A forging crack prediction method based on Deform

Abstract

The present invention discloses a forging cracking prediction method based on Deform, comprising the following steps: Step 1, establishing a damage cracking model: establishing a damage cracking model based on the obtained data on the influence of temperature and strain rate on fracture strain. Step 2, establishing a damage prediction module for predicting the forging hot forming process: writing user-defined variables through the secondary development of the Deform post-processing subroutine, and the established damage values ​​can quantitatively analyze the damage of the forging. Step 3, performing a simulation of damage and cracking of metal material forgings: performing finite element modeling analysis on forging forming, and using the damage module established in step 2 to predict whether fracture occurs during the forging process. The present invention is based on the software Deform, and establishes a damage prediction module through its post-processing secondary development, which can predict the formation of surface cracks of forgings during the forming process, and reduce the negative impact of the formation of surface cracks in the forging processing production process on production efficiency and production quality.

Description

Technical Field

[0001] The invention relates to the technical field of damage simulation of titanium alloy high temperature forging in metal plastic forming, and in particular to a deformation-based damage cracking simulation prediction method. Background Art

[0002] Surface cracks are easily generated during the forging process of materials. The titanium alloy forging process window is narrow. When processing exceeds the process window or severe plastic deformation occurs that exceeds the forming limit of the material, cracks will occur. Cracks often appear on the surface of forgings dominated by tensile stress. After the cracks appear, not only does it take a lot of time to polish, but it also reduces product quality and production efficiency. In severe cases, it will cause the product to be scrapped. Therefore, the prediction of surface cracks is of vital importance to actual production. However, the generation of surface cracks is affected by many factors, and how to achieve accurate prediction is still a difficult problem. Therefore, there is an urgent need for a simple, efficient and practical surface crack prediction and evaluation method.

[0003] With the development of finite element analysis software, more and more technicians use finite element analysis software such as Deform and Abaqus to simulate fracture damage during metal plastic forming. There are multiple toughness fracture criteria set internally. Users need to input relevant material parameters and critical damage coefficients, and the software automatically calculates the damage value through simulation. However, the system's own toughness fracture criterion is one-sided and insufficient to describe the fracture conditions of all forming processes in the field of high-temperature forging. For forging under high-temperature conditions, Deform and Abaqus cannot handle new fracture models well, and the model needs to be modified and then implemented through secondary development.

[0004] Based on the shortcomings of the above methods, the purpose of this method is to provide a simulation prediction method for surface cracking in the forging forming process. The established model is developed through Deform post-processing to establish a prediction module, which is used in forging forming simulation to improve the accuracy of simulation. Summary of the invention

[0005] The present invention aims to solve the problem that surface cracks in the forging forming process cannot be accurately predicted, and provides a simulation method that can improve the prediction accuracy of forging damage values ​​to solve a series of problems caused by surface cracks in engineering practice.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] Step 1: Establish the damage cracking model

[0008] A damage cracking model is established based on the obtained data on the influence of temperature and strain rate on fracture strain.

[0009] Step 2: Establish a damage prediction module for predicting the forging hot forming process

[0010] Through the secondary development of the Deform post-processing subroutine, user-defined variables are written, and the damage values ​​established can be used to quantitatively analyze the damage of forgings in the damage prediction module.

[0011] Step 3: Simulate the damage and cracking of metal forgings

[0012] Finite element modeling analysis is performed on the forging forming, and the damage prediction module established in step 2 is used to predict whether fracture occurs during the forging process.

[0013] The present invention discloses a deformation-based damage cracking simulation prediction method. In step 1, a tensile test of a forged target material under different temperatures, strain rates and stress states is designed to obtain a data set consisting of temperature, load and displacement; then the tensile test is reproduced by finite element simulation to obtain data on the influence of temperature and strain rate on fracture strain and establish a damage cracking model.

[0014] The present invention discloses a deformation-based damage cracking simulation prediction method. In step 1, the shape of the forged target material is designed to be a round bar.

[0015] The present invention provides a deformation-based damage cracking simulation prediction method. In step 1, stress and strain rate can be calculated through load and displacement.

[0016] The damage cracking model involved in step 1 is based on classic ductile fracture models such as Normalized C&L, Oyane and other models, and is corrected by temperature and strain rate.

[0017] The damage cracking model involved in step 1 of the present invention is based on the classic ductile fracture model, which is the Normalized C&L model modified by temperature and strain rate.

[0018] In the present invention, the Zener–Hollomon parameters are first used to integrate temperature and strain rate: Q is the thermal deformation activation energy, and R is a constant equal to 8.314 J mol -1 K -1 ; Establish equations through curve fitting Where a, b, c are constants; the equation It is calculated as temperature and strain rate correction curve / model.

[0019] The above constants a, b, and c are the results of curve fitting and are determined by the relationship between the damage value D and Lnz.

[0020] Preferably, the secondary development involved in step 2 includes:

[0021] Use Fortran language compiler software to open the USR_DEF_PST3 file in the Deform installation folder, and write user-defined variables for the pstusr3.f subroutine. After writing, save and compile to generate the USR_DEF_PST3.dll data sharing file.

[0022] As a further preference, the writing of user-defined variables involved in step 2 specifically includes:

[0023] The variables in the damage model are solved using the basic variables calculated by the Deform software, such as the maximum principal stress, stress triaxiality, etc.

[0024] Name the custom variable and assign an initial value; in actual operation, the initial value is generally assigned to 0.

[0025] The calculation rules of custom variables are formulated according to the final fracture criteria; e.g.

[0026]

[0027] where σ * is the maximum principal stress, is the equivalent stress, ε f is the fracture strain.

[0028] Preferably, when performing finite element analysis on the damage and cracking model established in step 1, the following parameters are analyzed: material constitutive model parameters, simulation control process parameters, mold movement speed and direction, assembly relationship, contact relationship, mesh division number, material initial temperature and other related parameters.

[0029] Preferably, when performing finite element analysis on the damage and cracking model established in step 1, it is necessary to locally refine the surface mesh, and the mesh size does not exceed 3 mm.

[0030] Preferably, the damage and cracking simulation involved in step 3 specifically includes:

[0031] Open the Deform software, open the data sharing file in the "User Variable Window", and then perform tracking calculations.

[0032] Furthermore, the Deform-based forging crack prediction method determines whether the damage value is accumulated. When the strain rate is greater than 0 and satisfies the formula conditions (such as the maximum principal stress or stress triaxiality is greater than 0), the damage value is accumulated, otherwise the current step is 0. The formula is: where σ * is the maximum principal stress, is the equivalent stress, ε f is the fracture strain; the is counted as the damage cracking model.

[0033] When judging whether the damage value accumulates, a strain rate greater than 0 ensures that deformation has occurred, and only when deformation has occurred can damage accumulate; according to the formula definition of the fracture criterion, such as:

[0034] where σ * is the maximum principal stress, is the equivalent stress; this criterion defines that damage accumulation will occur only when the maximum principal stress is greater than 0, because when the maximum principal stress is less than zero (representing compressive deformation), the damage value increment is negative, but in the deformation, it is considered that the damage value is an increasing cumulative process and a negative number cannot be accumulated. Therefore, the step of calculating the damage value as a negative number is defined as 0.

[0035] Preferably, the prediction of whether fracture occurs during the forging process involved in step 3 specifically includes: using 1 as the critical value for the damage value defined in the simulation result. The position where the damage value C of the forging surface is greater than 1 is the high-risk area for cracking, the position where 0.8 < C < 1 is the cracking risk area, and the position where C < 0.8 is the safe area. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is the damage value curve graph of the embodiment;

[0037] Figure 2 is the local forging simulation damage result graph of the embodiment;

[0038] Figure 3 is the schematic diagram of the actual cracking of the local forging of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0039] The present invention will be further described below with reference to the drawings and specific examples.

[0040] Taking the forged Ti-6Al-4V titanium alloy material as an example, the present invention will be described in detail. The process of this simulation method is as follows:

[0041] Step (1) Establish a damage cracking model through a series of experiments. The specific establishment method is as follows: Design tensile tests of the forged Ti-6Al-4V titanium alloy material at different temperatures and strain rates. The specimen is a smooth round bar specimen with a diameter of φ6 × 115 mm. The finite element analysis software Deform performs finite element simulation until the tensile test is reproduced, and obtains the influence data of temperature and strain rate on the damage value D, and establishes a damage cracking model.

[0042] In this embodiment, the test is only a round rod tensile test; the temperature is 400-950°C (specifically 400°C, 500°C, 600°C, 700°C, 750°C, 800°C, 850°C, 900°C, 950°C), the strain rate is 0.01-1S -1 (Specifically 0.01S -1 , 0.1S -1 , 1S -1 ); collect time-load-displacement-temperature data.

[0043] The damage cracking model involved in step 1 of this embodiment 1 is based on the classic ductile fracture model, which is the Normalized C&L model modified by temperature and strain rate.

[0044] In this example, the temperature and strain rate are first integrated using the Zener-Hollomon parameters: Q is the thermal deformation activation energy, and R is a constant equal to 8.314 J mol -1 K -1 In this embodiment, Q = 546.575 kJ / mol; the equation is established by curve fitting

[0045] Step (2) Establishing a damage prediction module for predicting the forging hot forming process

[0046] (2.1) User-defined variables are written through the secondary development of the Deform post-processing subroutine, and the established damage values ​​can be used to quantitatively analyze the damage of forgings. The specific establishment method is as follows: Use the Fortran language compiler software to open the USR_DEF_PST3 file in the Deform installation folder, and write the pstusr3.f subroutine. First, write a program to solve the relevant parameters in the damage cracking model (such as maximum principal stress, stress triaxiality, etc.), and then name the custom variables that need to be visualized (such as damage values, etc.) and assign initial values ​​(initial value is 0). Finally, according to the established damage cracking model, the calculation rules for the custom variables are set. The final correction model is (where σ * is the maximum principal stress, is equivalent stress). After completion, save and compile to generate USR_DEF_PST3.dll data sharing file.

[0047] (2.2) Simulate the damage and cracking of metal forgings: Use Deform software to perform finite element modeling calculations on forgings, where the surface mesh needs to be refined and the mesh size does not exceed 3 mm. After the calculation is completed, open the post-processing, select the compiled USR_DEF_PST3.dll data sharing file through the user variable window, and select the node as the target for custom variable tracking. By tracking the calculated damage value, the damage of the forging can be quantitatively analyzed.

[0048] Step 3: Simulate the damage and cracking of metal forgings

[0049] Finite element modeling analysis is performed on forgings, and the damage prediction module established in step 2 is used to predict whether fracture occurs during the forging process. The finite element modeling of forgings involved in step 3 includes the following parameters: material constitutive model parameters, simulation control process parameters, die movement speed and direction, assembly relationship, contact relationship, mesh division number, material initial temperature and other related parameters.

[0050] The specific operations are as follows: open the Deform software, open the data sharing file in the "User Variable Window", and then perform tracking calculations.

[0051] Furthermore, it is determined whether the damage value is accumulated. When the strain rate is greater than 0 and satisfies the formula conditions (such as the maximum principal stress or stress triaxiality is greater than 0), the damage value is accumulated, otherwise the current step is 0. The formula is: where σ * is the maximum principal stress, is the equivalent stress, ε f is the fracture strain; It is considered as the damage cracking model.

[0052] When judging whether the damage value is accumulated, the strain rate is greater than 0 to ensure that deformation has occurred. Only when deformation occurs can damage be accumulated; according to the formula definition of the fracture criterion, such as:

[0053] where σ * is the maximum principal stress, is the equivalent stress; this criterion defines that damage accumulation will occur only when the maximum principal stress is greater than 0, because when the maximum principal stress is less than zero (indicating compression deformation), the damage value increment is negative, but in deformation, the damage value is considered to be an increasing accumulation process and cannot accumulate a negative number. Therefore, the step of calculating the damage value as a negative number is defined as 0.

[0054] Step 4. Analyze the custom variables calculated by tracking: The damage value is used to determine whether cracking occurs during the forging process and the location where the cracking occurs. In the simulation results, the damage value takes 1 as the critical value. The position on the forging surface where the damage value C > 1 is the high-risk area for cracking, the position where 0.8 < C < 1 is the cracking risk area, and the position where C < 0.8 is the safe area.

[0055] Taking the Ti-6Al-4V titanium alloy special-shaped part as the forging target product, the prediction map obtained by the above prediction method is as Figure 1 shown, and there is an area where C > 0.8 in Figure 1 , it is judged that there is a cracking risk in this area. Through the forging experiment, the finished product is obtained (as Figure 2 shown), and in the area where cracking risk is predicted in Figure 1 , obvious cracking appears at the corresponding position in Figure 2 . This proves the accuracy of the prediction of the present invention.

[0056] In the above embodiment, the secondary development code is introduced:

[0057] When there is a letter C in front of each line of code in the software, it represents that this line is a comment line and does not participate in compilation. The following is the annotation of the main part. The code part is indicated before the code, and the code explanation is indicated before the code explanation.

[0058] Code:

[0059] SUBROUTINE USRVAR(TNOW,DTMAX,RZ,TEMP,EFEPS,TEPS,EFSTS,DAMG,

[0060] $RDTY,STS,EPS,TSR,WEAR,VAR1,VAR2,VNAME,ISTEP,INIT,

[0061] $IOBJ,NUMNP,NUMEL,ICURNE,NOUTINE)

[0062] Code explanation:

[0063] SUBROUTINE USRVAR: Establish a subroutine and name it

[0064] (): The variables in the parentheses are the basic variables provided for subsequent calculations

[0065] $: Represents a continuation line. Each line can only have 60 characters. For the excess part, add $ on the next line for continuation.

[0066] The basic variables represented by the characters in the parentheses:

[0067] C TNOW: Current time

[0068] C DTMAX:Max time step when set in simulation controls TNOW betweensuccessive steps can also be used to compute time step size.

[0069] C RZ:Element center coord.

[0070] C TEMP:Temperature

[0071] C EFEPS:Effective strain rate

[0072] C TEPS:Total accumulated strain

[0073] C EFSTS:Effective stress

[0074] C DAMGE:Damge factor

[0075] C RDTY:Relative density

[0076] C STS:Stress tensor

[0077] C EPS:Strain rate tensor

[0078] C TSR:Strain tensor

[0079] C WEAR(5):WEAR(1)=Interface temperature on master object nodes

[0080] C WEAR(2)=Sliding velocity on master object nodes

[0081] C WEAR(3)=Interface pressure on master object nodes

[0082] C VAR1(1-95):Initial state variables

[0083] C VAR2(1-95):Updated state variables

[0084] C VNAME(1 - 95): Name for each variables

[0085] C ISTEP: Step number

[0086] C INIT: Flag for Different Operations

[0087] C 0 - Define characteristic of the subroutine

[0088] C 1 - Initialize User Defined Variables(Initial Step)

[0089] C 2 - Calculate User Defined Variables(Subsequent Steps)

[0090] C IOBJ: Object number in current object

[0091] C NUMNP: Number of nodes in current object(when INIT=1,2)

[0092] C NUMEL: Number of elements in current object(when INIT=1,2)

[0093] C ICURNE: Current node / element number in object(when INIT=1,2)

[0094] C(depending on tracking at node or element)

[0095] Code:

[0096] IMPLICIT DOUBLE PRECISION(A - H,O - Z),INTEGER(I - N)

[0097] CHARACTER*24 VNAME(95)

[0098] Code explanation for DIMENSION RZ(3), STS(6), EPS(6), TSR(6), VAR1(95), VAR2(95), WEAR(5):

[0099] IMPLICIT DOUBLE PRECISION: Defines double-precision variables. The variables can retain values after the decimal point. Variables starting with A - H, O - Z are double-precision variables.

[0100] INTEGER: Integer variables. Variables starting with the letters I - N are integer variables.

[0101] CHARACTER*24 VNAME(95): Defines a character array that stores variable names.

[0102] DIMENSION: Defines an array that stores a series of variables. The number represents the number of variables inside. The above code is自带 in the file. The following is what needs to be written in the embodiments of the present invention.

[0103] Code:

[0104] PARAMETER(NM = 3)

[0105] DIMENSION A(NM, NM), C(9), W(NM), FV1(NM), FV2(NM)

[0106] B = ( / STS(1), STS(4), STS(6), STS(4), STS(2), STS(5), STS(6),

[0107] + STS(5), STS(3) / )

[0108] A = RESHAPE(C, ( / 3, 3 / ))

[0109] CALL TRED1(3, 3, A, W, FV1, FV2)

[0110] CALL TQLRAT(3, W, FV2, IERR)

[0111] IF(IERR.NE.0) GO TO 110

[0112] Code explanation:

[0113] The main function of this part of the code is to solve for three principal stresses through the partial stress tensor stored in the STS(6) array. The calculated principal stresses are stored in the W(3) array, where W(1) < W(2) < W(3). Two external functions, TRED1 and TQLRAT, are used.

[0114] Code:

[0115] Z=EFEPSE*EXP(546575 / (8.314*(TEMP+273)))

[0116] O=LOG(Z)

[0117] P=0.84068+29078.4*EXP((-1)*O / 5.4523)

[0118] Code Explanation

[0119] This part is based on the data obtained from the experiment, and compiles the functional relationship between critical fracture strain, temperature and strain rate P = f (lnz) = f (T, ε). EFEPSE strain rate, TEMP temperature.

[0120] Code:

[0121] IF(INIT.EQ.0)THEN

[0122] VNAME(1)='TNOW'

[0123] VNAME(2)='INCC'

[0124] VNAME(3)='C'

[0125] VNAME(4)='User Example-4'

[0126] VNAME(5)='User Example-5'

[0127] RETURN

[0128] ENDIF

[0129] Code explanation:

[0130] This part is to name the custom variables that we will eventually display in the post-processing interface. There are three of them: TNOW, INCC, and C. TNOW represents the current time. This variable is mainly used to calculate the strain increment. Strain increment = strain rate × (current time - previous time). INCC represents the damage increment. Damage increment = strain increment × damage criterion formula. C represents the cumulative damage value.

[0131] Code:

[0132] IF(INIT.EQ.1)THEN

[0133] VAR2(1)=0.

[0134] VAR2(2)=0.

[0135] VAR2(3)=0.

[0136] VAR2(4)=0.

[0137] VAR2(5)=0.

[0138] RETURN

[0139] ENDIF

[0140] Code explanation:

[0141] This part is to initialize the custom variables. Generally, they are all set to 0.

[0142] Code:

[0143] IF(INIT.EQ.2)THEN

[0144] IF(TNOW.GT.0)THEN

[0145] VAR2(1)=TNOW

[0146] ELSE

[0147] VAR2(1)=VAR1(1)

[0148] ENDIF

[0149] IF(EFSTS.GT.0)THEN

[0150] VAR2(2)=W(3)*(TNOW-VAR1(1))*EFEPS / EFESTS / P

[0151] ELSE

[0152] VAR2(2)=0

[0153] ENDIF

[0154] IF(W(3).GT.0)THEN

[0155] VAR2(3)=VAR1(3)+VAR2(2)

[0156] ELSE

[0157] VAR2(3)=VAR1(3)

[0158] ENDIF

[0159] ENDIF

[0160] Code explanation:

[0161] This part calculates the custom variables. TNOW only serves to store the previous time of the current time, where VAR1(1) represents the previous time and VAR2(1) represents the current time. INCC represents the damage increment. Taking the C&L criterion as an example, P represents the correction curve, and ε represents the strain increment. The calculation premise is that the equivalent stress EFESTS as the dividend cannot be 0, otherwise the calculation result will be wrong, so there is an IF statement (IF(EFSTS.GT.0)THEN) to calculate if EFSTS is greater than 0, and skip this step if it is less than or equal to 0. GT represents the greater than symbol. D represents the cumulative damage value, and it is only necessary to accumulate the damage increase. However, for the C&L criterion, damage accumulation is performed only when the maximum principal stress is greater than 0, so there is an IF statement (IF(W(3).GT.0)THEN) to calculate if the maximum principal stress W(3) is greater than 0, and skip this step if it is less than or equal to 0. At this point, the code is written.

Claims

1. A forging cracking prediction method based on Deform, characterized in that: It includes the following steps: Step 1: Establish a damage cracking model Establish a damage cracking model based on the obtained data of the influence of temperature and strain rate on fracture strain; Step 2: Establish a damage prediction module for predicting the forging hot forming process Write user-defined variables through secondary development of the Deform post-processing subroutine, and establish a damage prediction module that can quantitatively analyze the damage of forgings; The writing of user-defined variables involved in Step 2 specifically includes: Solve the variables in the damage model with the basic variables calculated by Deform software, and the variables include the maximum principal stress and stress triaxiality; Name the user-defined variables and assign initial values; the initial value is assigned as 0; The calculation rules of user-defined variables are formulated according to the finally determined fracture criterion; including: ； in is the maximum principal stress, is the equivalent stress, is the fracture strain; Step 3: Conduct simulation of damage cracking of metal material forgings Conduct finite element analysis on the forging forming, and use the damage prediction module established in Step 2 to predict whether fracture occurs during forging; The damage cracking simulation involved in Step 3 specifically includes: Open the Deform software, open the data sharing file in the "User Variable Window", and then conduct tracking calculations; At the same time, it is determined whether the damage value is accumulated. When the strain rate is greater than 0 and meets the formula conditions, the damage value is accumulated, otherwise the current step is 0; the formula is: ,in is the maximum principal stress, is the equivalent stress, It is considered as damage cracking model; is the fracture strain; When judging whether the damage value accumulates, a strain rate greater than 0 ensures that deformation has occurred, and only when deformation occurs can damage accumulate; according to the formula definition of the fracture criterion: ,in is the maximum principal stress, is the equivalent stress; this criterion defines that damage accumulation will occur only when the maximum principal stress is greater than 0, because when the maximum principal stress is less than zero, the damage value increment is negative, but in deformation, the damage value is considered to be an incremental accumulation process and cannot accumulate a negative number, so the step of calculating the damage value as a negative number is defined as 0; The prediction of whether fracture occurs during forging involved in Step 3 specifically includes: in the simulation results, the defined damage value takes 1 as the critical value, the position where the surface damage value C of the forging is greater than 1 is the high-risk area of cracking, the position where 0.8 < C < 1 is the risk area of cracking, and the position where C < 0.8 is the safe area.

2. The method for predicting forging cracking based on Deform according to claim 1, characterized in that: In Step 1, design tensile tests of the forged target material under different temperatures, strain rates and stress states to obtain a data set composed of temperature, load and displacement; then reproduce the tensile test through finite element simulation, and obtain the data of the influence of temperature and strain rate on fracture strain based on this and establish a damage cracking model.

3. The method for predicting forging cracking based on deform according to claim 1, characterized in that: The damage cracking model involved in Step 1 is modified from a classical ductile fracture model through temperature and strain rate; the classical ductile fracture model is selected from one of the Normalized C&L model and the Oyane model.

4. The method for predicting forging cracking based on Deform according to claim 1, characterized in that: First, the temperature and strain rate are integrated using the Zener–Hollomon parameters: , Q is the thermal deformation activation energy, and R is a constant equal to 8.314 J mol -1 K -1 ; Establish the equation D=f(lnz) through curve fitting , where a, b, and c are constants.

5. The method for predicting forging cracking based on deform according to claim 1, characterized in that; The secondary development involved in Step 2 includes: Open the USR_DEF_PST3 file in the Deform installation folder with a Fortran language compilation software, write user-defined variables for the pstusr3.f subroutine, save and compile it after completion to generate the USR_DEF_PST3.dll data sharing file.

6. The method for predicting forging cracking based on Deform according to claim 1, characterized in that: In Step 3, conduct finite element modeling on the forging forming and locally refine the surface mesh; the mesh size does not exceed 3mm.

Images