WO2016186135A1 - 破断予測方法、プログラム、記録媒体及び演算処理装置 - Google Patents
破断予測方法、プログラム、記録媒体及び演算処理装置 Download PDFInfo
- Publication number
- WO2016186135A1 WO2016186135A1 PCT/JP2016/064753 JP2016064753W WO2016186135A1 WO 2016186135 A1 WO2016186135 A1 WO 2016186135A1 JP 2016064753 W JP2016064753 W JP 2016064753W WO 2016186135 A1 WO2016186135 A1 WO 2016186135A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- mesh
- roughness
- fracture
- maximum principal
- value
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N3/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N3/28—Investigating ductility, e.g. suitability of sheet metal for deep-drawing or spinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N2203/0058—Kind of property studied
- G01N2203/006—Crack, flaws, fracture or rupture
- G01N2203/0067—Fracture or rupture
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N2203/02—Details not specific for a particular testing method
- G01N2203/0202—Control of the test
- G01N2203/0212—Theories, calculations
- G01N2203/0216—Finite elements
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N2203/02—Details not specific for a particular testing method
- G01N2203/026—Specifications of the specimen
- G01N2203/0262—Shape of the specimen
- G01N2203/0278—Thin specimens
- G01N2203/0282—Two dimensional, e.g. tapes, webs, sheets, strips, disks or membranes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/24—Sheet material
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
Definitions
- the present invention relates to a fracture prediction method, a program, a recording medium, and an arithmetic processing device for predicting a fracture site when molding analysis is performed by a finite element method.
- the present application claims priority based on Japanese Patent Application No. 2015-101311 filed in Japan on May 18, 2015, the contents of which are incorporated herein by reference.
- Patent Document 1 discloses a material whose stretch flangeability is improved by controlling a microstructure such as ferrite and bainite.
- Patent Document 2 discloses an aluminum alloy plate excellent in stretch flangeability obtained by defining plastic anisotropy and uniform elongation in a tensile test in a specific direction.
- Patent Document 3 discloses a method of predicting a molding defect at the time of molding using a finite element method. According to this, the analysis is performed using the finite element method, and the occurrence of the molding defect is determined using the strain and stress data of the element of interest. However, when such a method is used, it is necessary to divide the element into an appropriate size according to the analysis target. When the analysis is performed with an inappropriate element division, the prediction result is overestimated or underestimated. Therefore, there are cases that do not correspond to reality.
- Patent Document 4 unlike the conventional technique, the plate thickness reduction rate or the maximum principal strain distribution in two cases of finite element analysis with different element division sizes is quantitatively compared. As a result, we succeeded in increasing the accuracy of the prediction of the stretch flange cracked part.
- the technique of Patent Document 4 enables prediction of stretched flange cracking portions without necessarily improving the calculation accuracy of finite element analysis. Therefore, it can be performed in a short time and at a low calculation cost without limiting the finite element analysis software. There are significant advantages to what is possible.
- the geometric deformation amount of the plate thickness reduction rate or the distribution of the maximum principal strain is used as an evaluation index for specifying the stretch flange crack portion.
- the steel plate that is the analysis target component is a thin plate material that has a relatively high deformability and can be formed to a large strain region, it is possible to predict and specify the stretch flange portion without any problem.
- the steel plate as the analysis target part is a high-tensile steel plate or a hard-to-form metal plate
- the difference in the maximum principal strain distribution in the results of finite element analysis in two different element divisions becomes smaller, and the stretch flange cracks. The tendency for the detection accuracy of a site
- the present invention has been made in view of the above problems, and in a method for predicting breakage of a part obtained by forming a metal plate using a finite element method, breakage prediction capable of easily and reliably extracting a breakage site. It is an object to provide a method, a program, a recording medium, and an arithmetic processing device.
- the fracture prediction method is a method for predicting a fracture site of a part obtained by forming a metal plate, and the metal plate is divided by a first mesh roughness.
- a difference value between the maximum principal stress and the maximum principal stress in the case of the second mesh roughness is obtained, and the difference value corresponds to a portion where the difference value is larger than a predetermined value.
- two types of mesh roughness are used: a first mesh roughness and a second mesh roughness that is coarser than the first mesh roughness.
- the maximum principal stress in each mesh is averaged and output. Therefore, when there is a stress concentration site in a certain mesh, the maximum principal stress in the case of the finer first mesh roughness is less influenced by the averaging, and therefore in the case of the coarser second mesh roughness.
- the fracture site is extracted by quantitative comparison of the maximum principal stress in each case of the first mesh roughness and the second mesh roughness, it is possible to compare the two. It is not essential to set the first mesh roughness to a very small mesh roughness as in the prior art. Therefore, the molding analysis can be performed in a short time and at a low calculation cost.
- the maximum principal stress is used as an index. In the case of a metal plate having a high tensile strength and a small elongation, such as a high-strength steel plate, the stress changes greatly even when the deformation is small.
- the geometric deformation amount which is the plate thickness reduction rate or the maximum principal strain
- the deformation amount itself is small, so the difference between the index values for different mesh roughness is unclear, and it is difficult to predict the fracture site. become.
- the maximum principal stress which is the amount of mechanical change
- the difference between the index values becomes clear, and the fracture site can be predicted easily and reliably even for metal plates with high tensile strength and small elongation. Can do.
- the first mesh roughness and the second mesh roughness are determined based on an n value indicating work hardening characteristics of the metal plate.
- the steps may be further included.
- the first mesh roughness and the second mesh roughness are optimally set based on the n value. Therefore, it is possible to obtain excellent prediction accuracy without decreasing the prediction accuracy due to the mesh roughness being too coarse, and conversely without increasing the calculation time by using the mesh roughness that is unnecessarily fine.
- the third step may be performed again from the first step.
- the effect of averaging the maximum principal stress in the mesh is reduced, that is, the portion where the stress is concentrated is revealed.
- an adaptive mesh may be used when performing division based on the first mesh roughness.
- the analysis time can be reduced by performing analysis with a coarse mesh at the start of molding, and finely re-dividing the mesh only at a portion where deformation or stress change is large as the molding process proceeds.
- the mesh is made finer at a part where deformation or stress change is large, that is, a part where the risk of fracture is high, the prediction accuracy can be improved.
- the molding analysis in the first step may be terminated during the molding of the part.
- the so-called halfway stop evaluation is performed by stopping the forming analysis in the middle of forming the part.
- Interim stop evaluation enables evaluation of the differential value of the maximum principal stress as the molding process progresses, before the stress state approaches tensile strength or before deformation becomes excessive and failure occurs in the molding analysis. To do.
- a case where the second mesh is further divided by the first mesh roughness and the second mesh For each of the cases when divided by roughness, a shape index value that is at least one of maximum principal strain and sheet thickness reduction rate is obtained for each mesh; in the third step, A difference value between the shape index value in the case of the first mesh roughness and the shape index value in the case of the second mesh roughness is obtained, and the difference value of the shape index value is larger than a predetermined value. A portion in the case of the first mesh roughness corresponding to a portion satisfying at least one of being larger and a difference value of the maximum principal stress being larger than a predetermined value is extracted as the fracture portion; Like It may be.
- the prediction using at least one of the maximum principal strain and the sheet thickness reduction rate is also performed.
- the certainty of prediction can be improved.
- a metal plate with high tensile strength and small elongation a metal plate with low tensile strength and large elongation has a large amount of deformation.
- the prediction using maximum principal stress and the maximum principal It is desirable to use both predictions instead of using only one of the predictions using at least one of the strain and the plate thickness reduction rate.
- a high-strength metal plate with low elongation for example, a high-strength steel plate
- a metal plate particularly suitable for prediction using the maximum principal strain or plate thickness reduction rate for example, a mild steel plate
- various types of metal plates such as metal plates having intermediate strengths
- the fracture prediction method is a method for predicting a fracture site of a part obtained by forming a metal plate, the metal plate being divided by a predetermined mesh roughness,
- a first step for performing a forming analysis using a finite element method a second step for obtaining a maximum principal stress for each mesh; and a maximum principal stress for each coupled mesh obtained by joining two or more meshes adjacent to each other.
- the maximum principal stresses before and after joining the meshes are compared.
- the fracture site can be predicted easily and reliably.
- the molding analysis since the molding analysis is basically performed only once, the molding analysis can be performed in a shorter time and at a lower calculation cost.
- an adaptive mesh may be used when performing division based on the predetermined mesh roughness.
- the analysis time can be reduced and the prediction accuracy can be improved.
- the molding analysis in the first step may be terminated during the molding of the part.
- the difference value of the maximum principal stress is calculated. make it possible to evaluate.
- a shape index value that is at least one of a maximum principal strain and a sheet thickness reduction rate is further set. Obtained for each mesh; in the third step, the shape index value is further obtained for each coupled mesh; in the fourth step, the shape index value obtained in the second step; A difference value with respect to the shape index value obtained in step 3 is obtained for each part of the part, the difference value of the shape index value is larger than a predetermined value, and the difference value of the maximum principal stress is larger than a predetermined value.
- the part in the second step corresponding to the part satisfying at least one of the above may be extracted as the fracture part.
- the prediction using the maximum principal stress in addition to the prediction using the maximum principal stress, the prediction using at least one of the maximum principal strain and the sheet thickness reduction rate is also performed, so that various types of Excellent prediction accuracy can be obtained for the metal plate.
- a stretch flange crack occurrence site may be predicted as the fracture site. In this case, it is possible to predict the occurrence of stretch flange cracks which are particularly problematic as breaks.
- the fracture site at the end of the component may be extracted. In this case, it is possible to predict the breakage at the end portion of the component where the risk of breakage is high.
- the metal plate may be a steel plate having a tensile strength of 980 MPa or more.
- the fracture can be predicted for a steel plate having a tensile strength of 980 MPa or more, which is a difficult-to-form metal plate.
- a program according to an aspect of the present invention executes the fracture prediction method according to any one of (1) to (13) above.
- An arithmetic processing apparatus executes the program described in (14) above.
- FIG. 7A It is a figure which shows the outline
- the inventor pays attention to the fact that the stress gradient is large around the fracture site of the part obtained by forming the metal plate, which is the analysis target part, and the principal stress depends on the mesh roughness in the analysis by the finite element method.
- a new fracture prediction method that takes advantage of the fact that averaging is performed.
- two types of meshes (elements) having different mesh roughness hereinafter also referred to as mesh size or element size, which are used interchangeably
- mesh size or element size which are used interchangeably
- analysis is performed using finer mesh roughness as first mesh roughness and coarser mesh roughness as second mesh roughness.
- the principal stress in the mesh is averaged and output.
- the main stress output as an average value in the first mesh roughness in the case of the first mesh roughness and the second mesh roughness. However, it becomes larger than the main stress output as an average value in the second mesh roughness.
- two types of mesh roughness are used to analyze the first mesh roughness and the second mesh roughness at each part of the component. I do.
- the main stress output as an average value is different between the first mesh roughness and the second mesh roughness, it can be considered that there is a stress gradient in the mesh.
- This difference in main stress corresponds to the magnitude of the stress gradient. The greater the stress gradient, the higher the risk of rupture, and it is possible to predict the risk of rupture based on the difference in principal stress.
- the present invention instead of using two different types of mesh roughness as described above, after performing molding analysis by division with a predetermined mesh roughness, after obtaining the main stress for each mesh with a predetermined mesh roughness, Two or more meshes adjacent to each other are joined to form a joined mesh, the principal stress for each joined mesh is obtained, and the difference between the principal stress at a predetermined mesh roughness before joining and the principal stress in the joined mesh You may comprise so that it may obtain
- the main stress output as the average value in the mesh at the predetermined mesh roughness is different from the main stress output as the average value in the coupled mesh larger than the mesh at the predetermined mesh roughness.
- the maximum principal stress is used as the principal stress that is an index value for predicting fracture.
- the prediction of the fracture occurrence site at the time of press forming of the metal plate is realized by high-precision, low-cost, short-time numerical simulation.
- the maximum principal stress is calculated according to the following procedures 1 to 4 (see Non-Patent Documents 1 and 2).
- Procedure 1 Each component of the stress tensor is calculated by a numerical calculation method such as a finite element method.
- Procedure 2 Each component of the stress tensor can be expressed as a 3 ⁇ 3 matrix.
- Procedure 3 The principal stress (three numerical values of ⁇ 1, ⁇ 2, and ⁇ 3) is obtained from each component of the stress tensor.
- the principal stress is a value obtained as an eigenvalue of the stress tensor.
- Procedure 4 The principal stress having the maximum value among the three principal stresses obtained is treated as the “maximum principal stress”. For example, when there is a relationship of ⁇ 1> ⁇ 2> ⁇ 3, ⁇ 1 is regarded as the maximum principal stress.
- the distribution of the maximum principal stress in the finite element analysis with different mesh roughness is compared, and when the difference becomes sufficiently large, it is regarded as a fracture site. This makes it possible to predict a fracture site even in a high-strength steel sheet having high tensile strength and low elongation.
- the fracture prediction method of the present invention when performing fracture prediction of a part obtained by forming a metal plate, which is a part to be analyzed, shown in FIGS. 1A and 1B.
- the parts are divided by the first mesh roughness and the second mesh roughness which is coarser than the first mesh roughness, respectively, using the finite element method. I do.
- the calculation step S12 by the calculation means 12 the maximum principal stress is calculated and obtained for each mesh with the first mesh roughness and for each mesh with each second mesh roughness.
- the extraction step S14 by the extraction means 14 the difference value between the maximum principal stress in the case of the first mesh roughness and the maximum principal stress in the case of the second mesh roughness at each part of the part. And the part in the case of the first mesh roughness corresponding to the part whose difference value is larger than the predetermined value is extracted as the fracture part.
- the computer program causes the central processing unit (CPU) of the arithmetic processing unit (computer) to execute each step (division step S11, calculation step S12, extraction step S14).
- the computer program causes the central processing unit (CPU) of the arithmetic processing unit (computer) to function as each unit (dividing unit 11, calculating unit 12, and extracting unit 14).
- the computer program can be recorded on a computer-readable recording medium, such as a flexible disk or a CD-R.
- This arithmetic processing unit may have an input means 13 for inputting the maximum principal stress obtained for each divided mesh to another computer.
- an input means a keyboard, a mouse, various digitizers, etc. can be used.
- the input step S13 may be a step of inputting with a keyboard, or a step of automatically inputting the maximum principal stress calculated in the calculation step S12 into the extraction step 15 (reading data) in the program. But it ’s okay.
- a solid line indicates an essential means or process
- a broken line indicates an optional means or process.
- the dividing means 11 when a part is divided into a plurality of elements (that is, meshes), if solid elements (three-dimensional elements) are used, three-dimensional part shape digital data (CAD data or A part is expressed as shape measurement data), or when a shell element (two-dimensional element) is used, the part is expressed as a set of two-dimensional plane regions.
- CAD data three-dimensional part shape digital data
- a part is expressed as shape measurement data
- a shell element two-dimensional element
- the part is expressed as a set of two-dimensional plane regions.
- the corner portion of the part has a large shape change, it is divided by a sufficiently small mesh to ensure shape reproducibility.
- the entire part when performing mesh division by the first mesh roughness and the second mesh roughness having different roughness, the entire part may be subdivided (or roughened) uniformly, or the fracture prediction may be performed.
- the part to be performed may be subdivided or roughened. In terms of work man-hours, the former is convenient, and the latter is advantageous for shortening the calculation time. Therefore, it may be appropriately selected or combined in consideration of the entire load.
- the first mesh roughness and the second mesh roughness are determined in relation to the n value indicating the work hardening characteristic of the analysis target component.
- the mesh is sufficiently finely reproduced so as to reproduce the geometrical shape of the target portion, that is, for example, the curvature of the end or the radius of curvature of the corner. It is necessary to divide.
- the difference in the maximum principal stress between the first mesh roughness and the second mesh roughness is necessary to divide.
- n is an n value of the material
- a function f (n; k, L, L0) for adjusting the mesh size is given as follows.
- f (n; k, L, L0) (L-L0) x (2 / ⁇ ) x tan -1 (k x n) + L0 (3)
- L and L0 are an upper limit value and a lower limit value of the mesh size (mesh roughness), respectively.
- the variable k is a parameter for adjusting the rate of change of the mesh size with respect to the n value, and as a result of investigation, it is considered that a value of about 50 ⁇ k ⁇ 100 is appropriate.
- a value of k 65 is adopted.
- Equation (3) In the function f (n; k, L, L0) that defines the mesh size range, the three variables (k, L, L0) are used as constants, so the function f in Equation (3) is substantially It functions as a function that determines the mesh size depending only on the n value.
- This function f increases with the n value.
- n value is large, deformation localization is difficult to occur, so that the fracture prediction accuracy can be ensured even if the mesh division is large.
- the n value is small, deformation is likely to occur locally, and therefore, the deformation gradient of the fractured portion becomes large, and the fracture prediction accuracy decreases unless sufficient mesh division is performed. This is determined because the size of the division needs to be reduced.
- the ratio L coarse / L fine between L coarse and L fine is 1.5 or more, preferably 2 or more.
- L coarse that is, the second mesh roughness
- L fine that is, the first mesh roughness
- the prediction accuracy is not lowered by the mesh roughness that is too coarse.
- an excessively fine mesh roughness it is possible to obtain an excellent prediction accuracy without increasing the calculation time or reducing the prediction accuracy.
- the difference between the maximum principal stresses described above is determined based on an analysis result having the finest mesh division roughness (that is, an analysis result at the first mesh roughness) as a reference to another mesh closest to the target mesh position.
- the mesh of the analysis result (that is, the analysis result with the second mesh roughness) is extracted and calculated as a difference between them.
- the extraction means 14 extraction process S14
- the mesh whose difference value of the above-mentioned maximum principal stress is larger than a predetermined value is extracted as a fracture site.
- the above calculation (calculation means 12 (calculation step S12)) and extraction (extraction means 14 (extraction step S14)) may be executed in the same computer, or the calculation (calculation means 12 (calculation step S12)) is 1
- the maximum principal stress for each of two or more types of meshes whose mesh division roughness is changed is input to another computer (input means 13 (input step S13)) and extracted.
- extraction means 14 (extraction step S14)) may be executed.
- the input unit 13 and the extraction unit 15 are configured separately from the dividing unit 11 and the calculation unit 12
- Processing can be performed in parallel, and the effect of improving efficiency can be obtained.
- the first mesh roughness and the second mesh roughness are used. After performing at least one of resetting the mesh roughness to a finer roughness and resetting the predetermined value to a smaller value, division and forming analysis (dividing means 11 (dividing step 11) S11)), calculation of the maximum principal stress for each mesh (calculation unit 12 (calculation step S12)), and extraction of a fractured portion (extraction unit 14 (extraction step S14)) are sequentially executed.
- the influence of the averaging of the maximum principal stress in the mesh is reduced, that is, the portion where the stress is concentrated is revealed.
- the difference value between the maximum stress at the first mesh roughness and the maximum principal stress at the second mesh roughness can be obtained larger, so that the fracture site can be predicted more reliably.
- the predetermined value is reset to a smaller value, it is possible to predict, for example, a part where the difference value of the maximum principal stress is not so large as a part having a risk of fracture.
- the end of the analysis target part is divided into a plurality of meshes to perform molding analysis, and in the extracting unit 14 (extracting step S14), One of the end portions is extracted as a fracture site.
- the division is performed so that the roughness of the mesh division is surely changed particularly in the portion where the fracture is predicted. It is necessary that the end portion for which the fracture is predicted be smoothly connected without unevenness in both cases where the mesh division is coarse or dense. It is important to evaluate the stress gradient along the edge in order to reliably predict the fracture at the edge, and the mesh division roughness must change reliably in the direction along the edge. desirable.
- a mesh portion where the difference value of the maximum principal stress for each predetermined mesh is larger than a predetermined value is extracted as a fracture risk portion.
- a finite element method is performed in the dividing step S21 by the dividing means 21. Is used to divide the part with a predetermined mesh roughness and perform molding analysis.
- the maximum principal stress is calculated for each mesh.
- a second calculation step S24 by the second calculation means 24 two or more meshes adjacent to each other are combined to form a combined mesh, and the maximum principal stress is calculated and determined for each combined mesh.
- the computer program causes the central processing unit (CPU) of the arithmetic processing unit (computer) to function as each unit (the dividing unit 21, the first calculating unit 22, the second calculating unit 24, and the extracting unit 25).
- the computer program can be recorded on a computer-readable recording medium, such as a flexible disk or a CD-R.
- This arithmetic processing unit may have an input unit 23 for inputting the maximum principal stress obtained for each divided mesh to another computer.
- the input means 23 a keyboard, a mouse, various digitizers, and the like can be used.
- the input step S23 may be a step of inputting with a keyboard, or the maximum principal stress calculated in the first calculation step S22 is automatically input to the second calculation step 24 in the program. It may be a process (reading data).
- a part obtained by forming a metal plate which is a part to be analyzed, is divided into predetermined meshes (dividing means 21 (dividing step S21))
- a three-dimensional part shape is used.
- the part is expressed as digital data (CAD data or shape measurement data), or when a shell element is used, the part is expressed as a set of two-dimensional plane regions.
- CAD data or shape measurement data digital data
- the corner portion of the part has a large shape change, it is divided by a sufficiently fine mesh to ensure shape reproducibility.
- the mesh is divided so that the outer peripheral line of the part is smooth without unevenness.
- the same molding analysis as in the above embodiment is performed, and the molding process of the entire part is analyzed.
- the first calculation means 22 (first calculation step S22)
- the maximum principal stress for each mesh of interest is calculated.
- the calculation of the maximum principal stress is the same as the calculation in FIG. 1A and FIG. 1B (calculation means 12 (calculation step S12)).
- a calculated value (maximum) for each mesh to be combined is used. Information on the principal stress) and the position (coordinates) of each mesh is required.
- the calculated value for the combined mesh is the arithmetic average of the calculated values for each mesh.
- the position of the coupled mesh may be an arithmetic average of the positions of the meshes, or more simply, the position of the central mesh may be taken over as it is.
- the difference value of the maximum principal stress before and after the joining of the meshes is calculated as the difference value of the maximum principal stress in each mesh by extracting the meshes whose positions are closest to each other before and after the joining of the meshes. Then, an element in which the difference value of the maximum principal stress before and after the mesh connection is larger than a predetermined value is extracted as a fracture site (extraction means 25 (extraction step S25)).
- extraction means 25 extraction step S25
- the method for obtaining the predetermined value is the same as the extraction (extraction means 14 (extraction step S14)) in FIGS. 1A and 1B.
- first calculation first calculation means 22 (first calculation step S22)
- second calculation second calculation means 24 (second calculation step S24)
- first calculation first calculation means 22 (first calculation step S22)
- second calculation second calculation means 24 (second calculation step S24)
- input unit 23 input means 23 (input step S23)
- second calculation second calculation means 24 (second calculation step S24)
- extraction extraction means 25 (extraction step S25)
- the stretch flange crack portion of the press-formed product can be identified as an ultra high strength steel (for example, a tensile strength of 980 MPa). It is possible to estimate at low cost and in a short time even in the case of a high-strength steel sheet. Hereinafter, this point will be described in detail.
- high-tensile steel plates particularly high-strength steel types (for example, high-strength steel plates having a tensile strength of 980 MPa or more) have low elongation, and reach high stress states with a small amount of deformation. For this reason, it is difficult to identify a portion that may crack by using the geometric deformation amount as an index.
- the maximum principal stress which is the amount of mechanical change, is used as an index, the stress value changes greatly even with a small amount of deformation. Is possible.
- Finer mesh division means to calculate and evaluate the plate thickness reduction rate or maximum principal strain value of the strain concentration portion higher than when the mesh division is coarse. Similarly, it means that the value of the maximum principal stress at the stress concentration site is highly evaluated and calculated. From this point of view, it is considered possible to predict the stretch flange crack risk site by using any of the evaluation indices of plate thickness reduction rate, maximum principal strain, and maximum principal stress.
- the accuracy of stress is low in finite element analysis mainly for static explicit method, dynamic explicit method, one-step method, etc., compared to static implicit method that strictly solves the balance state of members.
- the maximum principal stress is not necessarily appropriate as a crack site prediction index.
- the stress is transmitted as a stress wave in the member as a time-dependent wave, and there is a disadvantage that an error from the balanced state is also generated. From this point of view, there is a problem in terms of calculation accuracy in using the stress state as a predictive index of the stretched flange cracking part in the low-strength material.
- Fig. 6 shows the maximum principal strain distribution when a finite element analysis is performed on a high-strength material having the same molding shape and tensile strength of 980 MPa as shown in Fig. 5.
- the difference of the maximum principal strain is recognized in the vicinity of the position 0 (mm), the difference is quantitatively smaller than the case of the low-strength material in FIG. Therefore, it becomes difficult to set a threshold value of ⁇ 2 (in FIG. 4) as to whether stretch flange cracks can occur or not, and it becomes difficult to predict crack sites.
- FIG. 7A is a graph obtained by plotting the maximum principal stress distribution with respect to the position from the analysis result of FIG. The overall stress level is increased by molding, and at first glance, the difference in maximum principal stress in the vicinity of position 0 (mm) appears small. However, from FIG. 7B in which the graph is enlarged, it can be seen that the difference in the maximum principal stress of the peak value is about 100 MPa. If such a difference in the maximum principal stress is obtained, it is possible to set the threshold value of the stretch flange crack estimation index of the high strength material at a significant level. In addition, it is apparent that the technique of the present invention can be applied to a solution that does not guarantee a balanced state of stress values, such as a dynamic explicit method or a one-step method.
- the present invention is particularly suitable for predicting breakage in a metal plate of a high strength and difficult-to-form material.
- a high-strength and difficult-to-form material include high-tensile steel plates, for example, ultra-high-strength steel plates having a tensile strength of 980 MPa or more.
- the application of the present invention is not limited to high-strength steel plates, but can be applied to other high-strength materials such as high-strength aluminum alloys, pure titanium, titanium alloys, and composite materials (metals / resins). Composite materials, dissimilar metal composite materials), and other high-strength materials such as carbon fibers.
- the inventor further advantageously combines the prediction using the difference in the distribution of the maximum principal stress and the prediction using the difference in the thickness reduction rate or the distribution of the maximum principal strain.
- the certainty of prediction can be improved by combining a plurality of predictions.
- a metal plate with high tensile strength and small elongation a metal plate with low tensile strength and large elongation has a large amount of deformation. It is desirable to use the quantity as an index.
- the prediction using maximum principal stress and the maximum principal It is desirable to use both predictions instead of using only one of the predictions using at least one of the strain and the plate thickness reduction rate. That is, by combining a plurality of predictions, it is particularly suitable for prediction using a metal plate having a high strength and low elongation (for example, an ultra-high strength steel plate having a tensile strength of 980 MPa or more) and a maximum principal strain or thickness reduction rate.
- metal plates for example, mild steel plates and aluminum alloy plates
- metal plates having intermediate strength between them for example, high-tensile steel plates having a tensile strength of about 490 MPa to 780 MP. It is possible to extract a fractured part with high prediction accuracy.
- the division is further performed with the first mesh roughness and with the second mesh roughness.
- a shape index value that is at least one of the maximum principal strain and the plate thickness reduction rate is obtained for each mesh, and in the extraction step S14 (extraction means S14), the first mesh in each part of the part is further obtained.
- a difference value between the shape index value in the case of roughness and the shape index value in the case of the second mesh roughness is obtained, the difference value of the shape index value is larger than a predetermined value, and the difference in maximum principal stress A part in the case of the first mesh roughness corresponding to a part satisfying at least one of the values larger than the predetermined value is extracted as a fracture part.
- the shape index value that is at least one of the maximum principal strain and the plate thickness reduction rate is further provided. Is obtained for each mesh, and in the second calculation step S24 (second calculation means 24), the shape index value is obtained for each coupled mesh, and in the extraction step S25 (extraction means 25), the first calculation is further performed. The difference value between the shape index value obtained in the step S22 (first calculation means 22) and the shape index value obtained in the step of the second calculation step S24 (second calculation means 24) is calculated for each part of the part.
- FIGS. 8A to 8D An outline of the adaptive mesh is shown in FIGS. 8A to 8D. Assuming that a strong stress or strain is generated at the center of the line segment AB as a result of applying a strong tensile deformation to the points A and B on the mesh size model as shown in FIG. 8A. In this case, if the simulation proceeds with the initial mesh size, the finite element model may not be able to fully express the concentration of the deformation field.
- an adaptive mesh As a method of avoiding this, a method of dividing the mesh size of a portion where deformation is concentrated so as to become smaller from the middle of the analysis as shown in FIG. 8B is called an adaptive mesh. As deformation and deformation concentration of a specific part progress, the area to which the adaptive mesh is applied is enlarged as shown in FIGS. 8C and 8D. The same adaptive mesh can be applied even if the deformation field is biaxial tension or compression. In the present invention, it is necessary to perform analysis with two different types of mesh roughness once, but when the size and shape complexity of the part to be evaluated are high, the mesh roughness with a small roughness is used. This analysis may require considerable time and analysis costs.
- an adaptive mesh can be used when performing division by the first mesh roughness in the embodiment shown in FIGS. 1A and 1B.
- an adaptive mesh can also be used when performing division with a predetermined mesh roughness.
- the calculation cost can be reduced by applying the present invention after finishing the analysis at an intermediate stage before the bottom dead center. Since stress concentration often starts before the bottom dead center at a site where the risk of rupture is high, it is possible to extract a dangerous site even by such an evaluation by stopping halfway.
- Example 1 The present invention will be described below with examples.
- stretch flange cracking is predicted. Burring molding divided into two pieces was carried out with the mold configuration shown in FIG. Square tube burring was performed with a punch having a square cross section with a side of 40 mm. The corner radius of the punch 13 is 5 mm, and the punch shoulder radius is also 5 mm.
- the base plate is held and fixed from above and below by the die 12 and the plate presser 10.
- the base plates 11A and 11B are obtained by cutting a square plate of 200 mm ⁇ 200 mm, making a rectangular hole in the center thereof by laser cutting, and then cutting the rectangular plate from the center.
- a base plate having a shape as shown in FIGS. 10A and 10B is obtained, and two of them are subjected to burring simultaneously.
- both the two corners R are stretched and subjected to flange deformation, and breakage occurs at one of the edge portions. If no cracking occurs, two molded products having a shape as shown in FIG. 11 are obtained.
- the finite element analysis was performed with two types of mesh roughness using the above punch and base plate shapes.
- the software used a shell element in a dynamic explicit solver in LS-DYNA.
- Two types of mesh roughness of 1.6 mm (see FIG. 10A) and 2.5 mm (see FIG. 10B) were adopted, and the deformation states of the edge portions were compared. The results are shown in Table 1.
- the fracture prediction method of the present invention (the dividing step S11 to the extracting step S14 in FIG. 1B, the dividing step S21 to the extracting step S25 in FIG. 2B, etc.) is applied to the RAM or ROM of the arithmetic processing unit (computer).
- This can be realized by a stored program.
- the program is recorded on a computer-readable storage medium.
- these programs, computer-readable recording media, and arithmetic processing devices (computers) will be described more specifically.
- the program is recorded on a recording medium such as a CD-ROM, or provided to a computer via various transmission media.
- a recording medium for recording the program in addition to the CD-ROM, a flexible disk, a hard disk, a magnetic tape, a magneto-optical disk, a nonvolatile memory card, and the like can be used.
- a program transmission medium a communication medium in a computer network system for propagating and supplying program information as a carrier wave can be used.
- the computer network is a WAN such as a LAN or the Internet, a wireless communication network, or the like
- the communication medium is a wired line such as an optical fiber or a wireless line.
- the program included in the present invention is not limited to the one in which the functions of the above-described embodiments are realized by the computer executing the supplied program.
- a program is also included in the present invention when the function of the above-described embodiment is realized in cooperation with an OS (operating system) or other application software running on the computer.
- OS operating system
- the program is also included in the present invention.
- FIG. 12 is a schematic diagram showing an internal configuration of an arithmetic processing device (personal user terminal device).
- reference numeral 1200 denotes a personal computer (PC) having a CPU 1201.
- the PC 1200 executes device control software stored in the ROM 1202 or the hard disk (HD) 1211 or supplied from the flexible disk drive (FD) 1212.
- the PC 1200 generally controls each device connected to the system bus 1204.
- the program stored in the CPU 1201, the ROM 1202, or the hard disk (HD) 1211 of the PC 1200 implements the procedures of the division step S11 to extraction step S14 in FIG. 1B and the division step S21 to extraction step S25 in FIG.
- the program stored in the CPU 1201, the ROM 1202, or the hard disk (HD) 1211 of the PC 1200 implements the procedures of the division step S11 to extraction step S14 in FIG. 1B and the division step S21 to extraction step S25 in FIG.
- Reference numeral 1203 denotes a RAM which functions as a main memory, work area, and the like of the CPU 1201.
- Reference numeral 1205 denotes a keyboard controller (KBC), which controls instruction input from a keyboard (KB) 1209, a device (not shown), or the like.
- KBC keyboard controller
- Reference numeral 1206 denotes a display controller (DC), which controls display on the display (D) 1210.
- Reference numeral 1207 denotes a disk controller (DKC).
- the DKC 1207 controls access to a hard disk (HD) 1211 and a flexible disk (FD) 1212 that store a boot program, a plurality of applications, an editing file, a user file, a network management program, and the like.
- the boot program is a startup program: a program for starting execution (operation) of hardware and software of a personal computer.
- Reference numeral 1208 denotes a network interface card (NIC) that exchanges data bidirectionally with a network printer, another network device, or another PC via the LAN 1220.
- NIC network interface card
Abstract
Description
本願は、2015年5月18日に、日本に出願された特願2015-101311号に基づき優先権を主張し、その内容をここに援用する。
(1)すなわち、本発明の一態様に係る破断予測方法は、金属板を成形して得られる部品の破断部位を予測する方法であって、前記金属板を、第1のメッシュ粗さで分割した場合と前記第1のメッシュ粗さよりも粗い第2のメッシュ粗さで分割した場合とのそれぞれにおいて、有限要素法を用いて成形解析を行う第1のステップと;前記第1のメッシュ粗さの場合と前記第2のメッシュ粗さの場合とのそれぞれにおいて、最大主応力をメッシュ毎に求める第2のステップと;前記部品の各部位における、前記第1のメッシュ粗さの場合での前記最大主応力と、前記第2のメッシュ粗さの場合での前記最大主応力との差分値を求め、前記差分値が、所定値よりも大きい部位に対応する、前記第1のメッシュ粗さの場合における部位を、前記破断部位として抽出する第3のステップと;を有する。
上記(1)に記載の態様では、第1のメッシュ粗さと第1のメッシュ粗さよりも粗い第2のメッシュ粗さとの二種類のメッシュ粗さを用いる。有限要素法では、各メッシュ内における最大主応力が平均化されて出力される。従って、あるメッシュ内に応力集中部位が存在する場合、より細かい第1のメッシュ粗さの場合における最大主応力は、平均化の影響が小さくなるため、より粗い第2のメッシュ粗さの場合における最大主応力よりも大きくなる。従って、各部位において、二種類のメッシュ粗さにおける最大主応力の差分値を求め、その差分値が所定値よりも大きい場合には、当該部位を応力集中部位とみなすことができる。応力が集中するほど破断が生じる危険度が高くなるので、最大主応力の差分値の大小で破断発生の危険度を予測することが可能になる。
また、従来の一種類のメッシュ粗さのみを用いる場合には、メッシュ粗さが粗い場合には、平均化の影響が強くなり、応力が集中している部位の寄与が平均値に埋もれてしまう。そのため、応力が集中していて破断の危険度が高い部位を抽出するためには、メッシュ粗さを極めて小さく設定しなければならなかった。これに対して、本態様では、第1のメッシュ粗さと第2のメッシュ粗さとのそれぞれの場合における最大主応力の定量的な比較によって破断部位を抽出するので、両者の比較を可能とする程度のメッシュ粗さで十分であり、第1のメッシュ粗さを、従来のように極めて小さいメッシュ粗さに設定することは必須ではない。従って、成形解析を短時間且つ低い計算コストで行うことができる。
しかも、本態様では、指標として最大主応力を用いている。高強度鋼板等の引張強度が高く伸びの小さな金属板の場合、変形量が小さい場合でも、応力は大きく変化する。そのため、板厚減少率又は最大主ひずみという幾何学的変形量を指標としても、変形量自体が小さいため、異なるメッシュ粗さにおける指標の値の差分が不明確になり、破断部位の予測が困難になる。一方、力学的変化量である最大主応力を指標とすることによって、指標の値の差分が明確になり、引張強度が高く伸びの小さな金属板についても、容易且つ確実に破断部位を予測することができる。
この場合、n値に基づいて、第1のメッシュ粗さ及び第2のメッシュ粗さが最適に設定される。従って、粗過ぎるメッシュ粗さによって予測精度を下げることなく、逆に、不必要に細か過ぎるメッシュ粗さを用いることによって計算時間を増大させることもなく、優れた予測精度を得ることができる。
この場合、少なくとも第1のメッシュ粗さをより細かい粗さに再設定することによって、メッシュ中における最大主応力の平均化の影響を小さくして、つまり、応力が集中している部位を顕在化させる。これによって、第1のメッシュ粗さでの最大応力と第2のメッシュ粗さでの最大主応力の差分値がより大きく得られるので、より確実に破断部位を予測することができる。
一方、所定値をより小さい値に再設定する場合には、例えば最大主応力の差分値がそれほど大きくない部位についても、破断が発生する危険性を有する部位として予測することを可能にする。
この場合、成形開始時においては粗いメッシュで解析を行い、成形過程の進行に伴い、変形又は応力変化が大きい部位のみにおいてメッシュを細かく再分割することによって、解析時間を減らすことができる。
また、変形又は応力変化が大きい部位、つまり破断の危険性が高い部位においてメッシュが細かくされるため、予測精度を向上させることができる。
この場合、成形解析を部品の成形途中で止めて、いわゆる途中止め評価を行う。途中止め評価は、成形過程が進行するにつれて、応力状態が引張強度に近くなる前や、変形が過大になり成形解析に不具合が生じる前に、最大主応力の差分値を評価することを可能にする。
この場合、最大主応力を用いた予測に加えて、最大主ひずみ及び板厚減少率の少なくとも一方を用いた予測も行う。複数の予測を組み合わせることによって、予測の確実性を向上させることができる。
また、引張強度が高く伸びの小さな金属板とは逆に、引張強度が低く伸びの大きな金属板においては、変形量が大きいため、板厚減少率又は最大主ひずみという幾何学的変形量を指標とすることが望ましくなる。また、引張強度が高く伸びの小さな金属板と引張強度が低く伸びの大きな金属板との間の中程度の引張強度及び伸びを有する金属板においては、最大主応力を用いた予測と、最大主ひずみ及び板厚減少率の少なくとも一方を用いた予測との一方のみを用いるのではなく、両者の予測を併用することが望ましくなる。つまり、複数の予測を組み合わせることによって、高強度で伸びの低い金属板(例えば、高張力鋼板)や、最大主ひずみ又は板厚減少率を用いた予測に特に適した金属板(例えば、軟鋼板)だけではなく、これらの中間の強度を有する金属板といった多様な種類の金属板について、優れた予測精度で破断部位を抽出することが可能になる。
上記(7)に記載の態様では、上記(1)に記載の態様の二種類のメッシュ粗さを用いた最大主応力の比較の代わりに、メッシュの結合前後での最大主応力を比較することによって、上記(1)に記載の態様と同様に、容易且つ確実に破断部位を予測することができる。
また、成形解析は基本的には1回しか行われないため、成形解析を更に短時間且つ低計算コストで行うことができる。
この場合、上記(4)に記載の態様と同様に、アダプティブメッシュを用いることによって、解析時間を減らすことができ、予測精度を向上させることができる。
この場合、上記(5)に記載の態様と同様に、途中止め評価を行うことによって、応力状態が引張強度に近くなる前や、成形解析に不具合が生じる前に、最大主応力の差分値を評価することを可能にする。
この場合、上記(6)に記載の態様と同様に、最大主応力を用いた予測に加えて、最大主ひずみ及び板厚減少率の少なくとも一方を用いた予測も行うことによって、多様な種類の金属板について優れた予測精度を得ることができる。
この場合、破断として特に問題となる伸びフランジ割れの発生について予測することができる。
この場合、破断が発生する危険性の高い部品端部における破断を予測することができる。
この場合、難成形性の金属板である引張強さが980MPa以上の鋼板について破断を予測することができる。
本発明では、応力勾配のある部位に対して、有限要素法によりメッシュ粗さ(以下、メッシュサイズ又は要素サイズとも言い、相互可換に用いられる)が異なる二種類のメッシュ(要素)(ここでは便宜上、より細かいメッシュ粗さを第1のメッシュ粗さ、より粗いメッシュ粗さを第2のメッシュ粗さとする。)を用いて解析する。有限要素法では、当該メッシュ内の主応力が平均化されて出力される。従って、あるメッシュ内に応力勾配の大きい部位が存する場合、第1のメッシュ粗さの場合と第2のメッシュ粗さの場合とでは、第一のメッシュ粗さにおいて平均値として出力される主応力が、第二のメッシュ粗さにおいて平均値として出力される主応力よりも大きくなる。
この場合、成形解析は基本的には1回しか行われないため、成形解析を更に短時間且つ低計算コストで行うことができる。
手続1:有限要素法等の数値計算手法により応力テンソルの各成分を算出する。
手続2:応力テンソルの各成分は3×3の行列として表現できる。
手続3:応力テンソルの各成分から主応力(σ1、σ2、σ3の3つの数値)を求める。主応力は、応力テンソルの固有値として得られる値である。
手続4:得られた3つの主応力のうち、値が最大の主応力を「最大主応力」として取り扱う。例えばσ1>σ2>σ3の関係にある場合には、σ1が最大主応力と見做される。
コンピュータプログラムは、コンピュータ読み取り可能な記録媒体、例えばフレキシブルディスク、CD-R等に記録され得る。
本演算処理装置は、分割したメッシュ毎に求めた最大主応力を他のコンピュータに入力する入力手段13を有し得る。入力手段として、キーボード、マウス、各種デジタイザ等を使用できる。これに対応して、入力工程S13は、キーボードで入力する工程でも良いし、プログラム内で、算出工程S12で算出した最大主応力を、自動的に抽出工程15に入力する(データを読み込む)工程でも良い。
なお、図1A、図1B及び後述の図2A、図2Bにおいて、実線は必須の手段又は工程を示し、破線は選択的な手段又は工程を示す。
本発明において、有限要素法によりメッシュ分割して解析を行う際には、対象部位の幾何学的形状、即ち例えば端部の曲率や角部の曲率半径等を再現するように、十分細かにメッシュ分割を行う必要がある。更に、本発明において、第1のメッシュ粗さ及び第2のメッシュ粗さの二種類のメッシュ分割で解析を行った後に、第1のメッシュ粗さと第2のメッシュ粗さとで最大主応力の差分値をとるに際して、二種類のメッシュ分割の粗さ(粗及び密)には十分な配慮を行う必要がある。本発明者らは、粗と密のメッシュ分割の大きさの設定方法について鋭意検討し、それが材料の加工硬化特性と関連していることを見出した。材料の加工硬化特性を一般に引張試験により求められるn値により代表させたときに、粗のメッシュ分割の平均的な粗さ(第2のメッシュ粗さ)L coarse(単位はmm)と、密のメッシュ分割の平均的な粗さ(第1のメッシュ粗さ)L fine(単位はmm)が以下の関係を満たすときに、優れた破断予測精度が得られることが判った。
ソリッド要素を使用する場合、以下の式(1A)及び式(2A)のパラメータ調整式で表されるパラメータ範囲で2種類のメッシュ粗さを定めることが望ましい。
f(n; k, 2.0, 0.2) ≦ L coarse ≦ f(n; k, 5.0, 2.0) (1A)
f(n; k, 1.5, 0.2) ≦ L fine ≦ f(n; k, 2.5, 1.5) (2A)
一方、薄板プレス成形で使用頻度の高いシェル要素を使用する場合、初期の板厚をt0[mm]とすると、メッシュサイズがt0以下になることは数値計算上の誤差拡大要因となるため、これを回避する以下の式(1B)及び(2B)の利用が望ましい。
f(n; k, 2.0×t0, 1.5×t0) ≦ L coarse ≦ f(n; k, 5.0, 2.0×t0) (1B)
f(n; k, 2.5×t0, t0) ≦ L fine ≦ f(n; k, 4.0×t0, 2.5×t0) (2B)
f(n; k, L, L0) = (L-L0)×(2/π)×tan-1(k×n)+L0 (3)
ここに、LとL0はそれぞれメッシュサイズ(メッシュ粗さ)の上限値と下限値である。変数kはn値に対するメッシュサイズの変化率を調整するパラメータであり、調査検討した結果50≦k≦100程度の値が適切と考えられる。以下、特に断りが無い場合はk=65の値を採用する。メッシュサイズの範囲を規定する関数f(n; k, L, L0)において(k, L, L0)の3変数は定数として値を定めて使用するため、式(3)の関数fは実質的にn値のみに依存してメッシュサイズを決定する関数として機能する。
以上のように、L coarse(つまり、第2のメッシュ粗さ)及びL fine(つまり、第1のメッシュ粗さ)を設定することにより、粗過ぎるメッシュ粗さによって予測精度を下げることがなくなる。他方、不必要に細か過ぎるメッシュ粗さを用いることによって計算時間を増大させたり、かえって予測精度を下げたりすることもなく、優れた予測精度を得ることができる。
そして、抽出手段14(抽出工程S14)では、上記した最大主応力の差分値が所定値より大きいメッシュを破断部位として抽出する。
ここで、入力手段13及び抽出手段15を、分割手段11及び算出手段12と別装置構成とする場合には、1つのコンピュータで成形解析した結果を元データとして他のコンピュータに入力することにより、処理を並列して行うことが可能となり効率が向上するという効果を得ることができる。
少なくとも第1のメッシュ粗さをより細かい粗さに再設定することによって、メッシュ中における最大主応力の平均化の影響を小さくして、つまり、応力が集中している部位を顕在化させる。これによって、第1のメッシュ粗さでの最大応力と第2のメッシュ粗さでの最大主応力の差分値がより大きく得られるので、より確実に破断部位を予測することができる。
一方、所定値をより小さい値に再設定する場合には、例えば最大主応力の差分値がそれほど大きくない部位についても、破断が発生する危険性を有する部位として予測することを可能にする。
端部の何れかを破断危険部位として抽出するには、上記実施形態と同様に、所定メッシュ毎の最大主応力の差分値が所定値より大きいメッシュの部位を破断危険部位として抽出する。
ここで、上記実施形態と同様に、コンピュータプログラムが、各工程(分割工程S21、第1の算出工程S22、第2の算出工程S24、抽出工程S25)を演算処理装置(コンピュータ)の中央処理装置(CPU)に実行させる。言い換えると、コンピュータプログラムが、演算処理装置(コンピュータ)の中央処理装置(CPU)を各手段(分割手段21、第1の算出手段22、第2の算出手段24、抽出手段25)として機能させる。
コンピュータプログラムは、コンピュータ読み取り可能な記録媒体、例えばフレキシブルディスク、CD-R等に記録され得る。
本演算処理装置は、分割したメッシュ毎に求めた最大主応力を他のコンピュータに入力する入力手段23を有し得る。入力手段23として、キーボード、マウス、各種デジタイザ等を使用できる。これに対応して、入力工程S23は、キーボードで入力する工程でも良いし、プログラム内で、第1の算出工程S22で算出した最大主応力を、自動的に第2の算出工程24に入力する(データを読み込む)工程でも良い。
そして、上記のメッシュの結合の前後における最大主応力の差分値が所定値より大きい要素を破断部位として抽出する(抽出手段25(抽出工程S25))。
所定値の求め方は、図1A及び図1Bの抽出(抽出手段14(抽出工程S14))と同様である。
ここで、入力手段23、第2の算出手段24及び抽出手段25を、分割手段21及び第1の算出手段22と別装置構成とする場合には、1つのコンピュータで成形解析した結果を元データとして他のコンピュータに入力することにより、処理を並列して行うことが可能となり効率が向上するという効果を得ることができる。
すなわち、複数の予測を組み合わせることによって、予測の確実性を向上させることができる。
上述のように、引張強度が高く伸びの小さな金属板とは逆に、引張強度が低く伸びの大きな金属板においては、変形量が大きいため、板厚減少率又は最大主ひずみという幾何学的変形量を指標とすることが望ましくなる。また、引張強度が高く伸びの小さな金属板と引張強度が低く伸びの大きな金属板との間の中程度の引張強度及び伸びを有する金属板においては、最大主応力を用いた予測と、最大主ひずみ及び板厚減少率の少なくとも一方を用いた予測との一方のみを用いるのではなく、両者の予測を併用することが望ましくなる。つまり、複数の予測を組み合わせることによって、高強度で伸びの低い金属板(例えば、引張強度が980MPa以上の超高張力鋼板)や、最大主ひずみ又は板厚減少率を用いた予測に特に適した金属板(例えば、軟鋼板やアルミニウム合金板)だけではなく、これらの中間の強度を有する金属板(例えば、引張強度が490MPa~780MP程度の高張力鋼板)といった多様な種類の金属板について、優れた予測精度で破断部位を抽出することが可能になる。
図8A~図8Dにアダプティブメッシュの概要を示す。図8Aに示すようなメッシュサイズのモデル上において点Aと点Bに強い引張変形を与えた結果線分ABの中央部に強い応力またはひずみが発生するケースを想定する。この場合初期のメッシュサイズのままシミュレーションを進行させると変形場の集中を有限要素モデルが十分に表現できない場合が起こり得る。これを回避する方法として変形が集中する部位のメッシュサイズを図8Bのごとく解析の途中から小さくなるように分割させる方法をアダプティブメッシュと呼ぶ。変形及び特定部位の変形集中が進行すると図8C、図8Dのようにアダプティブメッシュが適用された領域が拡大される。変形場が二軸引張や圧縮であっても同様のアダプティブメッシュ適用が可能である。
本発明においては二種類の異なるメッシュ粗さでの解析を各々一回ずつ実施する必要があるが、評価対象となる部品の大きさや形状複雑性が高い場合には粗さの小さいメッシュ粗さでの解析には相当の時間及び解析コストを要する場合がある。この場合粗さの小さいメッシュ粗さでの解析に代わり粗さの大きいメッシュ粗さでの解析にアダプティブメッシュを適用することにより破断評価対象である変形集中部位のメッシュのみ細分化することが可能である。変形集中部位のみ粗さの小さいメッシュ適用が可能なため大規模解析の実施を回避しつつ本発明における破断予測が可能である。
具体的には、上記図1A及び図1Bに示される実施形態において第1のメッシュ粗さによる分割を行う際に、アダプティブメッシュを用いることができる。
同様に、上記図2A及び図2Bに示される実施形態において所定のメッシュ粗さによる分割を行う際にも、アダプティブメッシュを用いることができる。
正常に完了した解析結果を二種類の粗さのメッシュサイズで得る必要があるため、どちらか一方あるいは両方の解析モデルで正常終了しない場合、本発明が適用できない。
これらの状況を回避するため必ずしも下死点までの解析結果を用いるのではなく、途中止め評価を行い、成形解析の途中段階での応力分布から破断予測を行うことが可能である。また下死点における計算不具合が事前に想定される場合は下死点より手前の途中段階で解析終了させて本発明を適用することにより計算コストの低減も可能である。破断の危険性が高い部位では下死点よりも手前で応力集中が開始している場合が多いため、このような途中止めによる評価でも危険部位の抽出が可能である。
以下に実例を挙げながら、本発明について説明する。
本実施例では、伸びフランジ割れを予測する。
図9に示す金型構成にて2枚に分割したバーリング成形を実施した。1辺40mmの正方形断面を有するパンチで角筒バーリング成形を行った。パンチ13のコーナー半径は5mm、パンチ肩半径も5mmである。ダイ12と板押さえ10で上下から素板を抑えて固定する。素板11A及び11Bは200mm×200mmの正方形形状の板を切り出し、その中央部に矩形の穴をレーザ切断加工で穴空けをした後、矩形板を中央から切断して得たものである。
上述のように、本発明の破断予測方法(図1Bの分割工程S11~抽出工程S14、及び図2Bの分割工程S21~抽出工程S25等)は、演算処理装置(コンピュータ)のRAMやROM等に記憶されたプログラムによって実現できる。当該プログラムは、コンピュータ読み取り可能な記憶媒体に記録される。以下、これらプログラム、コンピュータ読み取り可能な記録媒体、及び演算処理装置(コンピュータ)についてより具体的に説明する。
12 算出手段
13,23 入力手段
14,25 抽出手段
22 第1の算出手段
24 第2の算出手段
Claims (16)
- 金属板を成形して得られる部品の破断部位を予測する方法であって、
前記金属板を、第1のメッシュ粗さで分割した場合と前記第1のメッシュ粗さよりも粗い第2のメッシュ粗さで分割した場合とのそれぞれにおいて、有限要素法を用いて成形解析を行う第1のステップと;
前記第1のメッシュ粗さの場合と前記第2のメッシュ粗さの場合とのそれぞれにおいて、最大主応力をメッシュ毎に求める第2のステップと;
前記部品の各部位における、前記第1のメッシュ粗さの場合での前記最大主応力と、前記第2のメッシュ粗さの場合での前記最大主応力との差分値を求め、前記差分値が、所定値よりも大きい部位に対応する、前記第1のメッシュ粗さの場合における部位を、前記破断部位として抽出する第3のステップと;
を有することを特徴とする破断予測方法。 - 前記第1のメッシュ粗さ及び前記第2のメッシュ粗さを、前記金属板の加工硬化特性を示すn値に基づいて決定する第0のステップをさらに有することを特徴とする請求項1に記載の破断予測方法。
- 前記第3のステップで前記破断部位が抽出されなかった場合に、
前記第1のメッシュ粗さ及び前記第2のメッシュ粗さのうち、少なくとも前記第1のメッシュ粗さをより細かい粗さに再設定すること、及び、
前記所定値をより小さい値に再設定すること、
のうちの少なくとも一方を行った上で、前記第1のステップから前記第3のステップを再度実施することを特徴とする請求項1または2に記載の破断予測方法。 - 前記第1のメッシュ粗さによる分割を行う際に、アダプティブメッシュを用いることを特徴とする請求項1~3のいずれか一項に記載の破断予測方法。
- 前記第1のステップにおける前記成形解析を、前記部品の成形途中で終了させることを特徴とする請求項1~4のいずれか一項に記載の破断予測方法。
- 前記第2のステップにおいて、さらに、前記第1のメッシュ粗さで分割した場合と前記第2のメッシュ粗さで分割した場合とのそれぞれについて、最大主ひずみ及び板厚減少率の少なくとも一方である形状指標値をメッシュ毎に求め;
前記第3のステップにおいて、
さらに、前記部品の各部位における、前記第1のメッシュ粗さの場合での前記形状指標値と、前記第2のメッシュ粗さの場合での前記形状指標値との差分値を求め、
前記形状指標値の差分値が所定値よりも大きいこと、前記最大主応力の差分値が所定値よりも大きいこと、のうちの少なくとも一方を満たす部位に対応する、前記第1のメッシュ粗さの場合における部位を、前記破断部位として抽出する;
ことを特徴とする請求項1~5のいずれか一項に記載の破断予測方法。 - 金属板を成形して得られる部品の破断部位を予測する方法であって、
前記金属板を、所定のメッシュ粗さで分割して、有限要素法を用いて成形解析を行う第1のステップと;
最大主応力をメッシュ毎に求める第2のステップと;
互いに隣接する2以上のメッシュ同士を結合させた結合メッシュ毎に、最大主応力を求める第3のステップと;
前記第2のステップで求めた前記最大主応力と前記第3のステップで求めた前記最大主応力との差分値を、前記部品の部位毎に求め、さらに、前記差分値が、所定値よりも大きい部位に対応する、前記第2のステップでの部位を、前記破断部位として抽出する第4のステップと;
を有することを特徴とする破断予測方法。 - 前記所定のメッシュ粗さによる分割を行う際に、アダプティブメッシュを用いることを特徴とする請求項7に記載の破断予測方法。
- 前記第1のステップにおける前記成形解析を、前記部品の成形途中で終了させることを特徴とする請求項7または8に記載の破断予測方法。
- 前記第2のステップにおいて、さらに、最大主ひずみ及び板厚減少率の少なくとも一方である形状指標値をメッシュ毎に求め;
前記第3のステップにおいて、さらに、前記形状指標値を結合メッシュ毎に求め;
前記第4のステップにおいて、
さらに、前記第2のステップで求めた前記形状指標値と、前記第3のステップで求めた前記形状指標値との差分値を、前記部品の部位毎に求め、
前記形状指標値の差分値が所定値よりも大きいこと、前記最大主応力の差分値が所定値よりも大きいこと、のうちの少なくとも一方を満たす部位に対応する、前記第2のステップでの部位を、前記破断部位として抽出する;
ことを特徴とする請求項7~9のいずれか一項に記載の破断予測方法。 - 破断部位として、伸びフランジ割れの発生部位を予測することを特徴とする請求項1~10のいずれか一項に記載の破断予測方法。
- 前記部品の端部における、前記破断部位を抽出することを特徴とする請求項1~11のいずれか一項に記載の破断予測方法。
- 前記金属板は、引張強さが980MPa以上の鋼板であることを特徴とする請求項1~12のいずれか一項に記載の破断予測方法。
- 請求項1~13のいずれか一項に記載の破断予測方法を実行することを特徴とするプログラム。
- 請求項14に記載のプログラムが記録されていることを特徴とするコンピュータ読み取り可能な記録媒体。
- 請求項14に記載のプログラムを実行することを特徴とする演算処理装置。
Priority Applications (9)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP16796528.4A EP3299793A4 (en) | 2015-05-18 | 2016-05-18 | Breakage prediction method, program, recording medium, and arithmetic processing device |
CA2985456A CA2985456A1 (en) | 2015-05-18 | 2016-05-18 | Breaking prediction method, program, recording medium, and arithmetic processing device |
JP2016555622A JP6176410B2 (ja) | 2015-05-18 | 2016-05-18 | 破断予測方法、プログラム、記録媒体及び演算処理装置 |
KR1020177032614A KR101951587B1 (ko) | 2015-05-18 | 2016-05-18 | 파단 예측 방법, 프로그램, 기록 매체 및 연산 처리 장치 |
BR112017023533-1A BR112017023533A2 (ja) | 2015-05-18 | 2016-05-18 | A fracture prediction method, a program, a recording medium, and an arithmetic processing unit |
MX2017014626A MX2017014626A (es) | 2015-05-18 | 2016-05-18 | Metodo de prediccion de ruptura, programa, medio de grabacion y dispositivo de procesamiento aritmetico. |
RU2017139510A RU2678023C1 (ru) | 2015-05-18 | 2016-05-18 | Способ прогнозирования разрывов, программа, носитель записи и арифметическое обрабатывающее устройство |
CN201680028089.XA CN107532981A (zh) | 2015-05-18 | 2016-05-18 | 断裂预测方法、程序、记录介质以及运算处理装置 |
US15/573,613 US11016011B2 (en) | 2015-05-18 | 2016-05-18 | Breaking prediction method, program, recording medium, and arithmetic processing device |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2015-101311 | 2015-05-18 | ||
JP2015101311 | 2015-05-18 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016186135A1 true WO2016186135A1 (ja) | 2016-11-24 |
Family
ID=57320042
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2016/064753 WO2016186135A1 (ja) | 2015-05-18 | 2016-05-18 | 破断予測方法、プログラム、記録媒体及び演算処理装置 |
Country Status (11)
Country | Link |
---|---|
US (1) | US11016011B2 (ja) |
EP (1) | EP3299793A4 (ja) |
JP (1) | JP6176410B2 (ja) |
KR (1) | KR101951587B1 (ja) |
CN (1) | CN107532981A (ja) |
BR (1) | BR112017023533A2 (ja) |
CA (1) | CA2985456A1 (ja) |
MX (1) | MX2017014626A (ja) |
RU (1) | RU2678023C1 (ja) |
TW (1) | TWI609179B (ja) |
WO (1) | WO2016186135A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111177848A (zh) * | 2019-12-26 | 2020-05-19 | 中国航空工业集团公司西安飞机设计研究所 | 一种基于有限元模型的应变理论值的获取方法和装置 |
WO2020204059A1 (ja) * | 2019-04-01 | 2020-10-08 | 日本製鉄株式会社 | 鋼材の破断予測方法、破断予測装置、プログラム及び記録媒体 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6901008B2 (ja) * | 2017-12-25 | 2021-07-14 | 富士通株式会社 | 画像処理プログラム、画像処理方法、および画像処理装置 |
CN111209702A (zh) * | 2020-01-02 | 2020-05-29 | 中车青岛四方机车车辆股份有限公司 | 一种轨道列车车体强度仿真与试验对标方法及装置 |
CN112699585B (zh) * | 2020-12-29 | 2024-04-09 | 中国航空工业集团公司西安飞机设计研究所 | 一种复合材料厚层压板接头有限元建模方法 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08339396A (ja) * | 1995-04-12 | 1996-12-24 | Nippon Steel Corp | 金属板の変形過程の数値シミュレート結果の処理装置 |
WO2008133092A1 (ja) * | 2007-04-12 | 2008-11-06 | Nippon Steel Corporation | 破断予測方法、演算処理装置、プログラム及び記録媒体 |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5127181B2 (ja) | 1971-12-15 | 1976-08-11 | ||
JPH11191098A (ja) | 1997-12-26 | 1999-07-13 | Nissan Motor Co Ltd | スピニング加工における成形不良発生予測方法 |
JP2000107818A (ja) * | 1998-10-08 | 2000-04-18 | Toyota Motor Corp | 塑性加工シミュレーションの破断判定方法 |
JP4313507B2 (ja) | 2000-08-23 | 2009-08-12 | 新日本製鐵株式会社 | 自動車客室構造部品用高強度鋼板とその製造方法 |
US7286972B2 (en) | 2001-04-17 | 2007-10-23 | Livermore Software Technology Corporation | Implicit-explicit switching for finite element analysis |
JP3988688B2 (ja) | 2003-06-27 | 2007-10-10 | 株式会社大林組 | コンクリートのひび割れ診断装置および方法、コンピュータプログラム、コンピュータ読み取り可能な記録媒体 |
JP4415668B2 (ja) | 2003-12-22 | 2010-02-17 | 日産自動車株式会社 | 成形加工シミュレーションにおける面形状歪み量演算方法及びその装置 |
JP4495623B2 (ja) | 2005-03-17 | 2010-07-07 | 株式会社神戸製鋼所 | 伸びフランジ性および曲げ加工性に優れたアルミニウム合金板およびその製造方法 |
WO2006107066A1 (ja) * | 2005-03-31 | 2006-10-12 | Jfe Steel Corporation | 熱延鋼板、その製造方法および熱延鋼板成形体 |
RU2402010C2 (ru) | 2006-02-01 | 2010-10-20 | Ниппон Стил Корпорейшн | Способ прогнозирования разрушения |
RU2324918C1 (ru) | 2006-12-01 | 2008-05-20 | Государственное образовательное учреждение высшего профессионального образования Тульский государственный университет (ТулГУ) | Способ оценки предельной деформации при локальной листовой штамповке |
CN101788432B (zh) | 2010-01-08 | 2013-09-18 | 上海工程技术大学 | 一种便捷通用多参数金属板料拉深性能楔形试验方法 |
KR101227295B1 (ko) | 2010-04-07 | 2013-01-30 | 신닛테츠스미킨 카부시키카이샤 | 파단 판정 방법, 파단 판정 장치, 프로그램 및 컴퓨터 판독 가능한 기록 매체 |
US9405867B2 (en) | 2012-06-07 | 2016-08-02 | Dassault Systemes Simulia Corp. | Hydraulic fracture simulation with an extended finite element method |
EP3016009B1 (en) | 2013-06-26 | 2021-10-13 | Nippon Steel Corporation | Method for determining bending fracture in metal plate, program, and storage medium |
-
2016
- 2016-05-18 MX MX2017014626A patent/MX2017014626A/es unknown
- 2016-05-18 RU RU2017139510A patent/RU2678023C1/ru active
- 2016-05-18 CA CA2985456A patent/CA2985456A1/en not_active Abandoned
- 2016-05-18 US US15/573,613 patent/US11016011B2/en active Active
- 2016-05-18 TW TW105115333A patent/TWI609179B/zh not_active IP Right Cessation
- 2016-05-18 KR KR1020177032614A patent/KR101951587B1/ko active IP Right Grant
- 2016-05-18 WO PCT/JP2016/064753 patent/WO2016186135A1/ja active Application Filing
- 2016-05-18 CN CN201680028089.XA patent/CN107532981A/zh active Pending
- 2016-05-18 JP JP2016555622A patent/JP6176410B2/ja active Active
- 2016-05-18 EP EP16796528.4A patent/EP3299793A4/en not_active Withdrawn
- 2016-05-18 BR BR112017023533-1A patent/BR112017023533A2/ja not_active Application Discontinuation
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08339396A (ja) * | 1995-04-12 | 1996-12-24 | Nippon Steel Corp | 金属板の変形過程の数値シミュレート結果の処理装置 |
WO2008133092A1 (ja) * | 2007-04-12 | 2008-11-06 | Nippon Steel Corporation | 破断予測方法、演算処理装置、プログラム及び記録媒体 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3299793A4 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020204059A1 (ja) * | 2019-04-01 | 2020-10-08 | 日本製鉄株式会社 | 鋼材の破断予測方法、破断予測装置、プログラム及び記録媒体 |
JPWO2020204059A1 (ja) * | 2019-04-01 | 2021-11-25 | 日本製鉄株式会社 | 鋼材の破断予測方法、破断予測装置、プログラム及び記録媒体 |
JP7052918B2 (ja) | 2019-04-01 | 2022-04-12 | 日本製鉄株式会社 | 鋼材の破断予測方法、破断予測装置、プログラム及び記録媒体 |
CN111177848A (zh) * | 2019-12-26 | 2020-05-19 | 中国航空工业集团公司西安飞机设计研究所 | 一种基于有限元模型的应变理论值的获取方法和装置 |
CN111177848B (zh) * | 2019-12-26 | 2023-05-23 | 中国航空工业集团公司西安飞机设计研究所 | 一种基于有限元模型的应变理论值的获取方法和装置 |
Also Published As
Publication number | Publication date |
---|---|
US20180136100A1 (en) | 2018-05-17 |
TW201706582A (zh) | 2017-02-16 |
BR112017023533A2 (ja) | 2018-07-24 |
CA2985456A1 (en) | 2016-11-24 |
MX2017014626A (es) | 2018-03-01 |
EP3299793A1 (en) | 2018-03-28 |
RU2678023C1 (ru) | 2019-01-22 |
EP3299793A4 (en) | 2018-10-31 |
KR20170136605A (ko) | 2017-12-11 |
JPWO2016186135A1 (ja) | 2017-06-01 |
US11016011B2 (en) | 2021-05-25 |
CN107532981A (zh) | 2018-01-02 |
JP6176410B2 (ja) | 2017-08-09 |
KR101951587B1 (ko) | 2019-02-22 |
TWI609179B (zh) | 2017-12-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6176410B2 (ja) | 破断予測方法、プログラム、記録媒体及び演算処理装置 | |
KR101167764B1 (ko) | 파단 예측 방법, 연산 처리 장치 및 기록 매체 | |
KR101065502B1 (ko) | 파단 예측 방법 | |
JP6314626B2 (ja) | プレス成形性の評価方法、装置、プログラム及びコンピュータ読み取り可能な記憶媒体 | |
JP4621216B2 (ja) | 破断限界取得方法及び装置、並びにプログラム及び記録媒体 | |
JP4980499B2 (ja) | 破断判定方法、破断判定装置、プログラムおよびコンピュータ読み取り可能な記録媒体 | |
JP4814851B2 (ja) | 薄板プレス成形シミュレーションにおける伸びフランジ割れの推定方法 | |
JP6828476B2 (ja) | エッジ部破断予測方法、プログラム及び記録媒体 | |
JPWO2014208697A1 (ja) | 金属板の曲げ破断判定方法、プログラム及び記憶媒体 | |
JPWO2019064922A1 (ja) | 変形限界の評価方法、割れ予測方法及びプレス金型の設計方法 | |
RU2670575C1 (ru) | Способ прогнозирования разрывов, устройство прогнозирования разрывов, программа, носитель записи и способ вычисления критерия распознавания разрывов | |
JP2012033039A (ja) | 材料の曲げ破断予測方法および装置、ならびにプログラムおよび記録媒体 | |
JP6852426B2 (ja) | 成形性評価方法、プログラム及び記録媒体 | |
US10444732B2 (en) | Blank shape determining method, blank, press formed product, press forming method, computer program, and recording medium | |
JP6897413B2 (ja) | 成形性評価方法、プログラム及び記録媒体 | |
JP7110976B2 (ja) | 成形性評価方法、プログラム及び記録媒体 | |
JP7206902B2 (ja) | 成形性評価方法、プログラム及び記録媒体 | |
KR102053370B1 (ko) | 초고장력강판 금형 구조 설계 제작 시스템 및 방법 | |
CN115169027A (zh) | 一种材料安全裕度的预测方法和装置 | |
WO2020174841A1 (ja) | 曲げ割れ評価方法、曲げ割れ評価システム、及びプレス成形部品の製造方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2016555622 Country of ref document: JP Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16796528 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2985456 Country of ref document: CA |
|
ENP | Entry into the national phase |
Ref document number: 20177032614 Country of ref document: KR Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15573613 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: MX/A/2017/014626 Country of ref document: MX |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2017139510 Country of ref document: RU |
|
REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112017023533 Country of ref document: BR |
|
ENP | Entry into the national phase |
Ref document number: 112017023533 Country of ref document: BR Kind code of ref document: A2 Effective date: 20171031 |