Method for predicting press stiffness

The method predicts press machine rigidity by measuring load and displacement, creating an analytical model, and deriving stiffness coefficients, addressing the limitations of conventional methods with improved accuracy and efficiency.

JP7590646B2Active Publication Date: 2024-11-27NIPPON STEEL CORPORATION
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2020216687
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-12-25
Publication Date
2024-11-27
Estimated Expiration
2040-12-25

AI Technical Summary

Technical Problem

Conventional methods for predicting press machine rigidity require actual measurement of each press mold, are labor-intensive, and do not account for continuous rigidity distribution along the surface direction.

Method used

A method involving actual measurement of load and displacement at multiple points on a press machine, creation of an analytical model reflecting the machine's dimensions, and derivation of a stiffness coefficient using a two-dimensional Fourier series expansion to predict rigidity, allowing for continuous stiffness distribution.

Benefits of technology

Enables accurate prediction of press machine rigidity with reduced calculation effort and minimal deviation from actual values, facilitating high-precision press forming by anticipating deformation during mold design.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007590646000019
    Figure 0007590646000019
  • Figure 0007590646000020
    Figure 0007590646000020
  • Figure 0007590646000021
    Figure 0007590646000021
Patent Text Reader

Abstract

To provide a novel prediction method of a press machine.SOLUTION: A press machine rigidity prediction method comprises: an actual measurement step S1 for applying a load to a press machine through a representative metal mold, for actually measuring a relationship between the load and an actual displacement on a plurality of measurement points on the press machine; an analysis model creating step S2 for preparing a press machine model in which a long side length and a short side length of the press machine are reflected; a theory computational model creating step S3 for expressing a relationship between bending of the press machine model and the load, by a function using a rigidity coefficient indicating rigidity of the press machine; and a rigidity coefficient deriving step S4 for, on the basis of the actual displacement and the load, deriving the rigidity coefficient.SELECTED DRAWING: Figure 4
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a method for predicting the stiffness of a press machine. [Background technology]

[0002] Conventionally, in order to reflect the expected deformation of the press die during workpiece molding in the design of the press die, there has been a method in which local deformation of the press die attached to a press machine during workpiece molding is actually measured and the rigidity of a press molding device consisting of a press machine and a press die is predicted. However, the conventional method is a method of measuring the local deformation of a specific press die when the specific press die is used. Therefore, even if the press machine is the same, when press dies of different shapes are used, in order to predict the rigidity of the press machine, it was necessary to actually measure the deformation of each different press die. Furthermore, conventionally, there has been no method for predicting continuous stiffness distribution along the surface direction of a press. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2011-145247 A Summary of the Invention [Problem to be solved by the invention]

[0004] In view of the problems in the background art described above, an object of the present invention is to provide a novel method for predicting the stiffness of a press machine. [Means for solving the problem]

[0005] The gist of the present invention is as follows. the method includes an actual measurement step of applying a load to the press machine via a representative die and measuring a relationship between the load and an actual displacement at a plurality of measurement points on the press machine; an analysis model creation step of preparing a press machine model reflecting a long side length and a short side length of the press machine; a theoretical calculation model creation step of expressing a relationship between a deflection of the press machine model and the load as a function using a stiffness coefficient indicating the stiffness of the press machine; and a stiffness coefficient derivation step of deriving the stiffness coefficient based on the actual displacement and the load, A method for predicting stiffness of a press machine, comprising a correction step of correcting the stiffness coefficient based on a ratio between a load area of ​​the representative die and a load area of ​​a specified press die, and calculating a corrected stiffness coefficient of the press machine equipped with the specified press die.

[0006] (1) A method for predicting the stiffness of a press machine according to one embodiment of the present invention includes an actual measurement step of applying a load to the press machine via a representative die and measuring the relationship between the load and actual displacement at a plurality of measurement points on the press machine; an analytical model creation step of preparing a press machine model that reflects the long side length and short side length of the press machine; a theoretical calculation model creation step of expressing the relationship between the deflection of the press machine model and the load as a function using a stiffness coefficient that indicates the stiffness of the press machine; and a stiffness coefficient derivation step of deriving the stiffness coefficient based on the actual displacement and the load. (2) In the above (1), a correction step may be included in which the stiffness coefficient is corrected based on a ratio between a load area of ​​the representative die and a load area of ​​a specified press die, and a corrected stiffness coefficient of the press machine equipped with the specified press die is calculated. (3) In the above (1) or (2), the function may be a two-dimensional Fourier series expansion. (4) In the above (3), the function may be a two-dimensional Fourier series expansion using a sine function. (5) In the above (3), the function may be a two-dimensional Fourier series expansion of a Legendre polynomial. (6) In the above (3), the function may be a two-dimensional Fourier series expansion using Laguerre polynomials. (7) In the above (3), the function may be a two-dimensional Fourier series expansion of a Hermite polynomial. (8) In the above (4), the function may be expressed by the formula (A).

number

number

number

number

[0007] According to the present invention, a novel method for predicting the stiffness of a press machine can be provided. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is an explanatory diagram of a press molding apparatus including a press machine and a press die. [Diagram 2] FIG. 1 is a conceptual diagram of a press model. [Diagram 3] FIG. 2 is a cross-sectional view taken along the line A in FIG. 1, illustrating the external shapes of a press machine and a press die in plan view. [Figure 4] FIG. 1 is a flow diagram of a method for predicting press stiffness. [Diagram 5] FIG. 13 is a flow diagram of a stiffness coefficient derivation step. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] As shown in FIG. 1, a typical press molding apparatus 1, which is a machine that applies strong pressure to a blank (e.g., a metal material) not shown in the figure, and deforms the blank into the shape of a press die to produce a press-molded product, is basically composed of a press machine 10 and a press die 20 attached to the press machine 10.

[0010] The press 10 includes a bolster 11 to which a lower die 21 is attached at the upper part, a slide 12 to which an upper die 22 is attached at the lower part, and a drive device 13 that drives the bolster 11 and the slide 12 to move closer to or farther apart.

[0011] The press die 20 includes a lower die 21 and an upper die 22 .

[0012] The structure of the press machine 10 differs depending on the pressing capacity, specifications, design concept, etc. of the press machine 10. Even if the pressing capacity, specifications, design concept, etc. are the same, there are differences in the structure of the press machine 10 due to manufacturing errors, etc. Furthermore, the structure of the press die 20 attached to the press machine 10 differs depending on the specifications of the press-molded product to be manufactured. Therefore, even if the press load is the same, the rigidity of the press machine 10 differs due to differences in the structure of the press machine 10 and the structure of the press die 20, and therefore the state of deformation of the press machine 10 differs.

[0013] Incidentally, in order to manufacture a press-molded product with high accuracy by press molding, it may be necessary to estimate in advance the deflection ω (meaning the planar distribution of displacement, but hereinafter may also be referred to as displacement or deformation) of the press machine 10 in the shape of the press die 20, including the molding surface that contacts the blank of the press-molded product. In this case, it is preferable to grasp the rigidity of the press machine 10, which is the relationship between the load F (press load) and the deflection ω of the press machine 10. However, in order to accurately measure the deflection ω of the press machine 10 in order to grasp the rigidity of the press machine 10, a large number of measurement points are required, which requires a corresponding cost for equipment, etc. Furthermore, even if the rigidity of the press machine 10 is calculated using FEM or the like, there may be a deviation from the actual rigidity, or the calculation load may be excessive.

[0014] Therefore, the present invention is a method for predicting the stiffness of a press machine, including a measurement step of applying a load to the press machine via a representative die and measuring the relationship between the load and the actual displacement at a plurality of measurement points on the press machine, an analysis model creation step of preparing a press machine model reflecting the long side length and short side length of the press machine, a theoretical calculation model creation step of expressing the relationship between the deflection of the press machine model and the load as a function using a stiffness coefficient indicating the stiffness of the press machine, and a stiffness coefficient derivation step of deriving the stiffness coefficient based on the actual displacement and the load. This makes it possible to predict the stiffness of the press machine with a small calculation load and with a small deviation from the actual stiffness. Therefore, a novel method for predicting the stiffness of a press machine can be provided. Hereinafter, an embodiment of the present invention will be described in detail.

[0015] (Embodiment) Fig. 1 is an explanatory diagram of a press molding apparatus 1 equipped with a press machine 10 and a press die 20. Fig. 2 is an explanatory diagram of a press machine model 100. Fig. 3 is a cross-sectional view taken along the line A in Fig. 1, and is an explanatory diagram of the external shapes of the press machine 10 and the press die 20 in a plan view. Fig. 4 is a flow diagram of a method for predicting press machine rigidity. Hereinafter, the method for predicting the press stiffness according to this embodiment will be described with reference to the flow chart shown in FIG.

[0016] (Prediction method for press machine rigidity) As shown in FIG. 4, the method for predicting press machine stiffness according to this embodiment includes an actual measurement step S1, an analysis model creation step S2, a theoretical calculation model creation step S3, and a stiffness coefficient derivation step S4.

[0017] As a premise, first, a press machine 10 is prepared. The press machine 10 constitutes a general press forming apparatus 1 as described above, and includes a slide 12 and a bolster 11. The slide 12 or the bolster 11 is a plate-like body having a rectangular shape in a plan view, with a long side length a and a short side length b. The long side length a and the short side length b may be the same or different. In other words, the slide 12 or the bolster 11 may have a rectangular or square outer shape in a plan view. The long side length or the short side length of the slide 12 and the bolster 11 do not have to be the same. The slide 12 has a substantially flat lower surface to which the upper die 22 can be detachably attached. The bolster 11 has a substantially flat upper surface to which the lower die 21 can be detachably attached. That is, the lower surface of the slide 12 has a rectangular shape having a long side length a and a short side length b in a plan view. Similarly, the upper surface of bolster 11 has a rectangular shape having a long side length a and a short side length b in a plan view.

[0018] In addition, a representative die is prepared to be attached to the press machine 10. The representative die is a press die 20 for measuring the deformation of the press machine 10. The representative die includes an upper die 22 attached to the slide 12 and a lower die 21 attached to the bolster 11 .

[0019] 3, the upper die 22 and the lower die 21 have an outer shape having a die long side length α and die short side length β in a plan view. The upper die 22 faces and comes into surface contact with the lower surface of the slide 12, and has an attachment surface having the die long side length α and die short side length β. The lower die 21 faces and comes into surface contact with the upper surface of the bolster 11, and has an attachment surface having the die long side length α and die short side length β. The long side length or short side length of the upper die 22 and the lower die 21 do not have to be the same.

[0020] In a plan view, the outer shape of the press machine 10 and the outer shape of the representative die (press die 20) may be similar. That is, the ratio of the long side length a to the short side length b may be equal to the ratio of the die long side length α to the die short side length β. For example, a / α is 3 / 1 and b / β is 3 / 1.

[0021] Next, the representative die is attached to the press machine 10. Specifically, the upper die 22 is attached with the mounting surface of the upper die 22 of the representative die facing the lower surface of the slide 12 of the press machine 10 in surface contact. The lower die 21 is attached with the mounting surface of the lower die 21 of the representative die facing the upper surface of the bolster 11 of the press machine 10 in surface contact.

[0022] (1) Then, a load F is applied to the press machine 10 via the representative die, and the actual displacement ω at the coordinates (xi, yj) of multiple measurement points on the press machine 10 is calculated. ij The relationship between the load F and the actual displacement ω ij ω(x i ,y j , F). In the following, virtual displacements ωω at multiple coordinates (xi, yj) in the press machine model 100 when a load F is applied are ij ωω(x i ,y j ,F). In addition, the slide 12 and the bolster 11 of the press machine 10 usually have different plate thicknesses, reinforcement structures, etc., and the rigidity coefficient A mn Since the distribution of stiffness is different between the slide 12 and the bolster 11 of the press 10, the measurements are performed separately for the slide 12 and the bolster 11. In the following description, unless otherwise specified, the term "press" may refer to either the slide 12 or the bolster 11. The load F is a force acting in the vertical direction. The coordinates (xi, yj) of the multiple measurement points are i points in the long side direction (x direction) and j points in the short side direction (y direction). For example, if there are 10 points along each of the two lines in the long side direction and 10 points along each of the two lines in the short side direction, there are a total of 40 measurement points. The measurement points are at least five points including one point near the centroid including the centroid, two points along the long side direction passing through the centroid, and two points along the short side direction passing through the centroid. In reality, the measurement points include points where relatively large deformation is expected, such as near the long side and short side, and multiple points along the long side and short side directions to include the surface of the press machine 10 (slide 12 or bolster 11) facing the representative die, and are set symmetrically in a balanced manner without being unevenly distributed on the surface of the press machine 10 so that the deformation of the surface of the press machine 10 can be seen. Actual displacement ω ij is measured at the coordinates (xi, yj) of the measurement point by a displacement meter capable of measuring vertical displacement. The displacement meter may be, for example, a contact type using a spring or a non-contact type using a laser, and may calculate the displacement by analyzing the change in an image captured by an imaging device using a computer program.

[0023] (2) A press machine model 100 reflecting the long side length a and short side length b of the press machine 10 is prepared (analysis model creation step S2). As shown in FIG. 2, the press machine model 100 is a virtual analysis model simulating the press machine 10 (slide 12 or bolster 11). The press machine model 100 is a solid rectangular parallelepiped with four sides (two long sides and two short sides, i.e., peripheral portions) supported, having a long side length a and a short side length b, and having a uniform cross section (constant thickness). A reaction force acts on the press machine model 100 from a press die model 200 having a die long side length α and a die short side length β.

[0024] (3) Next, the relationship between the deflection ωω of the press model 100 and the load F is expressed as the stiffness coefficient A mn (theoretical calculation model creation step S3). Here, the press model 100 is a solid rectangular parallelepiped with a uniform cross section (constant thickness) having a long side length a and a short side length b as described above, and is supported at all four sides (two long sides and two short sides). Here, since the mounting surface of the press die facing the press 10 is actually a relatively simple paraboloid shape, it is considered that a deflection ω closer to reality can be calculated by applying Saint-Venant's principle to the transfer of load from the press die to the press 10 and reflecting this principle in the calculation of the press model 100. The deflection of a rectangular plate-like body such as the press model 100 is symmetrical in two orthogonal directions, the long side direction and the short side direction. Therefore, since it is clear that the shape will be like superposition of waves whose half period is the value obtained by dividing the long side length a or the short side length b by an odd number in each of the long side and short side directions, the deflection ω can be calculated using the analytical solution of the deflection of a uniform cross-section, four-sided supported flat plate, which is a function expanded into a two-dimensional Fourier series, as shown in equation (1).

number

[0025] Here, instead of pressure p0, we substitute load F, and group the other constants together as A. mn Then, equation (1) can be simplified to equation (2).

number

[0026] Furthermore, as m and n increase in equation (2), A mn Since is small enough to be ignored, when equation (2) is made into a finite series, it becomes the following equation.

number

number

[0027] In addition, the deflection ω or stiffness coefficient A of the actual press machine 10 mn When calculating, M and N are odd numbers to represent the geometrically symmetric shape of the deflection ω. Considering the balance between the calculation load and calculation accuracy, M and N are sufficient to be 15 or less, and may be 7 or less, respectively. In this way, the deflection ω can be expressed as a function expanded into a two-dimensional Fourier series by a sine function. This allows the calculation to be performed using an algebraic solution based on an analytical solution, and when the scaling parameters are constant, it is possible to calculate a deflection ω close to the actual value in a short analysis time, and a stiffness coefficient A close to the actual value. mn can be predicted.

[0028] (4) And the real displacement ω ij and stiffness coefficient A based on load F mn is derived (stiffness coefficient derivation step S4). For example, when the scaling parameter is constant, instead of formula (A), the matrix representing the deflection ω at each coordinate (xi, yj) is expressed in the form of the product of the matrix of the stiffness coefficient, the deformation mode in the x direction, and the deformation mode in the y direction and the load F, and the stiffness coefficient A is calculated by using singular value decomposition. mn can be derived.

[0029] Generally, as shown in Figure 5, the stiffness coefficient A is calculated by the following flow: mnFig. 5 is a flow diagram of the stiffness coefficient derivation step S4. (4-1) First, the provisional stiffness coefficient A as shown in equation (4) mnt , the provisional long side direction scaling parameter r mt a and the temporary short side direction scaling parameter r nt b Then, as the initial value, the provisional stiffness coefficient A mnt The initial parameter A mn0 and the initial long-side scaling parameter r m0 a and the initial short-side scaling parameter r n0 b And set it.

number

[0030] (4-2) Next, stiffness coefficient A mn Instead of the provisional stiffness coefficient A mnt The coordinates (x, y) of the measurement point are substituted into equation (A) into which the above equation has been substituted. This is performed for each of the coordinates (xi, yj) of the measurement points. That is, as shown in equation (5), the virtual displacement ωω ij (=ωω(x i ,y j ,F) is calculated.

number

[0031] (4-3) Next, the actual displacement ω ij and virtual displacement ωω ij Evaluate the residual e with The residual e is calculated by the first term of equation (6) as follows: ij and virtual displacement ωω ij The virtual displacement ωω calculated by Equation (5) is calculated as the square error, which is the square of the difference between ijSince the relationship between and the load F is relatively strongly nonlinear, the residual e may include a coefficient penalty term, such as the second term in equation (6), to improve the reproducibility and stability of the calculation.

number

[0032] (4-4) Here, if the residual e calculated by equation (6) is equal to or greater than a predetermined value tol, the provisional stiffness coefficient A mnt In other words, the previous temporary stiffness coefficient A mnt A new provisional stiffness coefficient A that is slightly different (for example, by 0.1%) from mnt Then, equations (4) to (6) are calculated again. This calculation is repeated until the residual e calculated in equation (6) becomes less than a predetermined value tol (repeated calculation loop of neighborhood search). On the other hand, if the residual e calculated by the formula (6) is less than the predetermined value tol, the calculation is terminated. mnt is the stiffness coefficient A mn Let us assume that. In this way, the stiffness coefficient A mn can be derived.

[0033] (4-5) Also, the virtual displacement ωω calculated by equation (5) ij Since the relationship between the load F and the load F has a relatively strong nonlinearity, in order to avoid falling into an inappropriate local optimum solution, it is preferable to further provide a large fluctuation iterative calculation loop (for example, the Basin Hopping method) outside the iterative calculation loop for neighborhood search described in (4-4) in order to approach the global optimum solution. That is, the provisional stiffness coefficient A set previously is mnt A new provisional stiffness coefficient A that is significantly different (for example, by 5%) from mnt Then, calculate equations (4) to (6) again.

[0034] (5) Then, the derived stiffness coefficient A mnUsing equation (A), the actual displacement ω ij Calculate.

[0035] According to this embodiment, since the above-mentioned steps (actual measurement step, analytical model creation step, theoretical calculation model creation step, and stiffness coefficient derivation step) are included, it is possible to obtain not only the relationship between load and displacement at the representative points (lattice points) of the press machine 10, but also a continuous stiffness distribution along the surface (xy plane) of the press machine 10. In addition, the stiffness of the press machine 10 (stiffness coefficient A mn ) can be predicted with a small calculation load and with a small deviation from the actual stiffness. In addition, the stiffness coefficient A calculated through each of these steps can be calculated with a small calculation load and with a small deviation from the actual stiffness. mn By using this method, the deflection (deformation) of the press machine 10 can be accurately taken into account in advance when designing the shape of the press die, including the molding surface that contacts the workpiece. Therefore, the workpiece can be manufactured with high accuracy by the press molding device.

[0036] By the way, when the shape of the representative die and the specified press die are different, that is, when the load area α*β of the representative die is different from the load area α2*β2 of the specified press die, for example, the stiffness coefficient A is calculated based on the ratio of the load area α*β of the representative die to the load area α2*β2 of the specified press die. mn In this way, the stiffness coefficient A , which indicates the stiffness of the press machine 10 equipped with a press die having a shape different from that of the representative die, can be calculated. mn Therefore, even if the press die is changed, the rigidity of the press 10 can be appropriately predicted.

[0037] (Function expanded in two-dimensional Fourier series using Legendre polynomials) Incidentally, the above-mentioned formula (A) is a function expanded into a two-dimensional Fourier series by a sine function, but instead of this, a function expanded into a two-dimensional Fourier series by Legendre polynomials, such as formula (B), may be used.

number

[0038] (Function expanded in two-dimensional Fourier series using Laguerre polynomials) Incidentally, instead of the function of the above formula (A) or (B), a function expanded into a two-dimensional Fourier series by Laguerre polynomials, such as formula (C), may be used.

number

[0039] (Function expanded in two-dimensional Fourier series using Hermite polynomials) Incidentally, instead of the function of the above formula (A), formula (B) or formula (C), a function obtained by expanding a two-dimensional Fourier series using a Hermite polynomial, such as formula (D), may be used.

number

[0040] (Press machine rigidity prediction program) Next, a press machine stiffness prediction program for causing a computer to execute the above-mentioned method for predicting the press machine stiffness will be described. The program for predicting the press stiffness is a program for predicting the stiffness of the press machine 10 by applying a load F to the press machine 10 via a representative die, and calculating the relationship between the load F and the actual displacement ω at a plurality of measurement points on the press machine 10. ij and a step of creating an analytical model of the press machine model 100 reflecting the long side length a and the short side length b of the press machine 10. The relationship between the deflection ωω of the press machine model 100 and the load F is calculated using a stiffness coefficient A indicating the stiffness of the press machine 10. mn The theoretical calculation model creation step expressed as a function using the actual displacement ω ij and stiffness coefficient A based on load F mn and the stiffness coefficient derivation step to derive stiffness coefficient A mn and a stiffness coefficient output step of outputting the stiffness coefficient. Here, the stiffness coefficient A output by the press stiffness prediction program is mn is stored in a memory associated with the computer. The stiffness coefficient A mn The stiffness coefficient A may be displayed on a monitor or printed out on a printer. mn For example, a three-dimensional diagram showing the change in the deflection (deformation) of the press machine 10 in response to an increase or decrease in the load F can be displayed on a monitor so that the user can visually grasp the change. In this way, the press stiffness prediction program uses the stiffness coefficient A mn Since the above can be output, it can be used for CAE, and the press die 20 can be designed taking into account the deformation of the press machine 10. Therefore, the press die 20 designed taking into account the deformation of the press machine 10 can manufacture press-molded products with high precision.

[0041] (Press machine rigidity prediction device) Next, a press machine stiffness prediction device equipped with the above-mentioned press machine stiffness prediction program will be described. The press stiffness prediction device is equipped with a storage device, a processor, a memory, an interface, and an output device, similar to a general computer. In detail, the press stiffness prediction device applies a load F to the press 10 via a representative die, and calculates the relationship between the load F and the actual displacement ω at a plurality of measurement points on the press 10. ij and an analytical model creation section that prepares a press machine model 100 that reflects the long side length a and short side length b of the press machine 10. The analytical model creation section creates an analytical model that creates an analytical model that represents the relationship between the deflection ωω of the press machine model 100 and the load F using a stiffness coefficient A that indicates the stiffness of the press machine 10. mn The theoretical calculation model creation part expresses the actual displacement ω ij and stiffness coefficient A based on load F mn and a stiffness coefficient derivation part that derives the stiffness coefficient A mn and a stiffness coefficient output unit that outputs the stiffness coefficient. Here, the stiffness coefficient output section is the stiffness coefficient A mn Memory to store stiffness coefficient A mn Monitor that displays stiffness coefficient A mn Includes a printer to print. In this way, the press stiffness prediction device uses the stiffness coefficient A mn Since the press die 20 is equipped with a stiffness coefficient output unit that outputs the stiffness coefficient, the stiffness coefficient can be used in CAE, and the press die 20 can be designed in anticipation of the deformation of the press machine 10. Therefore, the press die 20 designed in anticipation of the deformation of the press machine 10 can manufacture press-molded products with high precision.

[0042] (Other embodiments) Although the embodiment has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design modifications and the like are possible. For example, the above-mentioned two-dimensional Fourier series expanded functions (formula (A), formula (B), formula (C) and formula (D)) are polynomials in which the basis functions in the long side direction and the short side direction are of the same type (for example, formula (B) is a polynomial combining Legendre functions and Legendre functions), but are not limited to this. For example, the polynomials may be polynomials in which different types of basis functions are combined in the long side direction and the short side direction, such as a Legendre function (see formula (B)) as the basis function in the long side direction and a Hermite function (see formula (D)) as the basis function in the short side direction.

[0043] The method for predicting the stiffness of the press machine 10 according to the present embodiment is to apply a load F to the press machine 10 via a representative die, and to obtain a relationship between the load F and the actual displacement ω at a plurality of measurement points on the press machine 10. ij The method also includes an analysis model creation step of preparing a press machine model 100 that reflects the long side length a and short side length b of the press machine 10. The method also includes an analysis model creation step of preparing an analysis model of the press machine model 100 that reflects the deflection ωω of the press machine model 100 and the load F using a stiffness coefficient A mn The step of creating a theoretical calculation model expressed as a function using the actual displacement ω ij and stiffness coefficient A based on load F mn and a stiffness coefficient derivation step of deriving a stiffness coefficient. The stiffness prediction method for the press 10 according to this embodiment goes through each of these steps, so that the press stiffness can be predicted with a small calculation load and with a small deviation from the actual stiffness. In addition, a novel method for predicting the stiffness of the press 10 can be provided that can predict not only the stiffness at several representative points actually measured in the press 10, but also a continuous (planar) stiffness distribution along the planar direction of the press 10. [Explanation of symbols]

[0044] 1 Press forming equipment 10 Press machine 11 Bolster 12 Slides 13 Drive unit 20 Press mold 21 Lower mold 22 Upper mold 100 Press Machine Model 200 Press die model a Long side length b Short side length A mn Stiffness coefficient A mn0 Initial parameters A mnt Provisional stiffness coefficient e residual F load p0 pressure S1 Actual step S2 Analysis model creation step S3 Theoretical calculation model creation step S4 Stiffness coefficient derivation step tol predetermined value (x,y), (xi,yj) coordinates α Mold long side length β Mold short side length ω (press machine) deflection ωω (press machine model) deflection ω ij Actual Displacement ωω ij Virtual Displacement r m a Long side scaling parameter r n b Short side scaling parameter r m0 a Initial long side scaling parameter r n0 b Initial short side scaling parameter

Claims

1. A method for predicting stiffness of a press, comprising the steps of: a measuring step of applying a load to the press machine via a representative die and measuring a relationship between the load and actual displacements at a plurality of measurement points on the press machine; an analysis model creation step of preparing a press machine model reflecting the long side length and the short side length of the press machine; a theoretical calculation model creation step of expressing the relationship between the deflection of the press machine model and the load as a function using a stiffness coefficient indicating the stiffness of the press machine; and a stiffness coefficient derivation step of deriving the stiffness coefficient based on the actual displacement and the load, A correction step of correcting the stiffness coefficient based on a ratio between a load area of ​​the representative die and a load area of ​​a specified press die, and calculating a corrected stiffness coefficient of the press machine equipped with the specified press die. A method for predicting the stiffness of a press machine.

2. 2. The method for predicting stiffness of a press machine according to claim 1, wherein the function is a function obtained by performing a two-dimensional Fourier series expansion.

3. 3. The method for predicting stiffness of a press machine according to claim 2, wherein the function is a two-dimensional Fourier series expansion by a sine function.

4. 3. The method for predicting stiffness of a press machine according to claim 2, wherein the function is a two-dimensional Fourier series expansion of a Legendre polynomial.

5. 3. The method for predicting stiffness of a press machine according to claim 2, wherein the function is a two-dimensional Fourier series expansion of a Laguerre polynomial.

6. 3. The method for predicting stiffness of a press machine according to claim 2, wherein the function is a two-dimensional Fourier series expansion of a Hermite polynomial.

7. The function is represented by the formula (A):

4. The method for predicting stiffness of a press machine according to claim 3. [0010] ...Formula (A) Where, F: load x: Coordinate of the measurement point in the long side direction y: Coordinate of the measurement point in the short side direction A mn : stiffness coefficient a: Long side length b: Short side length r m a : Long side direction scaling parameter (m = 1, 3, ...) r n b : Short side direction scaling parameter (n = 1, 3, ...)

8. The function is represented by the formula (B):

5. The method for predicting stiffness of a press machine according to claim 4. [0025] ...Formula (B) Here, P m : m-th degree Legendre polynomial P n : nth degree Legendre polynomial F: Load x: Coordinate of the measurement point in the long side direction y: Coordinate of the measurement point in the short side direction A mn : stiffness coefficient a: Long side length b: Short side length r m a : Long side direction scaling parameter (m = 0, 1, ...) r n b : Short side direction scaling parameter (n = 0, 1, ...)

9. The function is represented by the formula (C):

6. The method for predicting stiffness of a press machine according to claim 5. [0030] ...Formula (C) Here, L m : m-th order Laguerre polynomial L n : nth degree Laguerre polynomial F: Load x: Coordinate of the measurement point in the long side direction y: Coordinate of the measurement point in the short side direction A mn : stiffness coefficient a: Long side length b: Short side length r m a : Long side direction scaling parameter (m = 0, 1, ...) r n b : Short side direction scaling parameter (n = 0, 1, ...)

10. The function is represented by the formula (D):

7. The method for predicting stiffness of a press machine according to claim 6. [0045] ...Formula (D) Here, H m : m-th degree Hermite polynomial H n : nth degree Hermite polynomial F: Load x: Coordinate of the measurement point in the long side direction y: Coordinate of the measurement point in the short side direction A mn : stiffness coefficient a: Long side length b: Short side length r m a : Long side direction scaling parameter (m = 0, 1, ...) r n b : Short side direction scaling parameter (n = 0, 1, ...)

11. The series of the two-dimensional Fourier series expansion is finite. The method for predicting stiffness of a press according to any one of claims 2 to 10.

Citation Information

Patent Citations

  • Press machine

    JP1992228295A

  • Bending apparatus and its method

    JP2004290992A

  • Method of acquiring deflection and apparatus for calculating deflection, and method of acquiring control load and apparatus for calculating control load

    JP2009113044A

  • Method and apparatus for measuring dynamic deformation of press die

    JP2011145247A

  • Formation simulation system

    JP2014097520A