JP4645399B2 - Shock absorber modeling method, shock absorber analysis method, shock absorber modeling device, and shock absorber analysis device - Google Patents

Description

The present invention relates to a shock absorber modeling method for creating a finite element model for analysis by a finite element method, such as a vehicle collision test barrier having a honeycomb structure, a shock absorber analyzing method, The present invention relates to a modeling device and a shock absorber analysis device .

A technique for reproducing a collision test of an automobile by a collision analysis by a finite element method is known (for example, see Non-Patent Document 1). In this technology, a finite element model is created by modeling with a solid (solid) element the honeycomb structure formed as an aggregate of many cylindrical bodies having a substantially regular hexagonal cross section. Good analysis results are obtained.
Koji Nojima, “DB Modeling and Accuracy in Offset Collision”, Automobile Engineering Society 2001 Symposium Text, p37

  However, in each of the conventional techniques described above, since the honeycomb structure is modeled with solid elements, the reproducibility of the deformation mode of the barrier and the colliding body colliding with the barrier is poor, and there is room for improvement in this respect.

In view of the above facts, the present invention provides a shock absorber modeling method and apparatus for creating a finite element model capable of accurately analyzing a shock absorber configured by assembling a plurality of cylindrical bodies. It is another object of the present invention to provide a shock absorber analysis method and apparatus for accurately analyzing a shock absorber using a finite element model created by the shock absorber modeling method.

In order to achieve the above object, a method for modeling an impact absorber according to the invention described in claim 1 is an impact absorber constructed by assembling cylindrical bodies having axes parallel to each other. A shock absorber modeling method performed by a computer for creating a finite element model for structural analysis by a finite element method in the case of receiving a compressive load in the axial direction, the cylindrical body in the actual object to be analyzed Based on the virtual dimension of the cylindrical body that is larger than the dimension of the cylindrical body, the compressive stress in the case where a compressive load in a predetermined axial direction is applied to the cylindrical body of the virtual dimension or an assembly of the cylindrical bodies The virtual material characteristics of the cylindrical body of the virtual dimensions are set so as to be equal to the compressive stress when the predetermined axial load is applied to the target cylindrical body or the aggregate of the cylindrical bodies. The finite element model As an aggregate of the tubular body that is to the virtual size has the virtual material properties modeled by shell elements.

  In the shock absorber modeling method according to claim 1, a finite element model of a shock absorber configured as an aggregate of a large number of cylindrical bodies such as a honeycomb structure is formed. When the model is set so that the virtual dimension of the cylindrical body is larger than the dimension of the actual cylindrical body, virtual material properties are obtained so that the compressive stress of the finite element model substantially matches the actual compressive stress. Then, the shock absorber is modeled as an aggregate of cylindrical bodies having the above-mentioned virtual material characteristics, ie, a structure having a hollow portion in the cylindrical body, with shell elements.

  In this finite element model, the virtual dimension of the cylindrical body is set to be larger than the actual dimension, so the cylinder constituting the shock absorber is compared with a model in which the cylindrical body is modeled with the same dimensions as the actual dimension. The number of bodies, that is, the number of model elements can be reduced. The finite element model is a virtual material obtained by applying a compressive stress equivalent to that of the real object to the virtual body of the virtual dimension when receiving a predetermined load in the compression direction (main direction of the cylindrical body). Since it is modeled using characteristics, it is possible to accurately reproduce the stress and deformation due to compression of the real object even though the number of elements is reduced as described above. That is, with the finite element model created by this modeling method, it is possible to perform accurate analysis while reducing the computer load. In addition, this finite element model uses shell elements to model the shock absorber as a structure having a hollow portion in the cylindrical body, so that the analysis accuracy is improved compared to the model using solid elements. be able to.

  As described above, in the shock absorber modeling method according to the first aspect, it is possible to create a finite element model that can accurately analyze the shock absorber formed by assembling a plurality of cylindrical bodies. It should be noted that the shock absorber to be modeled may be one in which adjacent cylindrical bodies share a part of the cylindrical wall, such as a honeycomb structure, and includes an independent cylindrical body. May be.

In order to achieve the above object, a method for modeling a shock absorber according to a second aspect of the present invention provides a shock absorber comprising a plurality of cylindrical bodies having axes parallel to each other. A shock absorber modeling method performed by a computer for creating a finite element model for structural analysis by a finite element method in the case of receiving a compressive load in the axial direction, the cylindrical body in the actual object to be analyzed Based on the imaginary number of the cylindrical bodies less than the number of cylinders, the compressive stress in the case where a compressive load in a predetermined axial direction is applied to the aggregate of the imaginary number of cylindrical bodies is the actual cylinder to be analyzed A virtual material property of the virtual number of cylindrical bodies is obtained so as to be equivalent to a compressive stress when a compressive load in the predetermined axial direction is applied to the aggregate of the cylindrical bodies, and the finite element model is determined as the virtual element. Said having material properties As an aggregate of virtual numbers of the tubular body is modeled with shell elements.

  3. A method for modeling a shock absorber according to claim 2, wherein a finite element model of the shock absorber configured as an aggregate of a large number of cylindrical bodies such as a honeycomb structure is formed. When the model is set so that the virtual number of cylindrical bodies is smaller than the number of real cylindrical bodies, virtual material properties are obtained so that the compressive stress of the finite element model substantially matches the real compressive stress. Then, the shock absorber is modeled as an aggregate of the virtual number of cylindrical bodies having the virtual material characteristics, that is, a structure having a hollow portion in the cylindrical body, which is a shell element.

  In this finite element model, since the virtual number of cylindrical bodies is set to be smaller than the number of real objects, the number of model elements constituting the shock absorber is compared with a model having the same number of cylindrical bodies as the real objects. Can be reduced. The finite element model is a virtual material obtained by applying a compressive stress equivalent to that of the real object to the virtual body of the virtual dimension when receiving a predetermined load in the compression direction (main direction of the cylindrical body). Since it is modeled using the characteristics, the stress and deformation of the real object can be accurately reproduced while the model has a reduced number of elements as described above. That is, with the finite element model created by this modeling method, it is possible to perform accurate analysis while reducing the computer load. In addition, this finite element model uses shell elements to model the shock absorber as a structure having a hollow portion in the cylindrical body, so that the analysis accuracy is improved compared to the model using solid elements. be able to.

  As described above, in the shock absorber modeling method according to the second aspect, it is possible to create a finite element model that can accurately analyze the shock absorber configured by assembling a plurality of cylindrical bodies. It should be noted that the shock absorber to be modeled may be one in which adjacent cylindrical bodies share a part of the cylindrical wall, such as a honeycomb structure, and includes an independent cylindrical body. May be.

The shock absorber modeling method according to claim 3 is the shock absorber modeling method according to claim 1 or 2, wherein the thickness of the cylindrical wall of the cylindrical body is used as the virtual material property. the Ru used.

  In the shock absorber modeling method according to claim 3, the thickness of the cylindrical wall of the cylindrical body is made different from the actual thickness as a virtual material characteristic, so that the finite element model when subjected to the same compressive load is used. The compressive stress is equivalent to the actual compressive stress. For this reason, it is easy to set the virtual material characteristic that matches the compressive stress when a predetermined compressive load is applied, that is, the thickness of the cylinder wall, between the real object and the finite element model.

A shock absorber modeling method according to a fourth aspect of the present invention is the shock absorber modeling method according to any one of the first to third aspects, wherein the cylindrical shape constituting the finite element model is used. the cross-sectional shape along a plane perpendicular to the axis of the body, the shape different from the sectional shape along a plane perpendicular to the axis of the tubular body that constitutes the real of the analysis.

  In the shock absorber modeling method according to claim 4, for example, the cross-sectional shape of the real cylindrical body is a regular hexagonal shape, whereas the cross-sectional shape of the continuous body of the finite element model is square or uniaxial direction. Modeled as a hexagon, circle, etc. Thereby, the number of elements of the finite element model can be further reduced.

A shock absorber modeling method according to a fifth aspect of the present invention is the shock absorber modeling method according to any one of the first to fourth aspects, wherein the virtual material property is expressed by the cylindrical body. Ru varied in the axial sections.

  In the shock absorber modeling method according to claim 5, since the virtual material characteristics are different (changed) in the axial direction (main direction) of the cylindrical body that coincides with the direction of the compression load, as the compression proceeds. The deviation of the compressive stress between the actual finite element model and the actual object can be corrected. As a result, an analysis result closer to the real part can be obtained.

Analysis method of the shock absorber according to the invention of claim 6 wherein the computer, finite elements of the shock absorber is created by the modeling method of the shock absorber according to any one of claims 1 to 5 and you have use the model, performing a structural analysis of the impact-absorbing member according to the finite element method.

In the analysis method of the shock absorber according to claim 6, a finite element model modeled by a shell element with a smaller number of cylindrical bodies than a real object is created, and a structural analysis by a finite element method is performed using this finite element model ( Strength, rigidity, stress, deformation, vibration, etc.). For this reason, compared with the case where the finite element model which has the same number of cylindrical bodies as the actual size is used, the analysis accuracy can be ensured while reducing the computer load. In addition, the analysis can be performed with higher accuracy than the analysis using the model modeled with solid elements.

As described above, in the shock absorber analysis method according to claim 6 , shock absorption is performed using the finite element model created by the shock absorber modeling method according to any one of claims 1 to 5. The body can be analyzed with high accuracy.
The shock absorber modeling device according to claim 7 is a case where a shock absorber configured by collecting cylindrical bodies having mutually parallel axes is subjected to a compressive load in the axial direction of the cylindrical body. , A shock absorber modeling apparatus for creating a finite element model for structural analysis by a finite element method, wherein the number of the cylindrical bodies is smaller than the number of the cylindrical bodies in the real object to be analyzed. On the basis of this, the compression stress in the case where a compressive load in the predetermined axial direction is applied to the aggregate of the virtual number of cylindrical bodies is compressed in the predetermined axial direction on the aggregate of the actual cylindrical bodies to be analyzed. Means for obtaining virtual material properties of the virtual number of cylindrical bodies so as to be equivalent to compressive stress when a load is applied; and the finite element model, the virtual number of cylindrical bodies having the virtual material properties As a collection of shell elements, It comprises means for Le of the.
The shock absorber modeling device according to claim 8 is a case where a shock absorber configured by collecting cylindrical bodies having mutually parallel axes is subjected to a compressive load in the axial direction of the cylindrical body. , A shock absorber modeling apparatus for creating a finite element model for structural analysis by a finite element method, wherein the number of the cylindrical bodies is smaller than the number of the cylindrical bodies in the real object to be analyzed. On the basis of this, the compression stress in the case where a compressive load in the predetermined axial direction is applied to the aggregate of the virtual number of cylindrical bodies is compressed in the predetermined axial direction on the aggregate of the actual cylindrical bodies to be analyzed. Means for obtaining virtual material properties of the virtual number of cylindrical bodies so as to be equivalent to compressive stress when a load is applied; and the finite element model, the virtual number of cylindrical bodies having the virtual material properties As a collection of shell elements, It comprises means for Le of the.
The shock absorber modeling device according to claim 9 is the shock absorber modeling device according to claim 7 or claim 8, wherein the virtual material property of the cylindrical body is calculated as the virtual material property. The thickness of the cylindrical wall of the cylindrical body is used.
The shock absorber modeling device according to claim 10 is the shock absorber modeling device according to any one of claims 7 to 9, wherein the means for modeling the finite element model with a shell element is provided. The cross-sectional shape along the plane perpendicular to the axis of the cylindrical body constituting the finite element model is the cross-sectional shape along the plane perpendicular to the axis of the cylindrical body constituting the real object to be analyzed. Use different shapes.
The shock absorber modeling apparatus according to claim 11 is the shock absorber modeling apparatus according to any one of claims 7 to 10, wherein the means for obtaining the virtual material property of the cylindrical body includes: The virtual material characteristics are made different at each part in the axial direction of the cylindrical body.
The shock absorber analysis device according to claim 12 is a shock absorber modeling device according to any one of claims 7 to 11 and a shock absorber finite element created by the modeling device. Means for analyzing the structure of the shock absorber by a finite element method using a model.

  As described above, the shock absorber modeling method according to the present invention is capable of creating a finite element model that can accurately analyze a shock absorber configured by collecting a plurality of cylindrical bodies. It has the effect. Further, the shock absorber analysis method according to the present invention has an excellent effect that the shock absorber can be analyzed with high accuracy using the finite element model created by the shock absorber modeling method.

  A collision analysis apparatus 10 to which a shock absorber modeling method according to an embodiment of the present invention is applied will be described with reference to FIGS. First, the moving barrier 12 for collision test (actual) to be analyzed and modeled will be described, then the modeling method for the moving barrier 12 will be described in detail, and then the collision analysis apparatus 10 will be described. . In addition, arrow FR, arrow UP, and arrow W appropriately shown in the drawings indicate the forward direction (traveling direction), the upward direction, and the vehicle width direction of the traveling carriage 14 provided with the moving barrier 12, respectively.

  FIG. 10 shows a perspective view of the traveling carriage 14 to which the actual 12A of the moving barrier 12 is attached. FIG. 11 shows a finite element model 12B of the moving barrier 12 in a perspective view. . As shown in these drawings, the moving barrier 12 is attached to the front end of the traveling carriage 14 and constitutes a collision portion with a collision target vehicle (actual vehicle) described later.

  The moving barrier 12 is formed such that a lower part 16 located at a position corresponding to the bumper height is projected forward of the upper part 18 in the shape of an actual vehicle. As shown in FIG. 11, the moving barrier 12 includes an aluminum honeycomb 20 as a main component of an impact absorber that has a honeycomb structure as shown in FIG.

  The aluminum honeycomb 20 is a structure in which a large number of regular hexagonal honeycomb lattices 22 each having a cylindrical body whose axial direction coincides with the front-rear direction (collision direction) are assembled in parallel. 22, each cylindrical wall is shared with different honeycomb lattices 22 adjacent to each other, thereby forming a hollow structure in which a space is formed only in each honeycomb lattice 22 with no gap between them. The aluminum honeycomb 20 is a structure that is deformed by a collision, and the moving barrier 12 is a deformable barrier.

  In the moving barrier 12, the rear ends of the aluminum honeycombs 20 constituting the lower part 16 and the upper part 18 are fixed to the rigid wall 26 of the traveling carriage 14, and the front ends of the aluminum honeycombs 20 constituting the lower part 16 and the upper part 18 are respectively covered. The panel 24 is covered and configured.

  Further, the aluminum honeycomb 20 constituting the lower portion 16 and the upper portion 18 of the moving barrier 12 has been pre-compressed before the collision test, and the rigidity on the front end side is low (soft) and the rigidity on the rear end side is high (hard). It is regarded as a characteristic. That is, the aluminum honeycomb 20 is configured to have different rigidity in each part in the main direction.

(Modeling of aluminum honeycomb)
A modeling method for creating a finite element model 20B for structural analysis by the finite element method of the actual aluminum honeycomb 20 constituting the moving barrier 12 will be described below.

  When the finite element model 20B of the aluminum honeycomb 20 as shown in FIG. 2 is created, first, the honeycomb lattice as shown in FIG. Consider an element expansion model 20C in which the virtual dimension b of 22 is larger than the dimension a of the honeycomb lattice 22 in the actual object 20A shown in FIG. In the element expansion model 20C, the imaginary number of the honeycomb lattices 22 is set to be smaller than the number of the actual honeycomb lattices 22A by expanding the honeycomb lattices 22 while maintaining the overall dimensions. In this embodiment, the dimension a of the honeycomb lattice 22 of the actual object 20A is 20 mm, whereas the virtual dimension b of the honeycomb lattice 22 in the element expansion model 20C is set to 35 mm.

  Therefore, since the virtual dimensions and virtual numbers of the honeycomb lattice 22 as the elements are different from the actual dimensions and numbers of the honeycomb lattice 22 as the element, the element expansion model 20C has the same material characteristics as those of the actual objects 20A. Cannot be reproduced. For this reason, the material characteristics used for the finite element model 20B of the aluminum honeycomb 20 are determined as virtual material characteristics in which the compression stress σ when a predetermined compressive load is applied is equal between the element expansion model 20C and the actual object 20A. A finite element model 20B is created (constructed) using the virtual material characteristics.

  In this embodiment, the virtual plate thickness tb of the cylindrical wall of the honeycomb lattice 22 is adopted as the virtual material characteristic. As shown in FIG. 3A, the strength Sp in the main direction (showing the finite element model 20B, see the arrow A shown in FIG. 2), which is the compression direction of the element expansion model 20C of the aluminum honeycomb 20, is It is almost directly proportional to the thickness t of the honeycomb lattice 22. On the other hand, the strength Ss in the lateral direction (see arrow B in FIG. 2), which is the compression direction of the element expansion model 20C, is insensitive to changes in the plate thickness compared to the main direction strength Sp, and the main direction strength Sp. In contrast, it is extremely small.

  Thus, the aluminum honeycomb 20 having a high main direction strength Sp and a low lateral direction strength Ss has a virtual plate thickness tb different from the plate thickness ta of the actual product 20A, as shown in FIGS. 3 (A) and 3 (B). By adopting, the main direction strength Sp of the element expansion model 20C is made to coincide with the main direction strength Spe (experimental value) of the actual product, and the lateral strength does not deviate from the lateral strength Sse (experimental value) of the actual product 20A. A model can be created.

  More specifically, using the test piece 20D of the aluminum honeycomb 20 in which the virtual plate thickness tb is constant in each part, a compressive load F is applied via the load transmission body 28 as shown in FIG. FIG. 4 is obtained by calculating and plotting the main-direction compressive stress σ (calculated value) for each virtual plate thickness tb while changing the virtual plate thickness tb. The main direction compressive stress σ is a stress obtained by σ = F / A, where A is the load support area of the test piece 20D (the area of the load transmitting body 28). An approximate expression representing the relationship between the plate thickness t and the main-direction compressive stress σ is obtained from a virtual straight line (curve) connecting the plots in FIG. An approximate expression when the virtual dimension of the honeycomb lattice 22 is 35 mm is shown as Expression (1).

            σ = 3.12t + 0.0232 (1)

  By substituting main-direction compressive stress σe, which is an experimental value of the actual product 20A of the aluminum honeycomb 20, into the approximate expression (1), the main-direction compressive stress σe equivalent to the main-direction compressive stress σe of the actual product 20A in the element expansion model 20C. A virtual thickness tb for obtaining the stress σ (≈σe) is obtained.

  In this embodiment, as described above, the virtual plate thickness tb is made different in the main direction in correspondence with the actual aluminum honeycomb 20 having different rigidity in each part in the main direction. Specifically, the virtual plate thickness tb at each position in the main direction corresponding to the actually measured part is obtained by substituting the main direction compressive stress σ actually measured at each part in the main direction of the actual object 20A into the approximate expression (1).

  Then, using the virtual dimension b and the virtual plate thickness tb of the honeycomb lattice 22 in the element expansion model 20C, as shown in FIG. 2, a finite element model 20B in which an aggregate of the honeycomb lattice 22 is modeled with shell elements is created. To do. That is, the finite element model 20B of the aluminum honeycomb 20 is modeled as a honeycomb structure which is an aggregate of honeycomb lattices 22 having a virtual dimension b larger than the dimension a of the honeycomb lattice 22 in the actual object 20A.

  12A and 12B show a schematic distribution of the virtual plate thickness tb of the finite element model 20B constituting the finite element model 12B of the moving barrier 12 (lower part 16, upper part 18). Yes. As can be seen from these figures, in the finite element model 20B, the virtual plate thickness tb is set small on the compression load input side (front side) in the main direction, and the virtual plate thickness tb is set large on the side opposite to the compression load input side. Yes.

(Example of verification of aluminum honeycomb finite element model)
FIG. 5 shows a finite element model 20B of an aluminum honeycomb 20 (a test piece that does not constitute the moving barrier 12, but is a model different from the test piece 20D in that the plate thickness distribution is set in the same manner as the moving barrier 12). It is an analysis result which shows a deformation | transformation state at the time of making predetermined compression load to. More specifically, FIG. 5 shows the finite state when a constant compressive load F is applied to each part of the finite element model 20B via the flat load transmitting body 28 as shown in FIG. The deformation shown in FIG. 6B of the element model 20B is shown with the load transmitting body 28 removed.

  From this figure, it can be seen that in the finite element model 20B, the deformation mode similar to that of the actual object 20A crushed from the end (tip) on the load input side is well reproduced. For example, in the solid model 20S in which the aluminum honeycomb 20 is modeled by solid elements as shown in FIG. 18, when the compressive load F is applied via the load transmitting body 28 as shown in FIG. As shown in FIG. 18 (B), only a deformation mode that collapses as a whole is provided, and the behavior of the aluminum honeycomb 20 cannot be reproduced (simulated). Thereby, the effectiveness of the finite element model 20B modeled by the shell element by the present shock absorber modeling method was confirmed.

  Further, in FIG. 7, a stress-strain diagram (experimental result) when a compressive load F is applied to the actual 20A of the aluminum honeycomb 20 via the load transmitting body 28 is shown by a thin line. A stress strain diagram (analysis result by the finite element method) in the case where the compressive load F is applied to 20B via the load transmitting body 28 is shown. FIG. 7A shows a stress strain diagram in the main direction, and FIG. 7B shows a stress strain diagram in the lateral direction. From these figures, it was confirmed that the analysis results of the stress against strain in both the main direction and the horizontal direction agree with the experimental results with high accuracy. In particular, the energy absorbed by the deformation of the aluminum honeycomb 20 (integrated value of each diagram) is in good agreement between the analysis results and the experimental results, and it was confirmed that the finite element model 20B is a model suitable for collision analysis. .

(Application to collision analysis equipment)
FIG. 8 shows the collision analysis device 10 in a block diagram. As shown in this figure, the collision analysis device 10 includes an input device 30 for inputting analysis target object data, boundary condition data, etc., and data for creating a finite element model 20B of the aluminum honeycomb 20 (actual object 20A). The modeled data storage device 32 configured by a memory storing the measured value of the main direction compressive stress σ in each part of the above and the approximate expression (1)), the finite element model 20B of the aluminum honeycomb 20, and the finite element model 20B is used to store an analysis device (strength, deformation, stress, etc.) of the aluminum honeycomb 20 calculated by the computer 34 and a calculation device 34 configured by a computer for performing a structural analysis of the aluminum honeycomb 20 by the finite element method. Using the analysis result storage device 36 composed of a memory for generating the finite element model 20B and the finite element model 20B created It is constituted by a display device 38 constituted by an LCD or the like for displaying the results of analysis by the finite element method.

  Next, a modeling / analysis routine executed by the computer-configured arithmetic unit 34 will be described with reference to a flowchart shown in FIG.

  In step S10, the collision mode to be analyzed is analyzed from the input device 30 such as the overall dimensions of the aluminum honeycomb 20 constituting the lower portion 16 and the upper portion 18 of the moving barrier 12, the virtual dimension b of the honeycomb lattice 22 used to create the finite element model 20B, (Full wrap, offset, side collision, etc.), the speed just before the collision is input. In step S12, a virtual plate thickness tb used for the finite element model 20B is determined based on the virtual dimension b input in step S10. For example, the virtual plate thickness tb is determined by referring to the main direction compressive stress σe in each main direction portion of the actual object 20A stored in the modeled data storage device 32 for each position in the main direction, and compressing this in the main direction. The calculation is performed by substituting directly into an approximate expression representing the relationship between the stress σ and the virtual plate thickness tb (in the present embodiment where the virtual dimension b is 35 mm, the approximate expression (1)).

  In step S14, a finite element model in which a honeycomb structure (aluminum honeycomb 20) is modeled by shell elements using the virtual dimension b input in step S10 and the virtual plate thickness tb of each part in the main direction obtained in step S12. 20B is created. Here, as shown in FIG. 11, the finite element model 12B of the moving barrier 12 having the overall dimensions of the actual object 12A of the moving barrier 12 input in step S10 and having the plate thickness distribution shown in FIG. 12 is created. .

  Next, the process proceeds to step S16. In step S16, a finite element model 40B simulating the actual vehicle 40A of the collision target vehicle 40 (after the collision, see FIG. 15A) is created based on the input data. . Then, the process proceeds to step S18, and these structures after the finite element model 12B of the moving barrier 12 (the finite element model 20B of the aluminum honeycomb 20) is collided with the finite element model 40B of the collision target vehicle 40 are analyzed.

  Hereinafter, an example in which the analysis is performed when the moving barrier 12 attached to the tip of the traveling carriage 14 is caused to collide with the side surface of the collision target vehicle 40 will be described.

  FIG. 15A is a perspective view showing a deformed state after the actual object 12A of the moving barrier 12 collides with the actual object 40A of the colliding vehicle 40 in a side collision, and FIG. 15B shows the collided vehicle after the collision. It is a perspective view which shows the deformation | transformation state of 40 real thing 40A.

  From FIG. 15 (A), a concave pillar collision trace 42 along the vertical direction of the vehicle body is formed at the center in the vehicle width direction of the actual object 12A of the moving barrier 12 due to the collision with the center pillar of the vehicle 40 to be collided. I understand. Further, it can be seen that a concave door beam collision trace 44 is formed on the actual 12A of the moving barrier 12 along the vehicle width direction due to a collision with the door impact beam of the colliding vehicle 40. In this actual vehicle test, mainly three door beam collision traces 44 are formed. On the other hand, from FIG. 15B, it can be seen that in the actual object 40A of the collision target vehicle 40, the door panel 46 is greatly recessed due to the collision, and the lower portion of the door panel 46 is turned up.

  FIG. 13A shows the finite element model 12B of the moving barrier 12 after the collision of the colliding vehicle 40 with the finite element model 40B. From this figure, it can be seen that in the finite element model 12B, the pillar collision trace 42 is formed and is deformed close to the actual object 12A due to the collision with the center pillar. Similarly, in the finite element model 12B, it can be seen that door beam collision traces 44 are formed at three locations corresponding to the actual object 12A, and the deformation is close to the actual object 12A due to the collision with each door impact beam.

  FIG. 13B shows an analysis result by a finite element method using the solid model 12S of the moving barrier 12 as a comparative example. In the analysis using the solid model 12S, the pillar collision trace 42 and the door beam collision trace 44 are both smaller than the pillar collision trace 42 and the door beam collision trace 44 of the actual object 12A, and the deformation behavior of the moving barrier 12 cannot be simulated.

  On the other hand, FIG. 14 (A) shows a finite element model 40B of the collision target vehicle 40 after the finite element model 12B of the moving barrier 12 collides. From this figure, it can be seen that in the finite element model 40B in which the finite element model 12B collides, the dent of the door panel 46 is reproduced well, and in particular, the turning of the lower part of the door panel 46 is reproduced well. Furthermore, as can be seen from a comparison with FIG. 15B, the deformation of the rocker 48 constituting the lower skeleton of the vehicle body inward of the vehicle body is also well reproduced.

  FIG. 14B shows, as a comparative example, a finite element model 40B of the collision target vehicle 40 after the solid model 12S collides. In this comparative example, although the dent of the door panel 46 of the finite element model 40B is reproduced to some extent, the turning of the lower part of the door panel 46 is not reproduced, and it can be seen that the analysis accuracy is low. Further, in this comparative example, the deformation of the rocker 48 is not reproduced.

(Operational effect of this embodiment)
Here, in the shock absorber modeling method according to the present embodiment, the finite element model 20B of the aluminum honeycomb 20 is formed as an aggregate of honeycomb lattices 22 having a virtual dimension b larger than the dimension a of the actual object 20A. Therefore, the number of elements of the finite element model 20B can be reduced while adopting the honeycomb structure. For this reason, since the number of honeycomb lattices 22 is reduced as compared with the case of creating a finite element model using the dimension a and the plate thickness ta of the honeycomb lattice 22 in the actual object 20A as they are, the analysis by the finite element method is performed. The load on the arithmetic unit 34 can be significantly reduced. Further, the load on the arithmetic unit 34 at the time of modeling (element division) in step S14 can be reduced.

  In the shock absorber modeling method according to the present embodiment, the finite element model 20B having the honeycomb lattice 22 larger than the actual 20A has the same main direction compressive stress when receiving the same compressive load as the actual 20A. Since it is modeled using the virtual plate thickness tb, it is prevented that the analysis accuracy when analyzed by the finite element method is impaired due to the enlargement of the honeycomb lattice 22. Therefore, the finite element model 20B created by this modeling method has equivalent analysis accuracy compared to the case of creating a finite element model using the dimension a and the plate thickness ta of the honeycomb lattice 22 in the actual object 12A as they are. Obtainable.

  In this embodiment, a simple approximate expression such as the approximate expression (1) described above is used in order to make the main direction compressive stress σ of the finite element model 20B substantially coincide with the real main direction compressive stress by the virtual plate thickness tb. Thus, a setting for maintaining the analysis accuracy can be obtained. In particular, since the finite element model 20B is a model in which the virtual thickness tb is different in each part in the main direction, in other words, since the compression stress of each part in the main direction of the real object 20A is set to be reproduced, It was possible to reproduce the deformation with higher accuracy and to improve the analysis accuracy without increasing the number of elements of the finite element model.

  Furthermore, the finite element model 20B created by the shock absorber modeling method according to the present embodiment, that is, the moving barrier 12 performed using the finite element model 12B of the moving barrier 12, and the vehicle to be collided with the moving barrier 12 are collided. In the analysis by 40 finite element method, as described above with reference to FIGS. 13 to 15, an actual vehicle collision test can be reproduced with high accuracy.

  Therefore, by modeling the aluminum honeycomb 20 by the modeling method according to this embodiment, the load on the arithmetic unit 34 is reduced compared with a case where the actual 20A of the aluminum honeycomb 20 is modeled as it is, and practical calculation is performed. The vehicle collision test can be reproduced with speed and calculation cost and with significantly better analysis accuracy compared to the case where the analysis is performed using the solid models 12S and 20S.

(Another example of virtual material properties)
In the above embodiment, the virtual plate thickness tb is used as the virtual material characteristic. However, the main direction compressive yield stress of the actual 20A of the aluminum honeycomb 20 and the finite element model 20B may be matched using other material characteristics. An example will be described.

  FIG. 16A shows a stress strain diagram of an aluminum alloy that is a constituent material of the aluminum honeycomb 20. In this figure, the 1.0-fold diagram indicated by the solid line is a stress-strain diagram obtained by experiments, and the other diagrams are obtained by multiplying the stress-strain diagram of the experimental values by a predetermined stress magnification n. Is.

  Then, by using a test piece 20D having virtual material characteristics given by multiplying this stress magnification n and having a constant plate thickness t of each part, a compressive load is applied via a load transmitting body 28 as shown in FIG. When the principal direction compressive stress σ (calculated value) when each part is uniformly compressed and the test piece 20D yields is plotted with respect to each stress magnification n, FIG. 16B is obtained. A correlation equation representing the relationship between the stress magnification n, which is a virtual material characteristic, and the principal direction compressive stress σ (yield stress) is obtained from a curve smoothly connecting the plots. When the plate thickness t of the honeycomb lattice 22 is 0.1 mm, the correlation equation is shown as equation (2).

σ = −0.0736n ² + 0.3349n (2)

  By substituting the main direction compressive stress σe, which is an experimental value of the actual product 20A of the aluminum honeycomb 20, into this correlation equation (2), a virtual material that obtains the main direction compressive stress σe equivalent to the actual product 20A in the element expansion model 20C. The characteristic stress magnification n is determined. Therefore, as in the above-described embodiment, the stress magnification n obtained by substituting the main direction compressive stress of the actual object 20A in each part in the main direction is used for modeling as a virtual material characteristic at a corresponding position in the main direction. Thus, a finite element model 20B having a distribution of material properties (rigidity) as shown in FIG. 12 in the main direction can be created.

  Thus, even if the stress magnification n is used as the virtual material characteristic instead of the virtual plate thickness tb, the virtual dimension b of the honeycomb lattice 22 in the finite element model 20B is set to the dimension a of the actual object 20A, as in the above embodiment. The collision test can be accurately reproduced by setting the main direction compressive stress equivalent to that of the actual object 20A while reducing the number of divided elements.

(Modified example of honeycomb lattice shape)
In the above embodiment, the finite element model 20B (element expansion model 20C) is modeled with the honeycomb lattice 22 having a shape similar to the honeycomb lattice 22 of the actual 20A in front view (cross-sectional view perpendicular to the main direction). However, the present invention is not limited to this. For example, as shown in FIG. 17A, the regular hexagonal honeycomb lattice 22 of the actual object 20A is stretched only in one direction or in two orthogonal directions. The finite element model 20B may be created by using the flat honeycomb lattice 50 formed in a long hexagonal shape by changing the enlargement ratio. As shown in FIG. Alternatively, the finite element model 20B may be created using a rectangular grid 52 having a different shape. In particular, in the latter case, it becomes easier to create the finite element model 20B. (The modeling program of the arithmetic unit 34 is simplified). Note that other shapes such as a pentagonal shape, a heptagonal shape, and an octagonal shape can also be adopted as the lattice model.

  In the above-described embodiment and the modification, the example in which the virtual material characteristics (virtual plate thickness tb, stress magnification n) of the finite element model 20B are different in each part in the main direction corresponding to the actual object 20A is shown. However, the present invention is not limited to this, and a finite element model may be created with the virtual material characteristics being constant in each part in the main direction as appropriate depending on the shape of the analysis target and the analysis application.

  Further, in the above-described embodiment, the example in which the moving barrier 12 for vehicle collision test is the object to be modeled and analyzed is shown, but the present invention is not limited to this and is configured as an aggregate of a large number of cylindrical bodies. The present invention can be applied to modeling of shock absorbers of various structures that are compressed in the axial direction of the cylindrical body, for example, modeling of components of aircraft and high-speed railways, and analysis by the finite element method.

(A) is a front view showing an actual object to be modeled by the shock absorber modeling method according to the embodiment of the present invention, and (B) is created by the shock absorber modeling method according to the embodiment of the present invention. It is a front view for expressing the virtual dimension of a finite element model. It is a figure which shows the finite element model created with the modeling method of the shock absorber which concerns on embodiment of this invention. It is a diagram which shows the relationship between the board thickness and the intensity | strength of the aluminum honeycomb which is a modeling object by the modeling method of the shock absorber which concerns on embodiment of this invention, Comprising: (A) shows main direction intensity | strength, (B ) Indicates lateral strength. It is a figure which shows the relationship between the plate | board thickness of the aluminum honeycomb which is a modeling object by the modeling method of the shock absorber which concerns on embodiment of this invention, and main direction compressive stress. It is a figure which shows the shape after the deformation | transformation of the finite element model created with the modeling method of the shock absorber which concerns on embodiment of this invention. It is a figure which shows the finite element model created with the modeling method of the shock absorber which concerns on embodiment of this invention, Comprising: (A) is a figure which shows the shape before compression, (B) shows the shape after compression. FIG. (A) is a stress strain diagram in the main direction of a finite element model created by the shock absorber modeling method according to the embodiment of the present invention, and (B) is a model of the shock absorber according to the embodiment of the present invention. It is a stress-strain diagram of the horizontal direction of the finite element model created by the optimization method. 1 is a block diagram illustrating a schematic configuration of an impact analysis apparatus to which a shock absorber modeling method according to an embodiment of the present invention is applied. It is a flowchart which shows the modeling and analysis routine performed by the arithmetic unit of the impact analyzer to which the shock absorber modeling method according to the embodiment of the present invention is applied. It is a perspective view which shows the moving barrier of the analysis object by the impact-analysis apparatus to which the modeling method of the shock absorber which concerns on embodiment of this invention was applied. It is a figure which shows the finite element model of the moving barrier modeled with the modeling method of the shock absorber which concerns on embodiment of this invention. It is a figure which shows distribution of the virtual board thickness of the moving barrier modeled with the modeling method of the shock absorber which concerns on embodiment of this invention, Comprising: (A) is a top view, (B) is a side view. (A) is a figure which shows the deformation | transformation state of the moving barrier analyzed by the finite element method using the finite element model modeled by the modeling method of the shock absorber which concerns on embodiment of this invention, (B) is a comparative example It is a figure which shows the analysis result of the solid model which is. (A) is a figure which shows the deformation | transformation state analyzed by the finite element method of the to-be-collised vehicle which is the collision symmetry of the moving barrier modeled with the modeling method of the shock absorber which concerns on embodiment of this invention, (B). It is a figure which shows the analysis result of the colliding vehicle which the solid model which is a comparative example collided. (A) is a perspective view showing a shape after a collision test of a moving barrier object to be analyzed by an impact analyzer to which the shock absorber modeling method according to the embodiment of the present invention is applied, and (B) is a moving barrier object. FIG. (A) is a stress-strain diagram showing another example of virtual material characteristics used in the method of modeling the shock absorber according to the embodiment of the present invention, and (B) is a virtual material characteristic and main direction compressive stress according to another example. It is a figure which shows the relationship. FIG. 8 is a diagram showing a modification of the shape of the honeycomb lattice used in the shock absorber modeling method according to the embodiment of the present invention, where (A) is a front view of the first modification, and (B) is a second modification. FIG. It is a figure which shows the solid model which is a comparative example with the finite element model which concerns on embodiment of this invention, Comprising: (A) is a figure which shows the shape before compression, (B) is a figure which shows the shape after compression. .

Explanation of symbols

12 Moving barrier (shock absorber)
12A Real 12B Finite element model 20 Aluminum honeycomb (Shock absorber)
20A Real 20B Finite element model 22 Honeycomb lattice (tubular body)
b Virtual dimension tb Virtual plate thickness

Claims (12)

Creates a finite element model for structural analysis by the finite element method when a shock absorber composed of a collection of cylindrical bodies having mutually parallel axes is subjected to a compressive load in the axial direction of the cylindrical body. A shock absorber modeling method performed by a computer for
Based on the virtual dimension of the cylindrical body that is larger than the dimension of the cylindrical body in the real object to be analyzed, a compressive load in a predetermined axial direction is applied to the cylindrical body or the aggregate of the cylindrical bodies. The virtual dimension so that the compressive stress in the case of the analysis is equivalent to the compressive stress in the case where the compressive load in the predetermined axial direction is applied to the real cylindrical body or the aggregate of the cylindrical bodies to be analyzed. The virtual material characteristics of the cylindrical body of
A shock absorber modeling method in which the finite element model is modeled by a shell element as an aggregate of the cylindrical bodies having the virtual material characteristics and the virtual dimensions. Creates a finite element model for structural analysis by the finite element method when a shock absorber composed of a collection of cylindrical bodies having mutually parallel axes is subjected to a compressive load in the axial direction of the cylindrical body. A shock absorber modeling method performed by a computer for
Compressive stress in the case where a predetermined axial load is applied to an assembly of cylindrical bodies based on the virtual number of the cylindrical bodies that is smaller than the number of cylindrical bodies in the real object to be analyzed The virtual material characteristic of the virtual number of cylindrical bodies is obtained so that the compression stress is equivalent to the compressive stress when the predetermined axial load is applied to the assembly of real cylindrical bodies to be analyzed. ,
A shock absorber modeling method in which the finite element model is modeled by shell elements as an aggregate of the virtual number of cylindrical bodies having the virtual material characteristics. The virtual as the material properties, the modeling method of the tubular body according to claim thickness of the cylindrical wall Ru with 1 or claim 2 shock absorber according. The cross-sectional shape along the plane perpendicular to the axis of the cylindrical body constituting the finite element model is different from the cross-sectional shape along the plane perpendicular to the axis of the cylindrical body constituting the real object to be analyzed. modeling method according to claim 1 shock absorber according to any one of claims 3 to shape. The virtual material properties, the modeling method of the tubular body shock absorber according to any one of axial claims Ru varied among the respective units 1 to claim 4. Computer, and have use a finite element model of the shock absorbing body created by the modeling method of the shock absorber according to any one of claims 1 to 5, structural analysis of the impact-absorbing member by the Finite Element Method Analysis method of shock absorber. Creates a finite element model for structural analysis by the finite element method when a shock absorber composed of a collection of cylindrical bodies having mutually parallel axes is subjected to a compressive load in the axial direction of the cylindrical body. A shock absorber modeling device for
  Based on the virtual dimension of the cylindrical body that is larger than the dimension of the cylindrical body in the real object to be analyzed, a compressive load in a predetermined axial direction is applied to the cylindrical body or the aggregate of the cylindrical bodies. The virtual dimension so that the compressive stress in the case of the analysis is equivalent to the compressive stress in the case where the compressive load in the predetermined axial direction is applied to the real cylindrical body or the aggregate of the cylindrical bodies to be analyzed. Means for determining the virtual material properties of the cylindrical body;
  Means for modeling the finite element model with shell elements as an aggregate of the cylindrical bodies having the virtual material properties and the virtual dimensions;
  Shock absorber modeling device with Creates a finite element model for structural analysis by the finite element method when a shock absorber composed of a collection of cylindrical bodies having mutually parallel axes is subjected to a compressive load in the axial direction of the cylindrical body. A shock absorber modeling device for
Compressive stress in the case where a predetermined axial load is applied to an assembly of cylindrical bodies based on the virtual number of the cylindrical bodies that is smaller than the number of cylindrical bodies in the real object to be analyzed The virtual material characteristics of the virtual number of cylindrical bodies are obtained so that the compression stress is equivalent to the compressive stress when the predetermined axial load is applied to the aggregate of real cylindrical bodies to be analyzed. Means,
Means for modeling the finite element model with shell elements as an aggregate of the virtual number of cylindrical bodies having the virtual material properties;
Shock absorber modeling device with 9. The impact absorber modeling apparatus according to claim 7, wherein the means for obtaining the virtual material characteristic of the cylindrical body uses a thickness of a cylindrical wall of the cylindrical body as the virtual material characteristic. The means for modeling the finite element model with a shell element has a cross-sectional shape along a plane orthogonal to an axis in the cylindrical body constituting the finite element model, and the cylindrical body constituting the real object to be analyzed. The shock absorber modeling device according to any one of claims 7 to 9, wherein the shock absorber has a shape different from a cross-sectional shape along a plane perpendicular to the axis of the shaft. The model of the shock absorber according to any one of claims 7 to 10, wherein the means for obtaining the virtual material characteristic of the cylindrical body varies the virtual material characteristic at each portion in the axial direction of the cylindrical body. Device. The shock absorber modeling device according to any one of claims 7 to 11,
Means for performing a structural analysis of the shock absorber by a finite element method using a finite element model of the shock absorber created by the modeling device;
Shock absorber analysis device with

Images