JP4466102B2 - Method for creating structure simulation model and structure simulation method - Google Patents

Description

  The present invention has at least two constituent members, and one of the constituent members is embedded in the other constituent member, or one structural member and the other constituent member are joined together. BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a simulation model creation method and simulation method, and more particularly to a tire simulation model creation method and simulation method in which a plurality of constituent members such as a belt member, a carcass member, a tread member, and a side member are stacked.

A finite element method is used to simulate the behavior of a tire formed by laminating a plurality of rubber constituent members and laminating reinforcing members such as a belt member and a carcass member as a reinforcing layer.
In the tire simulation model created as a finite element model, the model of the reinforcing layer formed by reinforcing members such as belt members and carcass members is represented by membrane elements and shell elements, and the model of the rubber member surrounding the reinforcing layer is a solid element Is represented by Then, the nodes of the membrane element and shell element of the model of the reinforcing layer are created so as to be shared with the nodes of the solid element surrounding the element. In addition, even on the joint surface where the two rubber members are joined to each other, the finite elements are created so as to share the nodes.

  In Patent Document 1 below, a triangular solid element is generated by an automatic mesh for a rubber member portion other than the reinforcing layer, and the nodes of the solid element are shared with the nodes of the membrane element and the shell element of the reinforcing layer. A method for creating a simulation model has been proposed. Thereby, it is said that the design and development of the structure, shape, etc. of the tire that achieves the predetermined target performance of the tire can be made efficient and easy.

  In Patent Document 1, since the finite element in the reinforcing layer and the finite element in the rubber member are set so that the nodes are shared at the boundary surface, the shape of the constituent member is changed for reinforcement for optimal design and robust design. When the position of the layer is changed, the finite element model of the rubber member must be reconstructed at the same time so that the nodes on the boundary surface are shared with each other. That is, the mutual finite element models must be reconfigured so that the nodes of the finite elements are shared by the surfaces to be joined in consideration of the joining of the finite element models to be joined. For this reason, the positions of the nodes provided in the model are limited so as to share the nodes in a portion where the models are coupled, and the arrangement of the finite elements may be limited in the reconstruction of the model to be created.

Such model reconstruction is a complicated process, and model reconstruction must be performed particularly frequently in optimal design or robust design that searches for a shape that satisfies a predetermined target performance while sequentially changing the shape of the model. However, there is a problem that the design efficiency is poor.
Further, in Patent Document 2 below, a tire tread pattern element model having a longitudinal groove and a lateral groove and a tire body part element model to be joined to the tire model are separately created, and then the tire finite element to which the two models are joined. A method for creating a model is disclosed. Here, when joining the created tread pattern element model to the tire body part element model, it is defined and joined so as to be forcibly displaced so that the relative position of the nodes does not change ([0008] in Patent Document 2, [0028]) does not disclose how to provide this forced displacement.

WO94-916877 Japanese Patent No. 3314082

  Therefore, the present invention provides a model that should be coupled as in the prior art when reconstructing a simulation model of a structure such as a tire that is configured by coupling a structural member such as a reinforcing member to a structural member such as a rubber member. How to create a structure simulation model and create this model without the need to create a model so that nodes are shared at the joint, and to easily connect this model to other models to be joined An object of the present invention is to provide a simulation method using the.

In order to achieve the above object, the present invention provides a method for creating a simulation model of a structure configured by combining at least two constituent members with an arithmetic device including a CPU , wherein the CPU includes a plurality of CPUs. of using finite element constitutes the whole structure of a simulation model comprising a plurality of components, from the simulation model, to separate the first member model simulates at least one of said structural member, the structural member And at least one of the first member model and the second member model, the step of distinguishing from the second member model reproducing at least one component member other than the one reproduced by the first member model Reconstituting the member model of the first part model, and after reconfiguring the member model, the separated first member model is converted to the second part. When the first member model is placed at a predetermined position with respect to the second member model to recombine with the material model, the nodes of the finite element of the first member model are The finite element of the second member model is specified when located inside or on the boundary of the finite element of the member model, and the shape of the finite element is shaped using a shape function from the reference shape in the parametric space. By determining as converted, position information of corresponding points in the reference shape of each node of the first member model is obtained, and each position of the first member model is obtained using the position information and the shape function. a step of determining a constraint condition for restricting the physical quantity of nodes, simulator which integrates by coupling together the first member model after reconstitution and said second member model using the constraint And creating a Deployment model, it provides a method of creating a simulation model of the structure, characterized by the execution.

Et al is, in the step of defining the constraint condition at the time of placing the first member model to the second member model, the boundary lines connecting the nodes of the finite elements of the first member model Finding an intersection that intersects the boundary of the finite element of the second member model, adding this intersection as a node in the first member model, rearranging the finite element of the first member model, Preferably, the constraint condition is determined using the first member model rearranged. In the step of determining the constraint condition, if the boundary line connecting the nodes of the finite element of the second member model intersects the boundary of the finite element of the first member model, the intersection is obtained and obtained. Preferably, an intersection point is added as a node in the first member model, and the finite elements of the first member model are rearranged.

Further, at least a portion of at least one member model of the first member model and the second member model represents the pre-SL components of a plurality of element points, when the center each of the plurality of element points The position information of a plurality of element points included in the predetermined range and a weight function in which a weighting factor is determined according to the distance from the center of these element points is used to determine a predetermined physical quantity within the predetermined range. It is also preferable to create a member model by defining an interpolation function representing the distribution. That is, the first member model may be a discretized model by a mesh-free method. Further, the reconstructed first member model may also be a discretized model by a mesh-free method. Furthermore, the model may be a combination of a model based on the mesh-free method and a model configured using finite elements.

  The constituent member reproduced by the first member model is, for example, a reinforcing member in which the elastic modulus in a predetermined direction of the constituent member is 100 times higher than the elastic modulus in any direction orthogonal to this direction. . For example, in a tire, it is a reinforcing member such as a belt member for a rubber member. In this case, the first member model is preferably a layered model having a rigidity equivalent to the rigidity of the reinforcing member.

Further, at least one of the first member model and the second member model is, for example, a model reproducing a rubber member, and the structure is, for example, a tire.
More specifically, the first member model is a model reproducing a reinforcing member including at least one of a belt member and a carcass member of a tire, and the second member model is a tire excluding the reinforcing member. This model reproduces the main body.
Alternatively, the first member model may be a model having a tread pattern provided continuously in the tire circumferential direction as a shape.

  In addition, the present invention includes a step of creating a simulation model of a structure using the method of creating a simulation model of the structure, and a step of performing a simulation operation by giving a predetermined condition to the simulation model A structure simulation method is provided.

  In this case, the simulation calculation is performed by, for example, simulating the behavior of the structure that is affected by the solid or liquid when the structure is in contact with at least one of a solid and a fluid. At that time, the contact surface of the structure simulation model that reproduces the contact surface of the structure that contacts at least one of the solid and the fluid has a friction coefficient that depends on at least one of the contact pressure and the sliding speed. It is preferable to perform the simulation calculation by assigning it. The predetermined condition simulates a use condition when the structure is used, and the performance of the structure is predicted using a predetermined physical quantity obtained by the simulation calculation. preferable.

In the present invention, the simulation model of the structure, to separate the first member model reconstituted predetermined position of the second member models to recombine the first member models this reconstituted When the first member model is disposed, the discrete points of the second member model surrounding the discrete points of the portion to be coupled with the first member model are determined, and the physical quantities in the first member model are regulated using the physical quantities at the discrete points. Define restraint conditions to be applied. For this reason, since it can be easily coupled to the model to be coupled using this constraint condition, it is not necessary to create a model so that the nodes are shared in the coupled portion with the model to be coupled as in the conventional case.

  Hereinafter, a method for creating a structure simulation model and a method for simulating a structure according to the present invention will be described in detail based on the preferred embodiments shown in the accompanying drawings.

FIGS. 1A to 1C are cross-sectional views of a finite element model of a passenger car tire created as a simulation model of a structure according to the present invention.
First, as shown in FIG. 1A, a tire simulation model in which each component of the tire is reproduced by a finite element is created, and a simulation calculation is performed using the tire simulation model. In this simulation, the finite element is subjected to extremely large deformation and cannot be calculated, and the calculation result is inconvenient when the calculation result diverges. In this case, it is necessary to reconfigure the model by rearranging the finite elements in the finite element model or changing the shape of the member model.

The tire simulation model shown in FIG. 1A is a model to be reconstructed, and is a model in which a tread member model 10, a base member model 11, and a reinforcing layer model 12 are combined and integrated. . 1 (b) and 1 (c) show that a tread member model 10, a base member model 11 and a reinforcing layer model 12 are separated from each other as a finite element model from a simulation model of a tire to be reconstructed. The removed state is shown.
The tread member model 10 and the base member model 11 are each composed of a three-dimensional finite element such as a hexahedral solid element, a pentahedral solid element, or a tetrahedral solid element, and the reinforcing layer model 12 is composed of a finite element of a line element. . Fig.1 (a)-(c) has shown the cross section of each model when cut | disconnecting along the rotating shaft of a tire. The tire simulation model is a three-dimensional model. However, in the present invention, a two-dimensional model having a tire cross-sectional shape may be used.

  As described above, when inconvenience occurs in the simulation calculation, at least one member model of the tread member model 10, the base member model 11, and the reinforcing layer model 12 shown in FIGS. The simulation model is taken out separately, and the member model is reconfigured by changing the configuration of the simulation model, for example, the number of finite elements, the position of the node, the shape of the member, or the like. The reconstructed member model may be a separated model (first member model) or a model after separation (second member model). Thereafter, the retrieved member model is again placed in the simulation model for recombination. For example, the reinforcing layer model 12 is arranged so as to be embedded in the base member model 11. Further, the tread member model 10 is disposed so as to be joined to the base member model 11. Then, the tire models are created by combining the member models using the constraint conditions described later.

FIG. 2 is a block diagram of an arithmetic unit 20 that performs a tire simulation calculation using a tire simulation model that is created by recombining member models that have been reconfigured.
The computing device 20 is configured by modularizing a condition setting unit 22, a model creation unit 24, a simulation computation unit 26, a model correction unit 28, and a performance prediction unit 30 as subroutines. In addition, it has CPU32 which performs the process of each said part substantially, and the memory 34 which memorize | stores the process result obtained by each part.

  The condition setting unit 22 is a part for setting various conditions such as simulation model creation conditions and simulation calculation conditions, target performance, and the like in accordance with conditions input from a keyboard or mouse (not shown). For example, when performing optimal design, specify conditions such as the performance to be optimized, target performance, design of areas such as tire shapes that are to be changed, designation of areas that are not to be changed, or ranges where design changes are possible. The procedure for changing the simulation model for optimal design is determined according to these conditions.

  The model creation unit 24 is a part that creates a tread member model 10, a base member model 11, and a reinforcing layer model 12 according to a predetermined procedure, and creates a tire simulation model. In the tire simulation model, as shown in FIG. 1A, the joints of the tread member model 10, the base member model 11, and the reinforcing layer model 12 are shared. In the simulation model to be created, as described later, the nodes may be non-shared at the portion where each member model is coupled. In this case, as will be described later, the models can be combined using a constraint condition that restricts the physical quantity of the node.

  The simulation calculation unit 26 is a part that handles the created tire simulation model as an axisymmetric model and performs the simulation calculation according to the simulation calculation conditions set by the condition setting unit 22. For example, a rim model (not shown) created separately is attached to a tire simulation model, and an internal pressure filling process is performed so as to reproduce the internal pressure filling by applying a certain load to the inner peripheral surface of the simulation model. Further, the simulation model after the internal pressure filling process is grounded with a load applied to a rigid road surface model (not shown) to create a grounded simulation model. Further, a translation model and a rotational angular velocity are given to the ground contact state simulation model to create a running state simulation model that reproduces the state in which the tire travels on the road surface.

Such a traveling state reproduces the traveling state of the dry road surface, and a simulation calculation is performed using this state. The simulation calculation is not particularly limited and may be a known calculation. For example, the camber angle, slip angle, and braking torque are added to the simulation model to reproduce the dry performance during cornering with camber angle and slip angle, and the braking / driving performance when braking torque and driving torque are applied to the tire rotation axis. Give drive torque. In addition, the stress / strain analysis of the simulation model at a predetermined traveling speed is performed.
In addition, a road surface model including a fluid model that reproduces the wet road surface is separately created, and a simulation calculation is performed in which the simulation model passes through the fluid model created on the road surface model. In other words, the behavior of the tire is calculated when the tire is in contact with at least one of a solid such as a road surface or a fluid such as a water film or air and the tire is affected by the solid or liquid.

  As described above, the simulation calculation uses conditions (various conditions such as load, internal pressure, travel speed, road surface condition, etc.) that simulate use conditions when the tire is used. Further, in the simulation calculation, eigenvalue analysis for obtaining the natural frequency of the tire in the ground contact state may be performed. Further, in order to reproduce the state of conduction of heat generated in a part of the tire, a heat conduction analysis may be performed by giving an initial temperature condition and boundary conditions. In this case, thermal conductivity or the like is given as a material constant.

The model correction unit 28 performs simulation calculation when the simulation model is locally deformed in simulation calculation and the calculation becomes impossible, the calculation diverges during the calculation, or the calculation result becomes a unique result. When inconvenience occurs, the simulation calculation is stopped, the member model in question is separated from the simulation model, and the member model is reconfigured. For example, when a part of the reinforcing layer model 12 is extremely deformed and simulation calculation is impossible, the reinforcing layer model 12 is separated and taken out, the position of the finite element, the number of elements, etc. are changed, Rearrange.
The reconfigured member models are recombined using constraint conditions described later, a simulation model is created, and provided to the simulation calculation unit 26.
In this way, the member model is reconfigured so that the result of the simulation calculation is not impossible, diverges, or produces a peculiar result.

  The performance prediction unit 30 generates a rolling simulation model, and uses this simulation model and a road surface model that reproduces a dry road surface or a road surface model that includes a fluid model that reproduces a wet road surface. The characteristic physical quantity of the simulation model 14 in the dynamic state is calculated based on the simulation calculation result that reproduces the behavior of the dynamic tire traveling on the wet road surface, and the dry performance or the wet performance is calculated based on the calculated characteristic physical quantity. It is a part which predicts the quality of etc.

For example, dry performance during cornering with camber angle or slip angle, braking / driving performance when braking torque or driving torque is applied to the tire rotation shaft, or wet performance such as hydroplaning performance on wet road surfaces. . When performing a wet performance simulation calculation, the buoyancy that the fluid model acts on the simulation, the tread force that the simulation model acts on the road surface model, the pressure distribution of the fluid model, or the flow velocity, flow rate, energy density in the fluid model, or The distribution of energy, the contact shape of the simulation model, the contact area or the contact pressure distribution, etc. are calculated as characteristic physical quantities.
The calculated characteristic physical quantity is output to an unillustrated display or printer as an index of the prediction result.

  The creation of the tire simulation model and the simulation calculation are sequentially performed in accordance with the procedure determined by the condition setting unit 22 while changing the simulation model, and performance prediction is performed.

  In the present invention, at the model correction unit 28, at least one member model of the tread member model 10, the base member model 11, and the reinforcing layer model 12 is taken out from the tire simulation model prepared in advance and taken out. The member model is reconfigured and recombined as a tire simulation model.

FIG. 3 is a flowchart of a method for creating a tire simulation model, which is an example of a method for creating a simulation model for a structure according to the present invention, and shows a flow of recreating a tire simulation model by reconfiguring a member model. ing.
The tire simulation model creation method described below reconstructs the separated member model. However, in the present invention, the remaining member model after separating the member model (first member model) ( The second member model) may be subject to reconstruction.

First, a member model to be reconfigured is specified from a simulation model obtained by combining and integrating each member model that reproduces each component member, and this member model is separated from the simulation model and extracted (step S10). . Identification of the member model to be reconfigured in the simulation model is performed by, for example, examining physical quantities acting on each node or finite element of the model, such as displacement or distortion, during the simulation calculation, and the value of this physical quantity becomes singular. This is performed by detecting a member model of a portion that diverges due to a particularly large portion or value.
As a method for connecting the member models in the simulation model, there are a method for sharing the nodes in the connecting portion, a method for defining a constraint relationship that does not change the relative displacement between the nodes in the connecting portion, and the like. For example, when member models are connected by sharing nodes, a new node is created at the shared node position, and one node is a node of the separated member model, and the node of the member model is separated Thus, the member models are divided. When the constraint relationship is defined, the member models are classified by removing the constraint relationship.

Next, the separated member model is reconstructed (step S12).
The reconfiguration of the member model is, for example, moving the node position of a finite element, changing the type of a finite element from, for example, a tetrahedral solid element to a pentahedral solid element or a hexahedral solid element, or adding a finite element This means changing the model configuration by changing the number of finite elements and the number of nodes. Of course, it is also possible to correct the shape of the member model by correcting the length of the constituent members.

Next, the first member model is combined to recombine the reconstructed member model (first member model) with the remaining member model (second member model) separated from the simulation model. Placed in position. At that time, the nodes of the second member model surrounding the nodes of the parts to be coupled of the first member model are determined, and a weighting factor that defines the constraint condition used when the member models are coupled is calculated using the nodes. (Step S14).
For example, the reinforcing layer model 12 is reconfigured by separating the reinforcing layer model 12 from the tire simulation model shown in FIG. 4A and 4B are diagrams illustrating the reconstructed reinforcing layer model 12.

  The reinforcing layer of the tire is formed of a cord embedded in one direction in the rubber coat member, and the cord is arranged in parallel at a constant interval. The cord in FIG. 4A is a cord in which a plurality of strands have a twisted structure. Assuming that the cross-sectional area of such a cord is A and the number of cords included in the width W is N, as shown in FIG. Is represented. The thickness h of the reinforcing layer model 12 is expressed as h = A · N / W and modeled. When such a reinforcing layer model 12 is represented by a shell element or a membrane element, the thickness h is not reflected in the substantial shape, but is calculated as having the thickness h in the simulation calculation. In addition, a reinforcement layer is comprised with a member whose elasticity modulus of a reinforcement direction is 100 times or more higher than the elasticity modulus of any direction orthogonal to this reinforcement direction like a steel wire.

Thus, reinforcing layer model 12 to correct the position of the nodes of the finite element, by changing the type of the finite element, or reconstructed by changing a finite number of elements and the number of nodes by adding a finite element . The reconstructed reinforcing layer model 12 is arranged at a predetermined position of the base member model 11 of the tire simulation model.

5 (a) to 5 (g) simplify the case where the reinforcing layer model 12 is arranged in a tire body model composed of a tread member model 10 and a base member model 11, and a single reinforcing layer model is a base member model. It is a figure explaining the case where it is embed | buried under.
FIG. 5A is a diagram in a case where a reinforcing layer model having finite elements 1001 and 1002 (nodes 101, 102, 103) is arranged on a tire body model having nodes 1-6. FIG. 5B shows the relationship between element numbers and nodes in the reinforcing layer model. The finite element 1001 is constituted by the nodes 101 and 102 and the line segment therebetween, and the finite element 1002 is constituted by the nodes 102 and 103 and the line segment therebetween.
Next, as shown in FIG. 5C, an intersection (x mark in the figure) where the reinforcing layer model intersects the side (boundary) of the finite element of the tire body model is obtained, and this intersection is determined as a node in the reinforcing layer model. In addition, the finite element of the reinforcing layer model is rearranged and reconfigured. As shown in FIG. 5D, the nodes 202 and 203 are added, and the finite elements 1001 to 1003 are rearranged. The finite element 1002 is composed of nodes 102 and 202 and a line segment between them. Furthermore, when the boundary line connecting the nodes of the finite element of the tire body model intersects the boundary of the finite element of the reinforcement layer model, the intersection is obtained, and the obtained intersection is added as a node in the reinforcement layer model for reinforcement. The layer model is reconstructed.
If a reinforcement layer model is placed in the tire body model and the reinforcement layer model protrudes from the tire body model, the protruding portion is not affected by the tire body model and is therefore deleted from the member model. .

Next, as shown in FIG. 5 ( e ), for each finite element of the tire body model, a node included in the finite element is detected and a list is created. Of course, this node is associated with position information in the model corresponding to the node number.
Further, a weighting factor W of each node described later is obtained for each finite element in the joint portion of the tire body model. For example, as shown in FIG. 5G, when the nodes of the reinforcing layer model are included in the tire body model element 1, the weighting factors W (102 of the nodes 1, 2, 5, 4 of the tire body model element 1 are included. ) _1, W (102) _2, W (102) _5, W (102) 4.

FIG. 6 is a diagram for explaining the conversion T used when calculating the weighting coefficient W. This transformation T is a square on the physical space formed by the nodes 1, 2, 5, 4 of the tire body model element 1 and the line segments connecting these nodes. A reference shape in the parametric space, for example, one side It is converted to a square with a length of 2, nodes 1, 2, 5, and 4 at each vertex of the square, each side of the quadrangle is each side of the square, and the inner area of the quadrangle is the inner area of the square. Mapped. Further, the inverse transformation T- ^{1 is used} to convert the square into a quadrangle, the square inner area is mapped to the square inner area, each vertex of the square is mapped to each node of the square , and each side of the square is mapped to each side of the square. The That is, the transformation T maps the tire body model elements in the XY coordinate space on a one-to-one basis on the squares in the RS coordinate space.
Therefore, the corresponding points on the square can be obtained by transformation T for the nodes included in the finite elements or located on the sides for each finite element of the tire body model.
More specifically, the position in the XY coordinate space using the shape functions N ₁ (r, s), N ₂ (r, s), N ₃ (r, s), and N ₄ (r, s). The coordinates (x, y) can be associated with the position coordinates (r, s) using the following formula (1). Here, x ₁ , y ₁ , x ₂ , y ₂ , x ₅ , y ₅ , x ₄ , y ₄ are the position coordinates of the nodes ₁ , ₂ , ₅ , ₄ of the finite element of the tire body model, respectively. .
Although a square is used as the reference shape in the parametric space, it may be a triangular shape and is not particularly limited in the present invention. Of course, if the transformation T transforms a three-dimensional shape, a cube, a triangular prism, or a tetrahedron can be used as the reference shape.

Here, the shape functions N ₁ (r, s), N ₂ (r, s), N ₃ (r, s), and N ₄ (r, s) are defined in FIGS. Function. Here, when the matrix in the expression (1) is M, the components of the matrix M are the weighting factors W at the nodes described above. Specifically, the weighting factor W (102) _1 of the node 102 (position coordinates (x ₀ , y ₀ )) for the node 1 is the position coordinate of the corresponding point in the RS space coordinate of the node 102 (r ₀ , S ₀ ) is N ₁ (r ₀ , s ₀ ). Therefore, the position coordinates (x ₀ , y ₀ ) of the node 102 can be expressed by the following equation (2).
In this way, the weighting factor W is calculated.

  The calculated weighting factor W is determined and listed for each node of the reinforcing layer model 12, but the weighting factor W is determined for each finite element of the tire body model. For the nodes (intersections) of the reinforcing layer model located at, the weighting factor W is set to be double with different finite elements. For this reason, one of the overlapping nodes is deleted (step S16).

Next, the physical quantity acting on each node of the reinforcing layer model is expressed by the following formulas (3) to (5) using the physical quantity of the node of the finite element of the tire body model including the node and the weight coefficient W. A constraint equation as shown is set as a constraint condition (step S18).
Assuming that predetermined physical quantities of nodes 1, 2, 5, and 4 in FIG. 5E, for example, displacement in the X direction are u (1), u (2), u (5), and u (4), node 102 The displacement u (102) in the X direction at is determined by the following equation (3). Similarly, the displacements u (101) and u (202) at the nodes 101 and 202 are determined as in the following equations (4) and (5).

  Thus, the behavior of each finite element of the reinforcing layer model can be regulated by expressing the physical quantity acting on the node of the reinforcing layer model by the physical quantity of the node of the finite element of the tire body model including the node.

Then, the reinforcing layer model is coupled to the tire body model using the set constraint equation (Step S20).
The model combination can be used for various known simulations.
For example, when the deformation behavior of a tire simulation model is dynamically calculated, that is, when dynamic analysis is performed, an explicit method is performed in which the following calculation is performed for each predetermined time step. At a certain time step, among the nodes of the reinforcing layer model as shown in FIG. 1C, the reinforcing layer model acts on a boundary node (a constrained node) that intersects a side (boundary) of the finite element of the tire body model. The distribution force and the mass of this boundary node are distributed to the nodes of the tire main body model (the nodes to be constrained) according to the weighting factor W, and the mass of these nodes (the nodes to which the tire main body model is constrained) and the action Correct the power. In this case, after the above distribution is performed for all the boundary nodes of the reinforcement layer model, the accelerations of the nodes of the reinforcement layer model and the tire body model are calculated excluding the boundary nodes of the reinforcement layer model. To do. Thereafter, the acceleration of the node restrained by the tire body model is distributed to the boundary node of the reinforcing layer model using the weighting factor W. Thus, the acceleration of the boundary node in the reinforcing layer model can be obtained. The force applied to the boundary node is obtained using the acceleration at the obtained boundary node and the mass at the boundary node. Thus, the force acting in the next time step is obtained. Of course, when a force is applied to the boundary node in the next time step as an external force, this force is also added. Such an explicit method is suitable for a simulation calculation that dynamically reproduces the behavior of the tire.

Moreover, when performing a static analysis as follows, an implicit method can also be performed.
For example, a model matrix in which the degrees of freedom of the boundary nodes of the reinforcing layer model are eliminated from the constraint equation is created. This method is suitable for simulation calculation that statically reproduces the behavior of the tire, and the degree of calculation required for analysis can be shortened because the degree of freedom in the entire matrix is reduced by eliminating the degree of freedom of the boundary nodes.
In addition, instead of the method of eliminating the degree of freedom of the boundary node from the constraint equation, the diagonal component of the boundary node of the constrained reinforcement layer model and the node of the tire body model to be constrained in the matrix of the entire model By adding the penalty number of the same value to, and subtracting the penalty number of the same value from the intersection component, a modified model matrix is created.

For example, in a model matrix such as the following equation (6), when a constraint equation of u ₃ = u ₂ is determined, the matrix component of u ₃ is deleted as shown in the following equation (7), and u ₁ , U ₂ , u ₄ .
Further, as shown in the following equation (8), a penalty coefficient K is added to the diagonal components k ₂₂ and k ₃₃ corresponding to u ₂ and u ₃ in the matrix, and the intersection component of u ₂ and u ₃ in the matrix A penalty coefficient K having the same value is subtracted from k ₂₃ and k ₃₂ . Here, the penalty coefficient K is, for example, a very large value obtained by multiplying 10 ¹⁰ times the larger of the diagonal components k ₂₂ and k ₃₃ of u ₂ and u _{3 in} the matrix.

  In the separation and recombination of the member models, the reinforcing layer model is embedded in the tire body model and joined, but the tread member model 10 as shown in FIG. 1B is joined to the base member model 11. It may be a thing.

In this joining method, first, the tread member model 10 reconfigured with respect to the base member model 11 is arranged at a predetermined position so as to join. At this time, the node of the tread member model 10 located on the boundary of the finite element of the base member model 11 is obtained as the boundary point.
In the example shown in FIG. 8, node _{a 4, a 5, a 6} of nodes _a 1 ~a ₆ is determined as a boundary point of the joint surface.

Next, using the position coordinates at this boundary point and the position coordinates of the nodes of the finite element of the base member model 11 having this boundary point on the boundary, the weighting coefficient W is calculated in the same manner as described above.
Next, using this weighting factor W, a constraint condition is defined in which the physical quantity of the boundary point is represented by the physical quantity of the node of the finite element of the base member model 11 having the boundary point on the boundary. For example, the physical quantity of the boundary point a ₅ of the tread member model 10 in FIG. 8 is represented by the physical quantities of the nodes 2 and 3 of the base member model 11.
Next, using this constraint condition, as in the above-described coupling method, the mass of the nodal point and the acting force of the base member model are corrected, and each model is combined in a matrix related to the mass and stiffness of the synthesized model. The tread member model 10 is combined by correcting the matrix by deleting the degrees of freedom of the boundary points or correcting the matrix by adding and subtracting a penalty coefficient to the matrix components.
In this way, a simulation model in which the tread member model 10 is coupled to the base member model 11 on the boundary can be created.
In the present invention, the relative displacement does not occur between the nodes of the finite elements of the joined parts, regardless of the joining method with the tire body model and the base member model before dividing the reinforcing layer model and the tread member model. By defining the constraint relationship, the reinforcing layer model and the tread member model can be combined with the tire body model and the base member model by the method according to Step 20 described above.

In the creation of the conventional simulation model, the joints of the tread member model and the base member model share nodes, and the nodes of the reinforcing layer model placed in the base member model share the nodes of the base member model. The model was created. For this reason, when the reinforcing layer model is changed, the finite element of the base member model has to be reconfigured so that the sharing of the nodes is maintained in accordance with the change. On the other hand, in the present invention, since the models are combined using the constraint condition, it is not necessary to configure the model so as to share the nodes in the embedded member model or the member model joined at the boundary surface. Can be created.
A simulation calculation is performed on the simulation model thus corrected.

  Note that the creation of the simulation model has been described by taking as an example the case where the simulation calculation becomes inconvenient or the finite element needs to be reconfigured, for example, the simulation model calculation becomes impossible or the result is unusual. The simulation model creation method can also be used when a tire shape search is performed for optimum design or robust design.

Moreover, although the tread member model 10 shown in FIG. 1B is a model composed of finite elements, the present invention is not limited to this. For example, the tread member is represented by a plurality of element points (discrete points), and the position information of a plurality of element points included in a predetermined range (support) when each of the plurality of element points is the center, and these elements Even if a tread member model is created by determining an interpolation function that represents a distribution of a predetermined physical quantity within a predetermined range (support) using a weighting function in which a weighting coefficient is determined according to the distance from the center of the point. Good. Such a model creation method is called a mesh-free method because there is no mesh division for forming a finite element. The model based on the mesh-free method is different from the interpolation function representing the physical quantity of the finite element model that is represented by the physical quantity of the nodes constituting the finite element, according to the movable element point included in the support. Is set. This is a major difference from the finite element model. In the present invention, such a mesh-free model may be used as a member model to be separated (first member model), or the remaining model after separation of the member model (second member model). May be used. The model based on the mesh-free method is described in detail in Japanese Patent Application No. 2003-179732 by the applicant of the present application. Further, the tread member model 10 may be a model in which a model based on a finite element and a model based on a mesh-free method are combined.
Reconstituted member model is partitioned from the tire body model Other Torre head member model and reinforcing layer model may be a side member models, etc., it is not particularly limited.

  The method for creating a simulation model of a structural body and the simulation method of the present invention have been described based on tires. However, the present invention is not limited to tires composed of rubber members, and has at least two structural members, of which Any structure may be used as long as one structural member is embedded in or joined to the other structural member.

  As described above, the method for creating the simulation model of the structural body and the simulation method of the present invention have been described in detail. However, the present invention is not limited to the above embodiment, and various improvements and modifications can be made without departing from the gist of the present invention. Of course.

(A)-(c) is a figure which shows an example of the shape of the finite element model of the tire for passenger cars created as a simulation model of a structure in this invention. It is a block diagram of the arithmetic unit which performs performance prediction using the tire simulation method which is an example of the simulation method of this invention. It is a flowchart which shows the preparation method of the simulation model of the tire which is an example of the preparation method of the simulation model of the structure of this invention. (A), (b) is a figure explaining the preparation method of the reinforcement layer model in the simulation model of a tire. (A)-(g) is a figure explaining the principal part of the preparation method of the simulation model of a tire. It is a figure explaining the principal part of the preparation method of the simulation model of the structure in this invention. (A)-(d) is a figure explaining the example of the shape function used in the preparation method of the simulation model of the structure in this invention. It is a figure explaining the example of the preparation method of the simulation model of the structure in this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Tread member model 11 Base member model 12 Reinforcement layer model 20 Computation device 22 Condition setting part 24 Model creation part 26 Simulation calculation part 28 Model correction part 30 Performance prediction part

Claims (14)

The arithmetic unit having a CPU, a method of making a simulation model of the at least two components is configured by coupling structure,
The CPU is
Using a plurality of finite elements constitute the whole structure of a simulation model comprising a plurality of components, from the simulation model, to separate the first member model simulates at least one of said components, said structure Partitioning a member from a second member model reproducing at least one component member other than the member reproduced by the first member model among the members;
Reconfiguring at least one member model of the first member model and the second member model;
After reconfiguring the member model, the first member model is placed in a predetermined position relative to the second member model in order to recombine the separated first member model with the second member model. When placing, the finite element of the second member model is specified when the node of the finite element of the first member model is located inside or on the boundary of the finite element of the second member model. By determining the shape of the finite element as a shape converted from a reference shape in a parametric space using a shape function, position information of corresponding points in the reference shape of each node of the first member model is obtained. Determining, using the position information and the shape function , a constraint condition for restricting a physical quantity of each node of the first member model;
And creating a simulation model that is integrated by bonding together the first member model after reconstitution and said second member model using this constraint,
How to create a simulation model of the structure, characterized by the execution. In the step of determining the constraint condition, a boundary line connecting nodes of the finite element of the first member model when the first member model is arranged with respect to the second member model is the second member model. An intersection that intersects the boundary of the finite element of the member model is obtained, the intersection is added as a node in the first member model, the finite element of the first member model is rearranged, and the finite element is rearranged. The method for creating a simulation model for a structure according to claim 1 , wherein the constraint condition is determined using the first member model. In the step of determining the constraint condition, if the boundary line connecting the nodes of the finite element of the second member model intersects the boundary of the finite element of the first member model, the intersection is obtained and obtained. 3. The method for creating a simulation model for a structure according to claim 2 , wherein an intersection is added as a node in the first member model, and the finite elements of the first member model are rearranged. Wherein at least a portion of the first member model and at least one member model of the second member model represents the pre-SL components of a plurality of element points, given when the center each of the plurality of element points Distribution of a predetermined physical quantity within the predetermined range using positional information of a plurality of element points included in the range and a weight function in which a weighting coefficient is determined according to the distance from the center of these element points. The method for creating a simulation model for a structure according to any one of claims 1 to 3 , wherein a member model is created by determining an interpolation function to be represented. The constituent member reproduced by the first member model is a reinforcing member whose elastic modulus in a predetermined direction of the constituent member is 100 times or more higher than the elastic modulus in any direction orthogonal to the direction. 5. A method for creating a simulation model of a structure according to any one of 1 to 4 . 6. The method for creating a simulation model for a structure according to claim 5 , wherein the first member model is a layered model having rigidity equivalent to that of the reinforcing member. The method for creating a structure simulation model according to any one of claims 1 to 6 , wherein at least one of the first member model and the second member model is a model reproducing a rubber member. The method of creating a structure simulation model according to claim 7 , wherein the structure is a tire. The first member model is a model reproducing a reinforcing member including at least one of a belt member and a carcass member of a tire, and the second member model is a model reproducing a tire main body excluding the reinforcing member. The method for creating a simulation model for a structure according to claim 8 . 9. The method of creating a structure simulation model according to claim 8 , wherein the first member model has a tread pattern provided continuously in a tire circumferential direction. Creating a structure simulation model using the method for creating a structure simulation model according to any one of claims 1 to 10 ;
And a step of performing a simulation operation by applying a predetermined condition to the simulation model. 12. The structure simulation method according to claim 11 , wherein the simulation operation is calculated by simulating the behavior of the structure that is affected by a solid or a liquid by contacting the structure with at least one of a solid and a fluid. A friction coefficient depending on at least one of the contact pressure and the sliding speed is applied to the contact surface of the simulation model of the structure that reproduces the contact surface of the structure that contacts at least one of solid and fluid. The structure simulation method according to claim 12 , wherein the simulation calculation is performed. Wherein the predetermined condition is obtained by simulating the use condition when the structure is used, the simulation calculation using a predetermined physical amount obtained by claim 11 the performance prediction of the structure - 14. The structure simulation method according to any one of items 13 to 13 .

Images