mesh elements and nodes of several objects to be collided; it can comprehensively acquire the basic geometric data required for simulation, providing complete data support for subsequent mesh processing, flow field solution, and collision simulation. The sub-mesh data is preprocessed to obtain preprocessed data; this standardizes the data format, removes anomalies and redundant information, and improves the computational efficiency and stability of subsequent mesh assembly and numerical solution. The preprocessed data is then assembled using a dynamic mesh system to obtain an assembled mesh; this can construct a dynamic mesh structure adapted to multi-body motion, ensuring the topological integrity and spatial adaptability of the mesh under multi-body relative motion scenarios. Based on the assembled mesh, the flow field control equations for the current physical time step are iteratively solved to obtain the flow field results; this accurately acquires the flow field distribution results at the corresponding time, providing a reliable flow field basis for calculating the aerodynamic forces and moments of the objects. Based on the flow field results, integral solutions are performed to obtain the motion parameters acting on each object. These motion parameters include aerodynamic and torque parameters. This enables the conversion of flow field physical quantities into rigid body mechanical parameters, establishing a coupling relationship between the flow field and rigid body motion. Based on these motion parameters, the position and attitude parameters of the object at the next physical time step are calculated by solving the rigid body's six-degree-of-freedom motion equations or according to a preset motion law. This allows for the scientific prediction of the object's spatial state at the next moment, providing core motion state data for subsequent collision detection. Based on the position and attitude parameters of the object at the next physical time step, collision detection and response are performed to determine whether geometric intersection occurs between objects. If so, the motion parameters are updated, and the system returns "Based on the motion parameters, the position and attitude parameters of the object at the next physical time step are calculated by solving the rigid body's six-degree-of-freedom motion equations or according to a preset motion law." If not, the motion parameters are directly output. This system can promptly identify multi-body collision behavior and correct the motion state through reasonable responses, ensuring the realism of the collision simulation and the correctness of the motion logic. Based on the output motion parameters, the current calculation result is obtained. This result includes: repositioning the spatial positions of each object and adjusting the spatial coordinates and topological relationships of the mesh accordingly. It enables coordinated updating of the spatial positions of multiple bodies and the moving mesh, adapting to the spatial morphology after multi-body motion. The system determines whether the current calculation result meets the set convergence or termination conditions. If not, it returns "Assemble the preprocessed data using the moving mesh system to obtain the assembled mesh"; if yes, it terminates the simulation process. This effectively controls the calculation accuracy and execution progress of the simulation, ensuring that the simulation results meet the standards and the process is terminated appropriately.This method, through a step-by-step systematic design, sequentially completes data acquisition, preprocessing, mesh construction, flow field solution, motion calculation, collision response, mesh update, and simulation convergence determination. It achieves an organic combination of flow field coupling, rigid body motion, and collision simulation, ensuring the geometric accuracy, flow field solution accuracy, and collision simulation realism of multi-body separation and collision simulation. It also improves the computational stability, execution efficiency, and logical rigor of the entire simulation process, enabling it to efficiently and reliably complete the dynamic simulation task of multi-body separation and collision. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a multi-body separation collision simulation method based on master-slave surface detection and impulse response, provided as an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of the overall process of a multi-body separation and collision simulation method based on master-slave surface detection and impulse response, provided as an embodiment of this application.

[0022] Figure 3 This is a schematic diagram of the collision impulse response between objects A and B according to an embodiment of this application.

[0023] Figure 4 This is a schematic diagram of a two-sphere collision model provided in an embodiment of this application.

[0024] Figure 5 This is a schematic diagram illustrating the effect of grid scale on calculation results in an embodiment of this application.

[0025] Figure 6 This is a schematic diagram illustrating the effect of time step on calculation results in an embodiment of this application.

[0026] Figure 7 This is a schematic diagram of a single sphere free fall bouncing model provided in an embodiment of this application.

[0027] Figure 8 This is a schematic diagram of the Y-direction displacement curve of the center of mass of a free-falling sphere provided in an embodiment of this application.

[0028] Figure 9 This is a schematic diagram of the trajectory error of a sphere during multiple collisions, provided as an embodiment of this application. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] To address the difficulty of handling multi-body collision effects in existing fluid numerical simulation software, this application proposes a multi-body separation and collision simulation method based on master-slave surface detection and impulse response. This method effectively enables continuous simulation of the motion state and fluid interaction after a collision event, avoiding the inability to fully evaluate the separation condition due to simulation termination caused by the collision.

[0031] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] In one exemplary embodiment, such as Figure 1 As shown, a multi-body separation collision simulation method based on master-slave surface detection and impulse response is provided. The method includes the following steps.

[0033] S1: Read all sub-mesh data; the sub-mesh data includes the mesh cells and nodes of several objects to be collided with.

[0034] S2: Preprocess the subgrid data to obtain preprocessed data.

[0035] S3: Perform dynamic mesh system assembly on the preprocessed data to obtain the assembled mesh.

[0036] S4: Based on the assembled mesh, the flow field control equations for the current physical time step are solved iteratively to obtain the flow field results.

[0037] S5: Based on the flow field results, perform integral solutions to obtain the motion parameters acting on each object; the motion parameters include: aerodynamic force and torque parameters.

[0038] S6: Based on the motion parameters, the position and attitude parameters of the object in the next physical time step are calculated by solving the rigid body six-degree-of-freedom motion equation or according to the preset motion law.

[0039] S7: Based on the position and attitude parameters of the object in the next physical time step, perform collision detection and response, determine whether geometric intersection occurs between objects, if yes, update the motion parameters, and return "Based on the motion parameters, calculate the position and attitude parameters of the object in the next physical time step by solving the rigid body six-degree-of-freedom motion equation or according to the preset motion law"; if no, directly output the motion parameters.

[0040] S8: Based on the output motion parameters, obtain the current calculation results; the current calculation results include: repositioning the spatial positions of each object and adjusting the spatial coordinates and topological relationships of the mesh accordingly.

[0041] S9: Determine whether the current calculation result meets the set convergence or termination conditions. If not, return "Assemble the preprocessed data using a dynamic mesh system to obtain the assembled mesh"; if yes, end the simulation process.

[0042] By implementing steps S1 to S9 above, this application can solve the problem of forced interruption of calculation when a collision occurs during the multibody separation process in existing numerical simulation methods, realize continuous simulation of motion state and fluid interaction after the collision event, and avoid the inability to fully evaluate the separation condition due to the termination of simulation caused by the collision.

[0043] As an optional implementation, in step S7, the execution of the collision detection and response specifically includes the following steps.

[0044] S71: For each slave node, traverse all master units sequentially to determine the final target master unit; the slave node is the node corresponding to the surface area on the subordinate object that participates in the contact; the master unit is the surface area on the master object that participates in the contact.

[0045] S72: Based on the spatial relationship between the target master surface unit and the master-slave surface, and combined with the engineering numerical tolerance setting, a preliminary penetration determination is made to obtain a preliminary determination result; the preliminary determination result includes: no penetration and potential penetration points.

[0046] S73: When the preliminary determination result is a potential penetration point, determine whether the projection point is located in the content of the final target unit; if not, determine that there is no penetration; if yes, determine that it is a valid contact point.

[0047] S74: Extract contact information based on the effective contact point; the contact information includes: contact point, penetration depth, and contact normal vector.

[0048] S75: Based on the contact information, calculate the collision response and perform secondary screening on the effective contact points to obtain the screened effective contact points; wherein, the secondary screening condition is: retain contact points that satisfy the condition that the penetration depth is greater than or equal to the minimum penetration depth and the relative normal component at the contact point is less than or equal to 0.

[0049] S76: Determine whether the filtered effective contact points are multi-point contacts; if so, construct an equivalent single-point contact model using penetration depth weighted average to obtain equivalent contact points and equivalent contact normal vectors, and calculate the normal impulse based on the equivalent contact points and the equivalent contact normal vectors; if not, calculate the normal impulse based on the filtered effective contact points and the contact point normal vectors.

[0050] S77: Update the motion state of the two objects based on the normal impulse; the motion state includes linear velocity and angular velocity.

[0051] As an optional implementation, in step S71, for each slave node, all master face units are traversed sequentially to determine the final target master face unit, which specifically includes the following steps.

[0052] S711: Calculate the spatial distance from the face node to the center of each main face unit, and perform preliminary screening based on a pre-set distance threshold to obtain a set of candidate units after screening.

[0053] S712: Calculate the signed distance from the face node to the plane containing each candidate cell.

[0054] S713: The element with the smallest absolute value of signed distance is taken as the final target principal surface element.

[0055] As an optional implementation, in step S72, based on the spatial positional relationship between the target master surface unit and the master-slave surface, and combined with the engineering numerical tolerance setting, a preliminary penetration determination is made to obtain a preliminary determination result, which specifically includes the following steps.

[0056] S721: Set penetration tolerance.

[0057] S722: Determine whether the signed distance corresponding to the target unit is less than the penetration tolerance; if not, determine that there is no penetration; if yes, determine that it is a potential penetration point.

[0058] As an optional implementation, in step S73, when the preliminary determination result is a potential penetration point, it is determined whether the projection point is located within the final target unit content; if not, it is determined that there is no penetration; if so, it is determined that it is a valid contact point, specifically including the following steps.

[0059] S731: Calculate the orthogonal projection points of the face nodes on the target main face element plane.

[0060] S732: Calculate the centroid coordinates of the orthogonal projection points.

[0061] S733: Determine whether the centroid coordinates of the orthogonal projection point are greater than the numerical tolerance; if not, determine that there is no penetration; if yes, determine that it is a valid contact point.

[0062] As an optional implementation, in step S74, contact information is extracted based on the effective contact point, which specifically includes the following steps.

[0063] S741: Use the orthogonal projection point as the contact point.

[0064] S742: Use the absolute value of the signed distance corresponding to the target principal element as the penetration depth.

[0065] S743: Use the unit outward normal vector of the target master face element as the contact normal vector.

[0066] In summary, the specific process of the technical solution of this application is as follows.

[0067] (1) Read the data of all subgrids.

[0068] (2) Preprocess the grid cells and nodes.

[0069] (3) Complete the assembly of the dynamic mesh system.

[0070] (4) Iteratively solve the first n The flow field control equations are calculated for each physical time step until convergence and the flow field results are output.

[0071] (5) Solve for the aerodynamic forces and torque parameters acting on each object by integrating the flow field results and output them.

[0072] (6) By solving the equations of motion of the rigid body with six degrees of freedom or according to the preset motion law, calculate and obtain the first... n +1 physical time step: the position and attitude parameters of the object.

[0073] (7) Perform collision detection and response: If it is determined that there is a geometric intersection between the objects, the collision response processing module is called to update the motion parameters and return to step (6); otherwise, the motion parameters are output directly.

[0074] (8) Based on the updated motion parameters, reposition the spatial position of each object and adjust the spatial coordinates and topological relationships of its mesh accordingly.

[0075] (9) Analyze the current calculation results. If the set convergence or termination conditions are met, end the simulation process; otherwise, return to step (3) to continue the calculation of the next time step.

[0076] The technical solution for collision detection and response in process (7) in this embodiment is as follows.

[0077] In multi-body collision numerical simulations, collision detection, solving for collision geometric features, and collision response constitute the three core components. This paper defines the two objects in contact as the controlling object and the subordinate object. The surface region on the controlling object involved in the contact is called the master surface, the outer surface of the element constituting the master surface is called the master piece, and the corresponding node is called the master point. Similarly, the surface region on the subordinate object involved in the contact is defined as the slave surface, the outer surface of the element is called the slave piece, and the corresponding node is called the slave point. Slave points located within the geometry of the controlling object and master points located within the geometry of the subordinate object are collectively classified as internal points.

[0078] In computational fluid dynamics, triangular or quadrilateral elements are typically used to discretize the surface of an object, and these surface meshes are topologically closed. The key to collision detection algorithms lies in determining the spatial relative positions of different mesh elements based on the topological properties of the surface mesh. This embodiment establishes a master-slave surface penetration detection search algorithm based on the parallel partitioned storage mechanism of the CFD solver and the topological relationships of the object surface.

[0079] The core task of penetration detection is to determine whether a slave node illegally intrudes into the impenetrable region defined by the master surface. By combining the geometric and topological features of the master and slave surfaces, spatial distance determination, and projection validity verification, the accuracy and engineering applicability of the detection results are ensured. Let the master surface be composed of... N m It is composed of triangular units (a quadrilateral unit is decomposed into two triangles), and contains from the surface N s Each node, for each face node P s ,like Figure 2 As shown, perform the following steps.

[0080] A1. Iterate through all master face elements in sequence, calculate the Euclidean distance between the slave face node and the center of each master face element, and perform preliminary screening based on a pre-set distance threshold (related to the local mesh scale); then, for the selected candidate element set, further calculate the signed distance from the slave face node to the plane where each element is located, and select the element with the smallest absolute value of the signed distance as the final target master face element. The specific steps are as follows.

[0081] A1.1 Calculate the nodes on the face. P s The spatial distance dcenter to the center point of each candidate principal face unit is initially selected if the distance is less than a threshold. d th (Take a cell that is 5 times the average size of the local mesh in the near flow field).

[0082] A1.2 For the filtered cells, calculate the number of face nodes. P s The signed distance to the plane containing the element is given by the formula: ,in, d For signed distance; The unit outward normal vector of the candidate principal face element; P s For surface nodes; V m,0 Let be any vertex of the target principal face element.

[0083] A1.3. The element with the smallest absolute value of the signed distance is selected as the target principal surface element (the target element in the figure). T m This ensures that subsequent testing is targeted.

[0084] A2. Based on the spatial relationship between the master and slave planes and combined with the engineering numerical tolerance setting, a preliminary penetration judgment is made. The specific steps are as follows.

[0085] A2.1 Setting Penetration Tolerance The value needs to be adapted to the mesh scale and simulation accuracy requirements. In this embodiment, it is set to 1% of the local mesh scale (i.e., , h local (The average side length of the target principal face unit).

[0086] A2.2, if satisfied This indicates the origin of the face node. P s Located on the negative side of the unit outward normal vector of the principal face, it is initially identified as a potential penetration point; if If the node fails to penetrate, it is considered to have no penetration, and subsequent detection of that node is terminated directly.

[0087] A3. After initially determining it to be a potential penetration, verify whether the projection point is located inside the target main surface unit to avoid non-effective contact scenarios such as "skimming over" or "approaching from the outside".

[0088] A3.1 Calculation P s exist T m Orthogonal projection points on the plane: . is the unit outward normal vector of the candidate principal face element.

[0089] A3.2 If the centroid coordinate method is used for judgment Is it located in a triangular target unit? T m The specific steps are as follows.

[0090] A3.2.1, Let... T m The three vertices are V 0, V 1, V 2. Calculate the edge vectors and relative position vector .

[0091] A3.2.2 Solving for the coordinates of the centroid ,satisfy: .

[0092] A3.2.3, Introducing Numerical Tolerance Depending on the situation, if all barycenter coordinates satisfy... If the projection point is valid, the contact is considered valid; otherwise, it is considered invalid contact, and the penetration determination of that node is excluded.

[0093] A4. Extract contact geometry and kinematic information to provide basic data for subsequent collision response calculations. The extracted content is shown below.

[0094] A4.1 Further, extract the contact point location, using the projection point... As the contact point, i.e. P c = P proj If the projection point is located at the element boundary, the contact point position can be corrected by weighted averaging of element vertices, thereby improving the accuracy of subsequent impulse calculation.

[0095] A4.2 Further, calculate the penetration depth, taking the absolute value of the signed distance; a positive value represents the amount of intrusion. The formula is: .

[0096] A4.3 Furthermore, define the unit normal vector, using the unit outward normal vector of the target principal face element, i.e. n c = n m The direction is from the main surface to the secondary surface, ensuring consistency with the force direction definition of the collision response.

[0097] A5. Based on the solution results of the rigid body's six-degree-of-freedom motion equations, extract the relative motion information at the contact point.

[0098] Based on the object's center of mass velocity and angular velocity, the contact point of the objects belonging to the master and slave surfaces is calculated using rigid body motion relationships. P c , P s The velocity at that point is given by the formula: ,in, Vc,m The velocity of the center of mass of the object to which the master and slave planes belong. Angular velocity, Let be the radius vector from the center of mass to the contact point; the relative velocity and normal component at the contact point are: ;in, , These are the nodal velocities at the contact points between the objects belonging to the face and the objects belonging to the main face, respectively. The cross product of the relative velocity and the normal vector of the contact surface is the relative velocity on the normal vector.

[0099] A6. In this embodiment, due to the extremely short duration of the collision and the lack of significant change in the geometric configuration before and after the collision, the effects of gravity, aerodynamic forces, and other interactions are not considered in the collision solution. Based on this premise, the effect of the collision on the object's motion state can be described by the effect of impulse on its linear and angular velocities, such as... Figure 3 As shown. Based on impulse-momentum theory and combined with the dynamic characteristics of multi-body separation collision, the collision response is calculated, and the following steps are performed to complete the collision response.

[0100] A6.1 To avoid interference from numerical noise, non-physical contact, or separation trends, all detected penetration points undergo secondary screening. When the penetration depth... At that time, among them , L ref Based on the main surface feature length, contact points are eliminated to avoid erroneous responses caused by minute numerical errors; when This means that the two objects are moving away from each other at the point of contact, and no collision impulse is required; the condition is retained. and The contact points ensure that responses are only made to valid intrusion attempts.

[0101] A6.2 When a face node corresponds to multiple effective master face elements, or a single object has multiple penetration points, an equivalent single-point contact model is constructed using a penetration depth-weighted average. The equivalent contact point is... ,in, For the first i Penetration depth of each effective contact point P c,i For the first i The position coordinates of the effective contact points; the equivalent contact normal vector is: , n c,i For the first i The normal vector of each effective contact point.

[0102] A6.3. Combining the above formula, we can obtain the impulse on the colliding objects at the moment of impact. Since the collision time is too short, there is basically no energy dissipation, so we assume that the collision is perfectly elastic. ,in,J Normal impulse; m A , m B These are the masses of the object belonging to the primary face and the object belonging to the main face, respectively. I A , I B It is the inertial tensor of the object belonging to the face and the object belonging to the main face in the inertial frame (obtained by transforming the body axis inertial tensor through the coordinate transformation matrix); The relative velocity in the normal direction before the collision (the effective contact point must satisfy) ); r A =P eq -C A , r B =P eq -C B These are the radius vectors pointing from the centroid of the object belonging to the main surface to the equivalent point of contact, respectively. C A Let the centroid of the object to which the surface belongs be denoted. C B The direction in which the center of mass of the object to which the main face belongs points.

[0103] A6.4 Furthermore, based on the calculated normal impulse... J Update the velocity of the center of mass of the two objects (the linear velocity in the diagram): Angular velocity update: .

[0104] It should be noted that there are some alternative solutions for implementing the method of this application.

[0105] (1) Alternative to detection algorithm - bounding box algorithm: The bounding box method is an acceleration technique widely used in collision detection, ray tracing, spatial query and other fields. Its core idea is to use a simple and easy-to-compute "bounding body" to approximate a complex geometric object. By first performing a fast intersection test on the bounding body, a large number of object pairs that are unlikely to interact can be efficiently eliminated, thereby significantly reducing the overhead of subsequent accurate geometric calculations.

[0106] (2) The alternative to the contact force model - the penalty function method: The penalty function method is a numerical approximation method used to deal with contact, collision or constraint problems. Its core idea is to transform the originally impenetrable hard constraint into a "soft" elastic force to simulate contact behavior. When two objects penetrate each other, the system will apply a repulsive force proportional to the penetration depth, thereby preventing further intrusion and pushing the objects to separate.

[0107] The key to this application lies in: by secondary development of a UDF program based on traditional commercial software, and by combining master-slave surface detection and impulse response to perform effective collision simulation, the collision problem that traditional multi-body separation numerical simulation cannot solve is solved, thereby improving the simulation accuracy.

[0108] The beneficial effects of the technical solution in this application are as follows.

[0109] This application provides a multi-body separation collision simulation method based on master-slave surface detection and impulse response, which offers an effective and complete solution to the pain point of forced interruption when dealing with collisions in existing numerical simulation techniques.

[0110] Existing technologies for handling multi-body separation in mid-air are interrupted by numerical calculation errors upon mesh penetration or contact, resulting in the inability to obtain the post-collision trajectory and aerodynamic characteristics of the objects, leading to an incomplete simulation process. This application introduces a "master-slave surface penetration detection-collision response calculation" module. Upon collision detection, the flow field solution is not interrupted; instead, the velocity and angular velocity of the objects after the collision are calculated in real-time based on impulse theory. This data is then used to update the rigid body motion state, allowing subsequent fluid-structure interaction calculations to continue. This enables the numerical simulation to realistically and continuously simulate the entire process from separation initiation, flight, collision, to continued motion after the collision. It provides complete dynamic and aerodynamic data for evaluating the safety and stability of separation schemes, greatly enhancing the engineering application value of numerical simulation.

[0111] An efficient and robust geometric detection and equivalent processing strategy is adopted, balancing computational accuracy and efficiency. By defining principal and secondary faces and employing screening strategies based on spatial and signed distances, the computational workload for unnecessary contact pair searches is significantly reduced. The method of determining the validity of projection points using centroid coordinates has clear geometric meaning, high numerical stability, and can accurately eliminate invalid contacts such as "edge grazing," ensuring the reliability of the detection results.

[0112] When complex contacts occur where multiple nodes penetrate simultaneously (such as surface-to-surface contact), this application employs a weighted average method based on penetration depth, treating multiple contact points as a single comprehensive contact point and normal vector. This strategy avoids the enormous computational overhead and potential numerical conflicts associated with calculating impulses individually for each contact point, while also reasonably reflecting the mechanical effects of the contact area as a whole, ensuring both computational efficiency and mechanical rationality in multi-body, densely separated scenarios.

[0113] The method of this application will be described below using specific embodiments.

[0114] This application focuses on simulating small ball collisions using the commercial software Fluent in engineering applications. First, a simulation of the collision process of two perfectly elastic, equally mass, eccentrically positioned balls is performed to verify the simulation accuracy and reliability of a single collision. The calculation model is as follows: Figure 4 shown. sphere a and sphere b All have a radius of 0.1m, spheres a Along at a speed of 1 m / s x The sphere moves in the positive direction of the axis. b The two balls were stationary before the collision. y Phase difference in direction ε =0.04m. If there is no slippage on the sphere's surface, then the sphere after the collision... b exist x , y The velocity components in the direction are shown below.

[0115] .

[0116] in, r a Sphere a radius; r b Sphere b radius; ε For two balls in y The difference in direction; m a Sphere a The quality; m b Sphere b quality V Sphere a Velocity before collision.

[0117] The error of the experiment is judged according to the position error discrimination formula, and the expression is as follows.

[0118] .

[0119] in, Error The relative error of the motion trajectory; x s , y s , z s (a sphere) b Location of the center of mass, ( x e , y e , z e (a sphere) b Analytical solution for the location of the centroid.

[0120] The simulation duration is 0.6s, with a collision occurring at 0.1s. The time step is set to 0.001s. Independence verification and error analysis are performed using mesh sizes of 0.02m, 0.01m, 0.005m, 0.002m, and 0.001m. Figure 5 As shown, the results indicate that the selection of the grid size directly determines the accuracy of the calculation results. After the object surface is discretized, the larger the average size of the grid cells, the more significant the deviation between the numerical calculation results of the collision point location and the theoretical reference value. During the collision of the small ball, when the grid size is no higher than 0.002m, the relative error of the trajectory within 0.5s after the collision is less than 10%, which is within the ideal acceptable range.

[0121] To verify the impact of time step on computational accuracy, simulation experiments were conducted with time steps of 0.003s, 0.002s, 0.001s, and 0.0005s, for a total simulation duration of 0.6s. A collision occurred at 0.1s, and the mesh size was 0.002m in all simulations. Figure 6 It is known that there is a certain time error in determining the moment of collision, making absolute precision difficult to achieve. After the initial error occurs, the error continues to accumulate and expand over time as the sphere undergoes its six-degree-of-freedom motion, causing the deviation between the object's trajectory and the analytical solution to gradually increase. Within a 0.5s time range after the collision, when the time step does not exceed 0.002s, its impact on the overall calculation accuracy is relatively small.

[0122] To verify the accuracy and robustness of this method in simulating multiple collisions, the trajectory error of the gravitational direction was analyzed by simulating the free fall and repeated bouncing of a sphere under gravity. Figure 7 As shown, the sphere's radius remains 0.1m, its center of mass is 5.1m above the ground, and the gravitational acceleration is set to 10m / s². 2 .Depend on Figure 8 , Figure 9 It can be seen that as the number of collisions increases, the simulated trajectory and the analytical solution show a continuous upward nonlinear growth trend due to the accumulation of errors. Although the error is amplified with the accumulation of collisions, the overall trend of the trajectory does not diverge. When the number of collisions is small, the simulation results are in good agreement with the analytical solution.

[0123] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0124] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A multi-body separation and collision simulation method based on master-slave surface detection and impulse response, characterized in that, The multi-body separation and collision simulation method based on master-slave surface detection and impulse response includes: Read all sub-mesh data; the sub-mesh data includes the mesh cells and nodes of several objects to be collided with; The subgrid data is preprocessed to obtain preprocessed data; The preprocessed data is then assembled using a dynamic mesh system to obtain the assembled mesh. Based on the assembled mesh, the flow field control equations for the current physical time step are iteratively solved to obtain the flow field results; Based on the flow field results, the motion parameters acting on each object are obtained by integral calculation; the motion parameters include: aerodynamic force and torque parameters. Based on the motion parameters, the position and attitude parameters of the object at the next physical time step are calculated by solving the rigid body six-degree-of-freedom motion equation or by following a preset motion law. Based on the position and attitude parameters of the object in the next physical time step, collision detection and response are performed to determine whether geometric intersection occurs between the objects. If so, the motion parameters are updated and the system returns "Based on the motion parameters, the position and attitude parameters of the object in the next physical time step are calculated by solving the rigid body six-degree-of-freedom motion equation or according to the preset motion law"; otherwise, the motion parameters are directly output. Based on the output motion parameters, the current calculation results are obtained; the current calculation results include: repositioning the spatial positions of each object and adjusting the spatial coordinates and topological relationships of the mesh accordingly; Determine whether the current calculation result meets the set convergence or termination conditions. If not, return "Perform dynamic mesh system assembly on the preprocessed data to obtain the assembled mesh"; if yes, end the simulation process. The execution of the collision detection and response specifically includes: For each slave node, all master face units are traversed sequentially to determine the final target master face unit; the slave node is the node corresponding to the surface area on the subordinate object that participates in contact; the master face unit is the surface area on the master object that participates in contact. Based on the spatial relationship between the target master surface unit and the master-slave surface, and combined with the engineering numerical tolerance setting, a preliminary penetration determination is made to obtain a preliminary determination result; the preliminary determination result includes: no penetration and potential penetration points; Based on the contact information, the collision response is calculated, and the effective contact points are further filtered to obtain the filtered effective contact points.

2. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 1, characterized in that, The execution of the collision detection and response also includes: When the preliminary determination result is a potential penetration point, it is determined whether the projection point is located within the final target unit content; if not, it is determined that there is no penetration; if so, it is determined that it is a valid contact point. Contact information is extracted based on the effective contact points; the contact information includes: contact point, penetration depth, and contact normal vector. Based on the contact information, the collision response is calculated, and the effective contact points are further filtered to obtain the filtered effective contact points; wherein, the secondary filtering condition is: retain contact points that satisfy the condition that the penetration depth is greater than or equal to the minimum penetration depth and the relative normal component at the contact point is less than or equal to 0. Determine whether the filtered valid contact points are multi-point contacts; if so, construct an equivalent single-point contact model using penetration depth weighted average to obtain equivalent contact points and equivalent contact normal vectors, and calculate the normal impulse based on the equivalent contact points and the equivalent contact normal vectors; if not, calculate the normal impulse based on the filtered valid contact points and the contact point normal vectors. Based on the normal impulse, the motion states of the two objects are updated; the motion states include linear velocity and angular velocity.

3. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 2, characterized in that, For each child face node, traverse all primary face units sequentially to determine the final target primary face unit, specifically including: Calculate the spatial distance from the face node to the center of each main face unit, and perform preliminary screening based on a pre-set distance threshold to obtain a set of candidate units after screening; Calculate the signed distance from the face node to the plane containing each candidate cell; The cell with the smallest absolute value of the signed distance is taken as the final target principal surface cell.

4. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 3, characterized in that, The formula for calculating the signed distance is: ； in, d For signed distance; The unit outward normal vector of the candidate principal face element; P s For surface nodes; V m,0 Let be any vertex of the target principal face element.

5. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 2, characterized in that, Based on the spatial relationship between the target master surface unit and the master-slave surface, and combined with the engineering numerical tolerance setting, a preliminary penetration determination is performed to obtain a preliminary determination result, specifically including: Set penetration tolerance; Determine whether the signed distance corresponding to the target unit is less than the penetration tolerance; if not, determine that there is no penetration; if so, determine that it is a potential penetration point.

6. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 2, characterized in that, When the preliminary determination result is a potential penetration point, determine whether the projection point is located within the final target unit content; If not, it is determined that there is no penetration; If so, it is determined to be a valid contact point, specifically including: Calculate the orthogonal projection points of the face nodes onto the plane of the target main face element; Calculate the centroid coordinates of the orthogonal projection points; Determine whether the centroid coordinates of the orthogonal projection point are greater than the numerical tolerance; if not, it is determined that there is no penetration; if so, it is determined that it is a valid contact point.

7. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 6, characterized in that, The formula for calculating the orthogonal projection points is: ； in, These are the orthogonal projection points of the face nodes onto the target main face element plane; d For signed distance; The unit outward normal vector of the candidate principal face element; P s For surface nodes.

8. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 2, characterized in that, Based on the effective contact points, contact information is extracted, specifically including: Use the orthogonal projection points as the contact points; The absolute value of the signed distance corresponding to the target principal surface unit is taken as the penetration depth; The unit outward normal vector of the target master face element is used as the contact normal vector.

9. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 2, characterized in that, The formula for calculating the normal impulse is: ； r A =P eq -C A ； r B =P eq -C B ； ； ； ； ； in, J Normal impulse; m A , m B These are the masses of the object belonging to the primary face and the object belonging to the main face, respectively. I A , I B Let be the inertial tensor of the object belonging to the main surface and the object belonging to the main surface in the inertial frame; The relative velocity in the normal direction before the collision; r A , r B These are the radius vectors pointing from the centroid of the object belonging to the main surface to the equivalent point of contact, respectively. C A Let the centroid of the object to which the surface belongs be denoted. C B The direction in which the centroid of the object to which the main face belongs points; P eq Equivalent contact point; This is the equivalent contact normal vector; For the first i Penetration depth of each effective contact point P c,i For the first i The coordinates of the positions of the effective contact points; The relative velocity at the point of contact; n c,i For the first i Normal vector of each effective contact point; , These are the node velocities at the contact points between the objects belonging to the primary surface and the objects belonging to the secondary surface.

10. The multi-body separation and collision simulation method based on master-slave surface detection and impulse response according to claim 2, characterized in that, The update expression for the motion state is: ； ； in, m A , m B These are the masses of the object belonging to the primary face and the object belonging to the main face, respectively. J Normal impulse; V A , V B These are the centroid velocities of the object belonging to the main face and the object belonging to the main face, respectively. This is the equivalent contact normal vector; I A , I B Let be the inertial tensor of the object belonging to the main surface and the object belonging to the main surface in the inertial frame; r A , r B These are the radius vectors pointing from the centroid of the object belonging to the main surface to the equivalent point of contact, respectively. These are the angular velocities of the object belonging to the secondary face and the object belonging to the primary face, respectively.

Images