JP2011251318A - Forming analysis method for press component - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a forming analysis method for a press component, with which a calculation time can be shortened as far as possible for a press component having a shape with locally different thicknesses at a flange part, such as a bent hut-shape material, for which press simulation is difficult from the view point of dimensional precision, and with which high forming analysis precision can be obtained.SOLUTION: In the simulation using a finite element method for evaluating formability of the press component, a die is modeled as an elastic solid model having the predetermined thickness from the contact surface of a blank. When the simulation is executed, the entire surface of a die model on a side opposite to a surface in contact with a blank is displaced uniformly.

Description

  The present invention relates to a molding analysis method for a pressed part, which is performed to evaluate the moldability of a pressed part such as an automobile.

  In order to evaluate the formability of press parts such as automobiles, forming analysis by simulation using a finite element method is widely performed. In modeling during execution of such a simulation, it has been common for a mold to model only a surface that contacts a blank (material) as a rigid body of a shell element (see, for example, Non-Patent Document 1). It was.

  The modeling cost (modeling time) and the calculation cost (calculation) are such that if the entire mold is modeled as an elastic body, it takes time and effort to model, and further computation time is required to perform the simulation. This is because it is not practical in terms of both time and a reason that almost sufficient formability evaluation can be obtained by comparing the forming analysis result by the simulation with the above modeling and the actual press.

  However, in the method represented by Non-Patent Document 1, the method of modeling only the surface in contact with the blank as a rigid body by the shell element is intended for shapes that are difficult to press-simulate from the viewpoint of dimensional accuracy, such as a bent hat material. In some cases, the material inflow did not match between the actual press and the finite element analysis, and for this reason, the springback analysis and the like did not match between the experiment and the analysis.

  On the other hand, Non-Patent Document 2 discloses a method of separately calculating the deflection of a mold for a bent hat material, modeling the entire mold as an elastic body in detail, and easily incorporating it into the finite element method calculation. Has been proposed.

Fukiharu Hiroshi, Nikkan Kogyo Shimbun, "Press-forming simulation", p12 (2003-5) Hideo Sasamori et al., Plasticity and Processing, Vol.44, No.513 (2003-10) p28-32

  However, since the method disclosed in Non-Patent Document 2 models the entire mold as an elastic body in detail, the above-described problem that modeling and simulation calculation takes a lot of time and is not practical. There is. Furthermore, there is a problem in that satisfactory analysis accuracy cannot be obtained with respect to accuracy because the deflection and load distribution of the mold are not coupled analysis (simultaneous analysis).

  The present invention has been made in view of such circumstances, and it is difficult to perform press simulation from the viewpoint of dimensional accuracy such as a bent hat material, and the calculation time is reduced for a shape with locally increasing or decreasing thickness at the flange portion. It is an object of the present invention to provide a forming analysis method for press parts that is shortened as much as possible and has high forming analysis accuracy.

  The above problems are solved by the following invention.

  [1] A press part forming analysis method by simulation using a finite element method for evaluating the formability of a press part, in which a mold is modeled as an elastic solid model having a predetermined thickness from a blank contact surface When performing simulation, a molding analysis method for a pressed part, characterized in that the entire mold model surface opposite to the blank contact surface is uniformly displaced.

  [2] In the forming analysis method for a pressed part according to [1] above, the die to be modeled is a die and / or holder, and the Young's modulus and / or the thickness of the die model is adjusted when executing the simulation. A method for forming and analyzing a pressed part.

  [3] A forming analysis method for a pressed part according to [1] or [2], wherein the blank is modeled as a solid model of an elastic-plastic material.

  According to the present invention, a press simulation is difficult from the viewpoint of dimensional accuracy such as a bent hat material, and in a forming analysis of a shape with local increase / decrease in thickness at the flange portion, a highly accurate result that closely matches the material inflow in the actual press Can be obtained. Also, using this result, it is possible to improve the prediction accuracy of plate thickness reduction, cracks, wrinkles and springback.

It is a figure which shows typically the metal mold | die model used in the Example. It is a figure which shows the flange remaining amount of the outer side (hat outer side) of the center part of simulation and an actual press. It is a figure which shows the flange remaining amount inside the center part of simulation and a real press (hat inner side). It is a figure which shows the comparison of simulation time.

  Prior to the description of the present invention, first, the knowledge that led to the present invention will be described below. As a result of investigating the die end face for the elastic deflection of the entire press die, the inventors have found that the die is bent in the vertical method only near the bottom dead center of the press (about 0 to 2 mm). I grasped that it was not bent. From this, it was found that the large deflection of the entire mold does not affect the material inflow.

  On the other hand, in the bent hat molding, it was confirmed that the plate thickness increased outside the bent portion and the plate thickness decreased inside during the material inflow process. From this, it was found that changes in the contact pressure distribution, such as the contact pressure increasing on the outside where the plate thickness increased and conversely decreasing on the inside, affected the material inflow.

  Therefore, the present inventors consider that the mold surface is locally elastically deformed due to the fluctuation of the contact pressure, and it is not necessary to model the entire mold in detail. It is thought that modeling is important, and the present invention has been conceived.

  As described in Non-Patent Document 1, in the present invention, the entire mold is defined as a rigid body. In the present invention, a predetermined thickness (for example, equivalent to an actual press) from the mold surface where the mold contacts the blank (material) is used. By defining (modeling) an elastic body having a thickness of 100 mm to 200 mm or the like, the elastic deformation of the mold surface in contact with the blank (material) is expressed. In other words, the frictional force, which is the inflow resistance of the material, is relaxed in the portion where the plate thickness is increased, and the frictional force is increased in the portion where the plate thickness is decreased. Can express the phenomenon. For this reason, the material inflow in the actual press matches the simulation result.

  By forming a solid model of both the die and the holder in the mold, it is possible to express the surface elastic deformation of the mold in contact with both the front and back surfaces of the blank. Further, by forming only one of the die and the holder as a solid model, it is possible to express surface elastic deformation due to pressure only by a mold on one side that contacts the blank.

  Further, in executing the simulation, a result close to an actual press can be obtained by adjusting the Young's modulus and the thickness of a mold model defined as an elastic body. This is because Young's modulus and thickness have a substantially linear relationship with surface contact deformation.

  For example, if the die thickness of the actual press die and the holder is 150 mm, if only the holder side is a solid element in the simulation, the Young's modulus is 1/2 or the surface elastic deformation is 300 mm. Are almost equal and a good inflow result can be obtained.

  In addition, by applying an elasto-plastic solid element to the blank, the blank will be deformed by receiving contact force (stress in the thickness direction), which reduces the abrupt change in contact surface pressure as in actual press. Thus, the calculation accuracy of the inflow amount is improved. When the solid element is applied to the blank, the inflow is improved compared to the case where the shell element is applied. The shell element does not deform directly due to the contact pressure, but the solid element deforms. This is because it is considered to be sensitive to.

  Examples in which the effectiveness of the present invention was examined by press molding simulation and actual press experiment of a bent hat material will be described below.

  FIG. 1 is a diagram schematically illustrating a mold model used in the example. In the figure, 1 represents a blank, 2 represents a die, 3 represents a holder, and 4 represents a punch. 1 (a) shows a solid model of both the holder and the die, and FIG. 1 (b) shows only the die as a solid model. FIG. 1 (c) shows only the holder as a solid model. 1 (d) is a thickened holder of FIG. 1 (c), and FIG. 1 (e) is a shell rigid body model, that is, a conventional rigid body is defined as a rigid body. Each shell model is shown.

  The boundary conditions in the molding simulation will be described. The mold operation in the draw molding of the rigid mold model (FIG. 1 (e)) with the conventional shell element is to fix the punch 4 and to apply the wrinkle holding force in the anti-press direction to the holder 3, and to the center of the rigid body of the die 2 Was moved in the pressing direction. In the present invention, the entire surface opposite to the surface in contact with the blank 1 of the mold modeled as an elastic body is uniformly displaced. This is to prevent the elastic deformation of the mold surface coming into contact with the blank from affecting the opposite side surface.

  That is, when the die 2 is made elastic (FIGS. 1A and 1B), the nodes on the upper surface side are uniformly displaced. When the holder 3 is made elastic (FIGS. 1 (a), (c), (d)), the nodes on the lower surface are uniformly displaced, and the force applied to this surface becomes the holder force. did. In executing the simulation, LS-DYNA manufactured by LSTC was used.

  2 and 3 are diagrams showing the remaining flange amount on the outside and inside (hat outside and inside) of the central part of the simulation and actual press. The actual press, Comparative Example 1, and Examples 1 to 6 are shown side by side from the left on the vertical axis and the horizontal axis on the remaining flange amount. In the actual press, the measured values in the actual press are compared, in Comparative Example 1, the result of simulation with the shell rigid body model in FIG. 1 (e), and in Examples 1-4, the gold in FIGS. 1 (a)-(d), respectively. The simulation result using a mold model with the Young's modulus of the mold set to 205.8 GPa is shown.

  Further, Example 5 has a Young's modulus of 102.9 GPa, which is half the former in the mold model of FIG. 1 (c), and Example 6 is a blank using the mold model of FIG. 1 (a). The simulation results are shown as solid elements of elasto-plastic bodies.

  The molding conditions of the simulation and the actual press are as follows: the blank is JSC590Y with a plate thickness of 1.2 mm, and the wrinkle pressing force is 686 KN. The friction coefficient in the simulation is 0.12, and the punch is a rigid shell element.

  FIG. 4 is a diagram showing a comparison of simulation times. For Comparative Example 1 and Examples 1 to 6, the time required for the simulation is shown. Since the simulation time varies depending on the performance of the computer used, it is a result of using the same computer, and each of them is made dimensionless by the time required for the simulation of Comparative Example 1.

  Example 6 (104 times), Example 1 (20 times), Example 4 (11 times), Example 3 (9 times), Example in order of the time required for the simulation of Comparative Example 1 5 (6 times) and Example 2 (5 times).

  First, comparing the holder with a solid model of the elastic body (Examples 3 to 5) and the model of the die with a solid model of the elastic body (Example 2), the holder is elastic because of the problem of load convergence. The solid model (Examples 3 to 5) takes longer calculation time.

  Also, in the comparison of Examples 3 to 5, Example 5 has a Young's modulus that is half that of Example 3, so that the time step (calculation time gap in the dynamic explicit method) can be increased, so the time is short. It has become. Similarly, since the holder of Example 4 is thicker than that of Example 3, the number of meshes is increased and the calculation time is increased.

  Furthermore, in Example 6, since the blank is also an elastic-plastic solid element, the calculation time is enormous due to the increase in the number of meshes and the decrease in the time size due to the decrease in the mesh size in the plate thickness direction. (104 times that of Comparative Example 1).

  From the viewpoint of accuracy, when comparing the actual press with Comparative Example 1, the flange remaining amount on the outer side of the hat is 48 mm in the actual press, but 56 mm in Comparative Example 1 is about 8 mm larger than the actual press (FIG. 2). I understand that. On the other hand, it can be seen that the flange remaining amount inside the hat shown in FIG. 3 is 34 mm in the actual press, but is 26 mm in the first comparative example, which is 8 mm shorter than the actual press. As described above, there is a large difference in the remaining flange amount between the actual press and the comparative example 1, and the problem of using the shell rigid body model for the press forming analysis of the bent hat material is reconfirmed.

  About Examples 1-6, although the flange remaining amount of a hat outer side is large compared with an actual press, and the flange remaining amount of a hat inner side is a small simulation result as a whole tendency, it is comparative example 1 and In comparison, it can be seen that the difference with the actual press is reduced and the simulation results are improved.

  Looking at the degree of agreement with the actual press, the best example is Example 6 (both holder and die are solid models of elastic bodies + blanks are also solid models of elasto-plastic bodies), then Example 1 (holders) Both dies are solid models of elastic bodies). This is a result as originally expected that the calculation time is very long but the accuracy is improved because more detailed modeling is performed as compared with other examples 2 to 5.

  However, even when only the die (Example 2) and only the holder (Example 3) are used as elastic bodies, the material inflow (flange remaining amount) is improved, and both are greatly different from the case of modeling (Example 1). It does not occur and is considered a practical effective means from the viewpoint of shortening the calculation time. If only the holder (Example 3, FIG. 1 (c)) is an elastic body, as described above, the calculation time is longer than the case of only the die due to the problem of load convergence. 1 (b)) is considered to be more practical if it is made elastic.

  Further, for Example 4 (thickening the holder, FIG. 1 (d)) and Example 5 (Having a Young's modulus of 102.9 GPa in half), further improvements were observed, respectively. By adjusting the Young's modulus and thickness (distance to the surface in the press direction) of the mold model defined as, a more accurate inflow amount close to the experimental result can be simulated.

  In this example, examination was made from both aspects of accuracy and calculation time, but there was also a trade-off relationship between the two, and in order to actually perform a press simulation, modeling and simulation execution corresponding to the required accuracy were required. is there.

  As described above, if the molding analysis method for a pressed part according to the present invention is used, it is difficult to perform a press simulation from the viewpoint of dimensional accuracy such as a bent hat material. Highly accurate results can be obtained that are in good agreement with material inflow. Furthermore, using this result, it is possible to improve the prediction accuracy of plate thickness reduction, cracks, wrinkles, and springback.

1 Blank 2 Die 3 Holder 4 Punch

Claims (3)

A press part molding analysis method by simulation using a finite element method for evaluating the formability of a press part,
The mold is modeled as an elastic solid model having a predetermined thickness from the blank contact surface,
A molding analysis method for a pressed part, wherein the entire die model surface opposite to the blank contact surface is uniformly displaced when executing the simulation. In the shaping | molding analysis method of the press parts of Claim 1,
The die to be modeled is a die and / or holder,
A molding analysis method for a pressed part, wherein the Young's modulus and / or the thickness of the mold model is adjusted when executing the simulation. In the molding analysis method for a pressed part according to claim 1 or 2,
A method for forming and analyzing a pressed part, wherein the blank is modeled as a solid model of an elastoplastic material.