Die steel rolling multi-field coupling simulation method, system, device and medium

Abstract

The application discloses a die steel rolling multi-field coupling simulation method, system, equipment and medium, and particularly relates to the technical field of metal plastic forming simulation, and comprises data acquisition, feature analysis, primary processing, secondary processing, coupling simulation and decoupling result output. The application realizes real-time feedback of dynamic recrystallization softening effect to stress solving by constructing two-way coupling conditions of temperature field, stress-strain field and microstructure evolution model; global sensitivity analysis and data preprocessing are introduced to realize input feature dimension reduction, improve solving efficiency and stability; adaptive time step and sequential coupling iteration strategy are adopted to enhance the calculation convergence under complex working conditions; the coupling results are decoupled, a mapping atlas is constructed, and structured data and analysis reports are output. The application realizes high-precision and high-efficiency coupling simulation of multi-physical fields in the die steel rolling process, and can be used for process optimization and quality prediction.

Description

Technical Field

[0001] This invention relates to the field of metal plastic forming simulation technology, and more specifically, to a multi-field coupled simulation method, system, equipment, and medium for die steel rolling. Background Technology

[0002] Hot rolling of die steel is the core plastic forming process for preparing high-quality die steel billets. This process involves the strong coupling evolution of temperature, stress-strain, and microstructure fields, directly determining the dimensional accuracy, microstructure uniformity, and mechanical properties of the rolled billet. Traditional process optimization relies heavily on trial and error and field experience, resulting in high costs, long cycles, and difficulty in precisely controlling the coordinated evolution of multiple physical fields. Finite element numerical simulation, however, has become a key technical means to reveal the rolling mechanism and optimize process parameters.

[0003] Existing multi-field coupled simulation methods for die steel rolling can meet the simulation requirements, but they still have the following technical shortcomings: Incomplete coupling mechanism: The microstructure evolution results are only used as post-processing outputs, and the material softening effect caused by recrystallization is not fed back into the stress-strain field solution, resulting in prediction deviations for macroscopic responses such as rolling force and width spread; The input parameter processing is crude: existing methods directly use all collected process parameters and geometric parameters to establish finite element models, lacking control over the quality of input data and sensitivity analysis, which affects solution efficiency and convergence stability; The solution strategy is singular: existing sequential coupling methods mostly use fixed time steps and fixed iteration order, without dynamically adjusting the solution parameters according to the convergence state of the physical field. In the stage of severe deformation during the rolling process, the fixed step size is prone to iteration divergence or low computational efficiency. The results lack decoupling and knowledge representation: the output of existing methods is mostly in the form of single physical field cloud maps or time history curves, lacking systematic decoupling processing of multi-physical field coupling results; the mapping relationship between process parameters, physical field response and microstructure has not formed a structured data map, which is difficult to directly support process window optimization and product quality control decisions.

[0004] In summary, existing simulation methods for die steel rolling have significant shortcomings in data preprocessing, multi-field bidirectional coupling mechanisms, adaptive solution strategies, and result knowledge representation. There is an urgent need for a multi-field coupled simulation method for die steel rolling that can achieve bidirectional full coupling of temperature field, stress-strain field, and microstructure, and has adaptive solution capabilities. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, the present invention provides a multi-field coupling simulation method, system, equipment, and medium for die steel rolling, which solves the problems mentioned in the background art through the following solutions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-field coupled simulation method for die steel rolling, comprising: S1: Data acquisition, obtaining the initial input data required for the simulation of die steel rolling, including the geometric data of the rolled part and the roll, rolling process parameter data, and material constitutive data; S2: Feature analysis, which involves preprocessing the initial input data and performing parameter sensitivity analysis to remove abnormal data that deviates from the normal distribution range. Based on the sensitivity analysis results, key process parameters that significantly affect the rolling results are selected, and a dimensionality-reduced input feature set is constructed. S3: First processing: Based on the geometric data, a three-dimensional finite element geometric model of the rolling process is established. The three-dimensional finite element geometric model is discretized into a mesh, and the temperature field and mechanical boundary conditions of the rolling area are initialized to generate the finite element model data after the first processing. S4: Secondary processing: Based on the input feature set and the finite element model data after primary processing, construct multi-physics coupling conditions including temperature field control equation, stress and strain field control equation and microstructure evolution model, define the coupling relationship and iterative solution order between multi-physics fields, and generate coupling solution configuration data. S5: Coupled simulation: Input the coupled solution configuration data into the finite element solver, perform transient multi-field coupled iterative solution, solve the displacement field and temperature field in turn in each time increment step, update the mold properties according to the real-time temperature field, repeat the iterative solution until the calculation converges, and obtain the multi-physics field coupled simulation results. S6: Decoupling result output. The coupled simulation results are subjected to field quantity decoupling processing. The single physical field distribution data, process parameter prediction values ​​and microstructure distribution data are extracted respectively. The decoupled data are then used to construct a mapping spectrum between process parameters, physical field response and microstructure and output.

[0007] Preferably, the multi-field coupled simulation system for die steel rolling is characterized by comprising: Acquisition module: configured to acquire the initial input data required for mold steel rolling simulation, including geometric data of the rolled part and rolls, rolling process parameter data, and material constitutive data; Analysis module: Performs data preprocessing and parameter sensitivity analysis on the initial input data acquired by the acquisition module, removes abnormal data that deviates from the normal distribution range, and filters key process parameters that have a significant impact on the rolling results based on the sensitivity analysis results, and constructs an input feature set after dimensionality reduction. Primary processing module: Based on the geometric data, a three-dimensional finite element geometric model of the rolling process is established, the three-dimensional finite element geometric model is discretized into a mesh, and the temperature field and mechanical boundary conditions of the rolling area are initialized to generate the finite element model data after primary processing. Secondary processing module: Based on the input feature set output by the analysis module and the finite element model data after primary processing generated by the primary processing module, construct multi-physics coupling conditions including temperature field control equation, stress-strain field control equation and microstructure evolution model, define the coupling relationship and iterative solution order between multi-physics fields, and generate coupling solution configuration data; Coupled simulation module: Input the coupled solution configuration data generated by the secondary processing module into the finite element solver, perform transient multi-field coupled iterative solution, solve the displacement field and temperature field in turn in each time increment step, update the material properties according to the real-time temperature field, repeat the iterative solution until the calculation converges, and obtain the multi-physics field coupled simulation results. Decoupling output module: Performs field quantity decoupling processing on the multi-physics coupling simulation results obtained by the coupling simulation module, extracts single physical field distribution data, process parameter prediction values ​​and microstructure distribution data respectively, and constructs the above decoupled data into a mapping spectrum between process parameters, physical field response and microstructure and outputs it.

[0008] Preferably, an electronic device includes a processor, a communication bus, a user interface, a network interface, and a memory. The processor executes computer-executable instructions stored in the memory to cause the electronic device to execute the computer program of the above-described multi-field coupling simulation method for die steel rolling.

[0009] Preferably, a computer-readable storage medium is characterized in that a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor as a computer program for the above-described multi-field coupling simulation method for die steel rolling.

[0010] The technical effects and advantages of this invention are as follows: This invention constructs a multi-physics coupling condition for temperature field control equations, stress-strain field control equations, and a JMAK-type dynamic recrystallization microstructure evolution model, and clearly defines the bidirectional coupling relationship between each physical field. In particular, the material softening effect caused by dynamic recrystallization is fed back to the stress-strain field solution in real time by modifying the hardening-softening balance term in the flow stress constitutive model. This invention introduces a variance-based global sensitivity analysis method in S2 to calculate the first-order and total sensitivity indices of each input parameter. Redundant parameters that do not significantly affect the rolling results are eliminated according to a set threshold. At the same time, abnormal data are eliminated by using the 3 principle or box plot method. Continuous parameters are interpolated, filled, and normalized, which improves the stability and computational efficiency of finite element solution. In this invention, an adaptive time step control strategy is adopted in S5. When convergence is difficult, the step size is automatically reduced, and when convergence is good, the step size is gradually increased. In each incremental step, the solution is iteratively solved in the order of displacement field, heat source term, temperature field, microstructure and material properties, which improves the robustness of the solution process and the overall computational efficiency. In this invention, the field quantity decoupling processing of the coupled simulation results is performed in S6, extracting the single physical field distribution data, process parameter prediction values, and microstructure distribution data respectively. The response surface methodology is used to construct a multidimensional quantitative mapping spectrum between process parameters, physical field response, and microstructure. Finally, the results are exported to structured formats such as Excel, PLY, and JSON, and cloud maps, animations, and PDF analysis reports are generated, making the simulation results have good interpretability and engineering applicability. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the overall method of the present invention.

[0012] Figure 2 This is a schematic diagram of the feature analysis structure of S2 in this invention.

[0013] Figure 3 This is a schematic diagram of the primary processing structure of S3 in this invention.

[0014] Figure 4 This is a schematic diagram of the overall system structure of the present invention.

[0015] Figure 5 This is a schematic diagram of the electronic device structure of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figures 1-3 As shown, embodiments of the present invention provide a multi-field coupled simulation method for die steel rolling, including: S1: Data acquisition, obtaining the initial input data required for the simulation of die steel rolling, including the geometric data of the rolled part and the roll, rolling process parameter data, and material constitutive data; This embodiment aims to obtain the initial input data required for multi-field coupling simulation of die steel rolling, laying the data foundation for subsequent finite element modeling and multiphysics coupling solution. Specific steps include: S101: Obtain geometric data of the rolled workpiece and rolls. Geometric data is the foundation for building a three-dimensional finite element model. The specific acquisition process includes: Geometric parameters of rolled parts: Determine the initial dimensions of the die based on the actual rolling conditions; Die dimensions: Measure and record the length of the rolled piece. ,width and height For rolled pieces with complex cross-sections, the coordinates of discrete points of the cross-section profile are obtained using a 3D scanner; Feature definition: Determines the initial mesh division region of the rolled part, including the location of the unstable region at the ends and the stable region in the middle.

[0018] Roll geometry parameters: Roll diameter parameters, which determine the work roll diameter. and roller body length The roll gap is set according to the target reduction amount; the initial roll gap value is set. ; Surface morphology: If the effect of roll roughness on friction needs to be considered, the average deviation of the roll surface profile should be collected. data.

[0019] S102: Obtain rolling process parameter data Process parameters determine the boundary conditions during the simulation process, which are determined by the field data acquisition system. Extract the following key parameters: Speed ​​parameters: Collect roll rotation speed and rolling speed , as the boundary conditions for motion in the finite element model; Temperature parameters: Data collection of initial rolling temperature That is, the initial temperature of the workpiece before entering the first pass, and the preheating temperature of the rolls. Collect ambient temperature , used for heat exchange boundaries in finite element models; Deformation parameters: Collect the reduction amount for each pass. reduction rate and front and back tension If there is no tension during rolling, the tension value is set to zero. Friction parameters: Obtain the rolling contact surface and preset the friction coefficient. The range of values ​​is used to determine the friction coefficient curve.

[0020] S103: Obtain material constitutive data for the mold The accuracy of material constitutive data directly determines the accuracy of stress-strain field calculation. Isothermal compression experiments were conducted using a thermal simulation testing machine to obtain flow stress data of mold steel at different temperatures and strain rates. Experimental design: Set the experimental temperature range, typically covering from room temperature to above the material's phase transition point, with intervals of [missing information]. or Set the strain rate range to cover low to high strain rates; Data acquisition and fitting: Acquire stress-strain curves, perform friction correction and adiabatic temperature rise correction on experimental data, and remove abnormal fluctuation points; A high-temperature rheological stress constitutive model for mold steel is established, and, as a preferred option, a model incorporating a thermal activation mechanism is adopted. Parameter fitting of the hyperbolic sine model: in Indicates strain rate, Indicates flow stress, Indicates material constants, Indicates deformation activation energy, Represents the gas constant, Indicates absolute temperature; All the above data were organized into a standardized format that the finite element software could recognize, and an initial input database was constructed to complete the data acquisition.

[0021] S2: Feature analysis, which involves preprocessing the initial input data and performing parameter sensitivity analysis to remove abnormal data that deviates from the normal distribution range. Based on the sensitivity analysis results, key process parameters that significantly affect the rolling results are selected, and a dimensionality-reduced input feature set is constructed. S201: Data Preprocessing and Outlier Removal Initial input data often contains outliers that deviate from the normal distribution due to measurement errors, sensor noise, or fluctuations in operating conditions. Statistical methods are used to clean the data. Missing value handling: Traverse the initial input database constructed by S1 to identify the missing values ​​of each parameter; for discrete parameters, use mode filling; for continuous parameters, use cubic spline interpolation filling to ensure the continuity of the data sequence.

[0022] Outlier Detection and Removal: Based on In principle, calculate the average value of each process parameter (such as rolling force and rolling temperature). with standard deviation The value falls within Samples outside the specified range are considered errors and are discarded. For non-normally distributed data, box plots are used to extract data exceeding the upper quartile. and lower quartile Data points with a difference of 1.5 times are identified as outliers and are removed or corrected. Data normalization: To avoid the influence of different dimensional parameters in subsequent sensitivity analysis, the dataset after outlier removal is subjected to Min-Max normalization, mapping the features to... The interval, specifically the normalization formula is: in, These are the original parameter values. and These represent the minimum and maximum values ​​of the parameter, respectively.

[0023] S202: Parameter Sensitivity Analysis To screen the process parameters that affect the rolling results, a variance-based global sensitivity analysis method is used to quantify the contribution of each input parameter and output the response. Define input and output: Input parameters include the geometric parameters obtained by S1 (initial height of the rolled piece). ,width Process parameters (roll speed) Pressure reduction Rolling temperature coefficient of friction and material constitutive parameters (deformation activation energy) strain rate ); Output response: Key evaluation indicators for rolling simulation were selected, including maximum rolling separation force. Average temperature at the exit of the rolled piece and the average grain size from the surface to the core .

[0024] Global sensitivity analysis: Based on Monte Carlo sampling, a probability distribution space for the input parameters is constructed, and a generation is performed. 10 sample points, of which For the number of input parameters, Base sample size; The output response value of each sample point is calculated using a finite element solver, and the first-order sensitivity index is calculated. With the total sensitivity index ,in This represents the independent contribution of a single parameter. It represents the combined contribution of the parameter and the interaction of other parameters; Sensitivity threshold setting: Set the overall sensitivity index threshold. Remove The parameters were determined to be redundant parameters that had no significant impact on the rolling results.

[0025] S203: Constructing the dimensionality-reduced input feature set Based on the sensitivity analysis results of S202, the process parameters with significant impact were screened out, and a low-dimensional input feature set was constructed for subsequent finite element modeling. Feature selection: retain the total sensitivity index The parameters are used as key process parameters; for example, if the analysis results indicate that the reduction amount... With rolling temperature Total sensitivity index The coefficients are 0.68 and 0.52 respectively, while the coefficient of friction... If the overall sensitivity index is only 0.03, then retain... and and will Set it to a constant value.

[0026] Feature set construction: The selected key process parameters are integrated with the fixed geometric data and material constitutive parameters in S1 to form the dimensionality-reduced input feature set. The feature set Stored in a structured data format, it contains three types of information: Fixed geometric features: roll diameter, roll length, initial cross-sectional dimensions of the rolled workpiece; Key variable process features: reduction amount, initial rolling temperature, and rolling speed; Intrinsic characteristics of materials: constitutive model parameter matrix, thermophysical property parameter curves; Data output: The constructed input feature set The data is then transferred to S3; simultaneously, logs of the removed abnormal data and sensitivity analysis reports are recorded for reference during subsequent process window optimization.

[0027] S3: First processing: Based on the geometric data, a three-dimensional finite element geometric model of the rolling process is established. The three-dimensional finite element geometric model is discretized into a mesh, and the temperature field and mechanical boundary conditions of the rolling area are initialized to generate the finite element model data after the first processing. S301: Establish a three-dimensional finite element geometric model Dimensionality reduction input feature set constructed based on S2 Using fixed geometric features, a three-dimensional geometric model of the rolling process is established in the finite element preprocessing; Geometric model construction: Geometric modeling of rolled parts based on feature set Initial length of the rolled piece ,high ,width A hexahedral solid model is established; to reduce computational redundancy, 1 / 2 or 1 / 4 of the length of the rolled piece is used for modeling, and symmetric boundary conditions are applied at the symmetry plane.

[0028] Roll geometry modeling: based on feature set The diameter of the rolls in and roller body length Establish a cylindrical solid model; simplify the roll into an elastic body, and select according to the actual analysis requirements. If the main focus is on the deformation of the rolled part, set the roll as an elastic body to reduce the amount of calculation.

[0029] Assembly and contact definition: Assemble the workpiece and rolls along the rolling centerline and set the initial roll gap value. This ensures that there is no initial penetration between the workpiece and the roll; Define the contact pair between the workpiece and the roll, adopt a face-to-face contact algorithm, and set the contact friction model as the modified Coulomb friction model with a friction coefficient of... Taken from feature set The variable process characteristics in the process, such as the friction coefficient If it is removed from S2, then the empirical constant value of 0.3 is taken.

[0030] S302: Mesh Discretization Processing The established three-dimensional geometric model is meshed, and the number of meshes is controlled while ensuring computational accuracy in order to achieve a balance between computational efficiency and accuracy. S3021: Cell Type Selection The rolled piece adopts an eight-node hexahedral linear reduced integral element (C3D8R), which can effectively avoid shear self-locking under rolling conditions with large shear deformation. If the roll is an elastic body, it is discretized using a four-node tetrahedral element (C3D4); if it is a rigid body, the contact surface needs to be meshed.

[0031] S3022: Mesh density control Thickness direction: Divide into sections not less than [number] along the thickness direction of the rolled piece. Layered mesh to accurately capture temperature and strain gradients; Contact Area: The mesh of the contact area between the workpiece and the roll is locally refined. The mesh size of the refined area is [missing information - likely a fraction of the base mesh size]. This ensures the accuracy of the calculations for contact pressure and frictional stress. Transition region: Gradient mesh technology is used to smoothly transition from the refined region to the non-contact region, avoiding computational non-convergence caused by abrupt changes in mesh size.

[0032] S3023: Mesh Quality Inspection After the mesh is divided, check the mesh quality indicators. The Jacobian determinant value should be greater than 0.6, and the warpage should be less than 0.6. The aspect ratio should be controlled within Within; for units that do not meet the quality standards, local re-division is carried out until the requirements are met.

[0033] S303: Initialize the temperature field and mechanical boundary conditions Based on feature set The process parameters and thermophysical parameters are initialized, as are the temperature field distribution and mechanical boundary conditions of the rolling zone. S3031: Temperature field initialization Initial temperature of the rolled piece, based on the feature set Key variable process features in rolling, including initial rolling temperature Assign initial temperature values ​​to all nodes of the rolled piece; The initial temperature of the rolls, based on the feature set Preheating temperature of the rolls Assign initial temperature values ​​to all nodes of the roll; set ambient temperature. It is used for the boundary between subsequent convective and radiative heat transfer.

[0034] S3032: Initialization of mechanical boundary conditions Symmetry constraint: Apply normal displacement constraints on the symmetry plane of the model established using symmetry. Rolled piece constraint: Applying constraints to the tail end face of the rolled piece to limit its rigid body displacement, while allowing free movement along the rolling direction; Roll motion: based on feature set Key variable process features in rolling, rolling speed Or the speed of the rolling mill An angular velocity boundary condition of rotation about the axis is applied to the roll.

[0035] S3033: Initialization of thermal boundary conditions Convection heat transfer: A convective heat transfer coefficient is set on the free surface of the rolled workpiece in contact with air. The typical value range is It should be set according to the actual working conditions; Radiation heat transfer: For high-temperature rolling, radiation heat transfer needs to be considered, and the radiation heat exchange between the surface of the rolled piece and the environment needs to be defined. Contact thermal conduction: The contact thermal conductivity coefficient is defined between the contact pair of the workpiece and the roll. The range of values ​​is This indicates the heat exchange capacity of the contact interface.

[0036] S304: Generate finite element model data after primary processing The geometric model, mesh information, material property assignments, and initial boundary conditions established above are integrated to generate a standard format finite element model data file for secondary processing by S4. Data encapsulation: Export an INP file containing node coordinates, element topology, section properties, and material mapping relationships to ensure the portability of model data across different solvers; Material property assignment: Assign the constitutive data of the mold steel material obtained by S1 fitting to the rolling element, and assign the elastic properties of the roll material to the roll element; Output transfer: The generated finite element model data after one-time processing, including complete geometry, mesh, material, initial field and boundary condition information, is transferred to S4 as the basic carrier for configuring multiphysics coupling conditions.

[0037] S4: Secondary processing: Based on the input feature set and the finite element model data after primary processing, construct multi-physics coupling conditions including temperature field control equation, stress and strain field control equation and microstructure evolution model, define the coupling relationship and iterative solution order between multi-physics fields, and generate coupling solution configuration data. This embodiment takes the finite element model data generated by S3 after one processing step and uses the dimensionality-reduced input feature set constructed by S2. Based on the process parameters and material constitutive data, multi-physics coupling conditions are constructed, including temperature field control equations, stress-strain field control equations, and microstructure evolution models. By defining the coupling relationships and iterative solution order between each physical field, coupling solution configuration data is generated for use by the S5 finite element solver.

[0038] S401: Constructing the temperature field control equations The temperature field is the fundamental physical field in the rolling process, and its distribution affects the material's flow stress and microstructure evolution. Based on Fourier's law of heat conduction, the transient temperature field control equation is established: in Indicates material density, Indicates specific heat capacity, Thermal conductivity as a function of temperature Represents temperature field, Indicates time, The heat source term representing the conversion of plastic deformation work into heat is taken as the plastic work. Converted into heat This represents the heat source term that indicates the conversion of frictional work at the contact interface into heat, and is distributed to the contact surface according to the proportion of frictional power. Boundary condition inheritance: The convective heat transfer, radiative heat transfer, and contact thermal conduction boundary conditions initialized in S3033 are embedded into the governing equations to form a complete temperature field boundary value problem.

[0039] S402: Constructing the governing equations of the stress-strain field The stress-strain field describes the elastoplastic deformation behavior of the workpiece during the rolling process. Based on the large deformation elastoplastic theory, equilibrium equations and geometric equations are established: Equilibrium equations: ,in Represents stress tensor, Represents volume force; Geometric equations: ,in Indicates the reduction rate. Represents displacement field, Indicates transpose; S4021: Constitutive Model Based on the constitutive model of high-temperature rheological stress for mold steel fitted by S103, the Arrhenius hyperbolic sine model is used to describe the relationship between flow stress and temperature, strain, and strain rate: in Indicates equivalent stress, Represents the equivalent rate of change, Indicates material constants, Indicates deformation activation energy, Represents the gas constant, This indicates absolute temperature.

[0040] S403: Constructing a Microorganism Evolution Model Microstructure evolution models are used to predict recrystallization behavior and grain size changes during rolling, and are a key link connecting macroscopic process parameters and material properties. S4031: Dynamic recrystallization model Dynamic recrystallization occurs in die steel during hot rolling, and the volume fraction of dynamic recrystallization is described by a JMAK-type equation. : in For reduction rate, For critical strain, The strain is the strain when the recrystallization volume fraction reaches 50%. , is a material constant.

[0041] S4032: Temperature-Strain Coupling Mechanism The rate of microstructure evolution is jointly controlled by the temperature field and the strain rate field. This is achieved by relating the temperature T to the equivalent strain rate in the above model. The solutions are linked to the temperature field and the stress-strain field respectively, thus achieving coupling between the microstructure model and the macroscopic field.

[0042] S404: Define multiphysics coupling relationships This embodiment clarifies the bidirectional coupling relationship between the temperature field, stress-strain field, and microstructure field, and establishes the transmission path between the field variables. S4041: Thermal and Stress Coupling The effect of stress on heat: work of plastic deformation With friction work As a heat source term input into the temperature field control equation, it causes the temperature of the rolled piece to rise. Influence of heat on stress: The temperature field calculation results are updated in real time to update the material properties in the stress-strain field control equations, including elastic modulus, stress intensity and constitutive model parameters, to realize the simulation of thermal softening effect.

[0043] S4042: Coupling of Heat and Microstructural Evolution The influence of microstructure evolution on heat: The latent heat released during recrystallization acts as an additional internal heat source, superimposed on the temperature field governing equations, resulting in the latent heat of recrystallization. ,in This represents the latent heat of recrystallization per unit volume; The influence of heat on the evolution of microstructure: The temperature field distribution determines the recrystallization kinetic parameters, which directly affect the recrystallization volume fraction and grain size.

[0044] S4043: Stress-Microstructure Evolution Coupling The influence of microstructure evolution on stress: The softening effect caused by dynamic recrystallization is reflected by modifying the hardening-softening balance term in the flow stress constitutive model, by introducing the recrystallization volume fraction into the flow stress expression: in Indicates the current equivalent flow stress of the material, Indicates work hardening stress, Indicates the flow stress of grains after recrystallization. Indicates the volume fraction of dynamic recrystallization. This indicates the volume fraction in which dynamic recrystallization has not occurred. The Influence of Stress on Microstructure Evolution: Cumulative Strain With equivalent change rate As a driving variable in the microstructure evolution model, it determines the initiation and evolution process of recrystallization.

[0045] S405: Define the iterative solution order To achieve a stable solution for the above multi-field coupling relationship, the iterative solution order of each physical field in each time increment step is established; S4051: Solving order setting: A staggered sequential coupling strategy is adopted, with each time increment step... The solution is obtained iteratively in the following order: Displacement field solution: Based on the temperature field of the previous increment step With micro-organization state Update the material constitutive model, solve the equilibrium equations, and obtain the displacement field for the current increment step. Strain field and stress field; Heat source term calculation: Calculate the heat source term based on the obtained plastic work and frictional work. and ; Temperature field solution: Based on the updated heat source term and the temperature field of the previous increment step, solve the transient heat conduction equation to obtain the temperature field of the current increment step. ; Microstructure update: Based on the strain field, strain rate field and temperature field obtained from the solution, calculate the dynamic recrystallization volume fraction and grain size; Material property update: Based on the updated temperature field and microstructure, update the material properties required for the next incremental step.

[0046] S406: Generate coupled solution configuration data The various control equations, coupling relationships, and iterative solution order constructed above are encapsulated into configuration data that the solver can recognize; Configure data format: Generate a text file containing the solution control card, and define the analysis type: transient, thermal, mechanical, microstructure coupling, solution time step, iterative algorithm; Generate a field variable association table to clarify the data transfer path and update frequency between various physical fields; Generate a user subroutine interface to write the temperature field, stress-strain field and microstructure evolution models constructed from S401 to S403 into user subroutines that can be called by the finite element solver, ensuring the accurate execution of the custom constitutive model and microstructure model.

[0047] S5: Coupled simulation: Input the coupled solution configuration data into the finite element solver, perform transient multi-field coupled iterative solution, solve the displacement field and temperature field in turn in each time increment step, update the mold properties according to the real-time temperature field, repeat the iterative solution until the calculation converges, and obtain the multi-physics field coupled simulation results. S501: Solver Initialization and Data Loading Start the finite element solver and load the finite element model data generated in S3 and the coupled solution configuration data generated in S4; Establish a solution environment, allocate CPU computing resources, and set up parallel computing strategies to adapt to the needs of large-scale grid computing.

[0048] S502: Time Increment Step Cyclic Control Set the total simulation time The duration is determined based on the actual rolling process time, and is usually [time value missing]. Second; An adaptive time step control strategy is adopted: the initial time increment step is set to... The time limit is set to 1 second. When convergence is difficult, the step size is automatically reduced; when convergence is good, the step size is gradually increased, with a maximum step size of 1 second. Initialize the time counter t=0 and enter the increment step loop.

[0049] S503: Sequential Iterative Solution within Incremental Steps Increment step at each time Within this process, iterations are performed strictly according to the solution order defined in S405; Displacement field solution: Based on the temperature field T and microstructure state of the previous increment step The material constitutive model is updated via a user subroutine, and the equilibrium equations are solved using an iterative method to obtain the displacement field of the current increment step. Strain field and stress field; Calculation of heat source term: Based on the plastic work and frictional work obtained from the solution, calculate the conversion of the plastic work of the heat source term into heat; Temperature field solution: Substitute the heat source term into the transient heat conduction equation, and use the temperature field of the previous increment step as the initial condition to solve for the temperature field of the current increment step. ; Microstructure update: Based on the strain field, strain rate field and temperature field obtained from the solution, calculate the dynamic recrystallization volume fraction and grain size; Material property update: based on the updated temperature field Based on the microstructure state, update the material properties required for the next incremental step.

[0050] S504: Convergence Judgment and Incremental Step Progression After each iteration of each increment step, calculate the displacement field residual and the temperature field residual: If the displacement field residual (Relative residual) and temperature field residual (Relative residual) Determine if the current increment step has converged, and proceed to the next increment step; If the residual does not meet the convergence criterion and the number of iterations has not exceeded the maximum number of iterations, continue iterating; If the number of iterations exceeds the maximum number of iterations and convergence is still not achieved, the time increment step size is automatically reduced and the current increment step is recalculated. Update time counter Repeat steps S503 to S504 until... .

[0051] S505: Simulation Result Output and Storage During the solution process, the field variable data are recorded at the set output frequency: Nodal field variables: temperature Displacement Equivalent plastic strain, equivalent plastic strain rate; element field variables: stress tensor, strain tensor, dynamic recrystallization volume fraction. Average grain size; global variables: rolling separation force, rolling torque, exit temperature.

[0052] Save the above results as a standard results file to form a complete multiphysics coupling simulation result, and transfer it to S6 for decoupling output.

[0053] S6: Decoupling result output: The coupled simulation results are subjected to field quantity decoupling processing. The single physical field distribution data, process parameter prediction values ​​and microstructure distribution data are extracted respectively. The decoupled data are then used to construct a mapping spectrum between process parameters, physical field response and microstructure and output. This embodiment focuses on the post-processing and application transformation of simulation results. It decouples the coupled simulation results obtained from S5 through analysis, constructing a quantitative mapping spectrum that can be directly used for process optimization. Specifically, it includes: S601: Field decoupling processing, which separates and extracts the spatiotemporal distribution data of each individual physical field from the full result matrix of the coupled solution, including: the cloud map and characteristic curve of the displacement field, equivalent stress field, equivalent strain field, and temperature field of the whole rolling process, and extracts the predicted values ​​of process parameters such as rolling force, rolling torque, billet exit temperature, and width expansion. Microstructure data extraction: Extract the dynamic recrystallization fraction and average grain size distribution data of the entire cross section of the billet after rolling from the coupling results to obtain the spatial distribution law and characteristic values ​​of the microstructure; Mapping spectrum construction: Using the key process parameters obtained from S2 screening as input dimensions and the billet exit temperature, peak rolling force, and average grain size after rolling as output dimensions, Box-Behnken experimental design combined with response surface methodology is used to construct a second-order response surface model between each input parameter and output index, forming a multi-dimensional quantitative mapping relationship, and constructing a queryable mapping spectrum of process parameters, physical field response, and microstructure. S602: Results Output and Archiving Data output format: The decoupled physical field data, predicted process parameters, and microstructure data are exported in a structured format, including: Excel spreadsheet: Quantitative indicators such as predicted values ​​of storage process parameters, statistical values ​​of grain size, and peak rolling force; PLY files store geometric visualization data of temperature field, stress-strain field, and microstructure distribution for use by third-party visualization software; JSON configuration files store the construction parameters and interpolation model coefficients of the mapping map.

[0054] Visualization output: Generates high-resolution images, including temperature field cloud maps, stress field cloud maps, recrystallization volume fraction distribution maps, process parameter response surface maps, and process parameter window maps; generates dynamic animations to show the time evolution of temperature field, stress field, and microstructure during rolling.

[0055] Report generation: Automatically generates simulation analysis reports, summarizing input parameters, key output indicators, mapping graphs, and analysis conclusions, and supports export in PDF format.

[0056] Based on the above method, this application also discloses a multi-field coupled simulation system for die steel rolling, referencing... Figure 4 The multi-field coupled simulation system for die steel rolling shown includes: Acquisition module: configured to acquire the initial input data required for mold steel rolling simulation, including geometric data of the rolled part and rolls, rolling process parameter data, and material constitutive data; Analysis module: Performs data preprocessing and parameter sensitivity analysis on the initial input data acquired by the acquisition module, removes abnormal data that deviates from the normal distribution range, and filters key process parameters that have a significant impact on the rolling results based on the sensitivity analysis results, and constructs an input feature set after dimensionality reduction. Primary processing module: Based on the geometric data, a three-dimensional finite element geometric model of the rolling process is established, the three-dimensional finite element geometric model is discretized into a mesh, and the temperature field and mechanical boundary conditions of the rolling area are initialized to generate the finite element model data after primary processing. Secondary processing module: Based on the input feature set output by the analysis module and the finite element model data after primary processing generated by the primary processing module, construct multi-physics coupling conditions including temperature field control equation, stress-strain field control equation and microstructure evolution model, define the coupling relationship and iterative solution order between multi-physics fields, and generate coupling solution configuration data; Coupled simulation module: Input the coupled solution configuration data generated by the secondary processing module into the finite element solver, perform transient multi-field coupled iterative solution, solve the displacement field and temperature field in turn in each time increment step, update the material properties according to the real-time temperature field, repeat the iterative solution until the calculation converges, and obtain the multi-physics field coupled simulation results. Decoupling output module: Performs field quantity decoupling processing on the multi-physics coupling simulation results obtained by the coupling simulation module, extracts single physical field distribution data, process parameter prediction values ​​and microstructure distribution data respectively, and constructs the above decoupled data into a mapping spectrum between process parameters, physical field response and microstructure and outputs it.

[0057] Please see Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 1000 may include: at least one processor 1001, at least one communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between the above components, and to ensure the stable transmission of data and instructions.

[0058] The user interface 1003 may include a display screen and a touch interaction unit, and optionally also includes a standard wired debugging interface and a wireless Bluetooth interaction interface. The user interface 1003 is used to realize functions such as simulation operator interaction, visualization of geometric and mesh models, configuration of process parameters, simulation status monitoring, display of coupled field cloud maps and microstructure distribution maps, etc. Optionally, it can also be connected to a high-performance graphics workstation to realize post-processing rendering and dynamic animation display of large-scale finite element models.

[0059] The network interface 1004 may include a standard Ethernet interface, an industrial fieldbus interface, a high-speed cluster interconnection interface (such as InfiniBand), or a wireless interface (such as 5G or Wi-Fi). The network interface 1004 is used to realize bidirectional communication between electronic equipment and material databases, process parameter acquisition systems, high-performance computing clusters, and cloud simulation platforms, and to complete the uploading of geometric data, process parameters, and material constitutive data, the submission and scheduling of finite element solution tasks, and the retrieval and archiving of simulation result data.

[0060] The processor 1001 may include one or more processing cores. The processor 1001 is connected to various functional units within the electronic device through various interfaces and lines. By running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005, the processor 1001 executes all the functions and data processing steps of the multi-field coupling simulation method for die steel rolling described in this invention, including but not limited to: parsing of geometric data of rolled parts and rolls, preprocessing and sensitivity analysis of rolling process parameters and material constitutive data, construction and mesh discretization of three-dimensional finite element geometric models, initialization of temperature field and mechanical boundary conditions, construction of multi-physics field coupling conditions of temperature field / stress-strain field / microstructure and definition of iterative solution order, execution of transient multi-field coupling iterative solution, and field quantity decoupling and mapping spectrum output of coupling simulation results.

[0061] Optionally, the processor 1001 can be one or more combinations of a central processing unit (CPU), a graphics processing unit (GPU), a field-programmable gate array (FPGA), and a digital signal processor (DSP). Among them, the CPU mainly handles general computing tasks such as operating system, simulation process control, user interaction program, and data preprocessing; the GPU is used to handle computationally intensive tasks such as parallel computing of finite element meshes, large-scale matrix solving, and real-time rendering of temperature and stress field contour maps; the FPGA can be used to accelerate the numerical calculation and iterative solution process of specific constitutive models.

[0062] The memory 1005 may include random access memory (RAM) or read-only memory; optionally, the memory 1005 may include non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, code, code sets, or instruction sets, and includes a program storage area and a data storage area. The program storage area can store instructions for implementing an operating system, instructions for implementing at least one function, and instruction programs for implementing all the steps of the above-described method embodiments of the present invention. The data storage area can store various types of data involved in the implementation of the method of the present invention, including but not limited to: geometric data of the rolled part and roll, rolling process parameter data, material constitutive data, preprocessed input feature set, three-dimensional finite element geometric model data, mesh discretization data, temperature field and mechanical boundary condition data, multi-physics coupling condition configuration data, coupling simulation result data, decoupled single physics distribution data, process parameter prediction values, microstructure distribution data, and mapping spectrum of process parameters-physical field response-microstructure, etc.

[0063] In the electronic device disclosed in this embodiment, the processor 1001 can call the executable program of the multi-field coupling simulation method for die steel rolling stored in the memory 1005. When the executable program is executed by one or more processors 1001, the electronic device executes all the steps of the multi-field coupling simulation method for die steel rolling as described in any of the above embodiments, so as to achieve high-precision, high-efficiency, and high-reliability multi-field coupling simulation and process parameter optimization of the die steel rolling process.

[0064] An electronic device readable storage medium stores instructions that, when executed by one or more processors, cause the electronic device to perform one or more of the methods described in the above embodiments.

[0065] It should be noted that the computer-readable storage medium may include, but is not limited to, USB flash drives, portable hard drives, magnetic disks, optical disks, solid-state drives (SSDs), embedded read-only memory, random access memory, and various non-volatile memories, as well as other media capable of storing program code. When the computer program product runs on hardware carriers such as electronic devices, high-performance computing clusters, and cloud servers, it can completely execute the entire process of data acquisition, analysis, modeling, coupled solution, and decoupled output of the method embodiments of the present invention, realizing accurate multi-physics simulation of the die steel rolling process, and providing high-precision simulation support for process window optimization, microstructure prediction, and product quality control.

[0066] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-field coupled simulation method for die steel rolling, characterized in that, include: S1: Data acquisition, obtaining the initial input data required for the simulation of die steel rolling, including the geometric data of the rolled part and the roll, rolling process parameter data, and material constitutive data; S2: Feature analysis, which involves preprocessing the initial input data and performing parameter sensitivity analysis to remove abnormal data that deviates from the normal distribution range. Based on the sensitivity analysis results, key process parameters that significantly affect the rolling results are selected, and a dimensionality-reduced input feature set is constructed. S3: First processing: Based on the geometric data, a three-dimensional finite element geometric model of the rolling process is established. The three-dimensional finite element geometric model is discretized into a mesh, and the temperature field and mechanical boundary conditions of the rolling area are initialized to generate the finite element model data after the first processing. S4: Secondary processing: Based on the input feature set and the finite element model data after primary processing, construct multi-physics coupling conditions including temperature field control equation, stress and strain field control equation and microstructure evolution model, define the coupling relationship and iterative solution order between multi-physics fields, and generate coupling solution configuration data. S5: Coupled simulation: Input the coupled solution configuration data into the finite element solver, perform transient multi-field coupled iterative solution, solve the displacement field and temperature field in turn in each time increment step, update the mold properties according to the real-time temperature field, repeat the iterative solution until the calculation converges, and obtain the multi-physics field coupled simulation results. S6: Decoupling result output. The coupled simulation results are subjected to field quantity decoupling processing. The single physical field distribution data, process parameter prediction values ​​and microstructure distribution data are extracted respectively. The decoupled data are then used to construct a mapping spectrum between process parameters, physical field response and microstructure and output.

2. The multi-field coupled simulation method for die steel rolling according to claim 1, characterized in that, The S1 data acquisition specifically includes: S101: Obtain the geometric data of the rolled part and the roll, including the initial length, width, height, and discrete point coordinates of the cross-sectional profile of the rolled part, the roll diameter, roll body length, initial roll gap value, and surface roughness. S102: Obtain rolling process parameter data, including roll speed, rolling speed, initial rolling temperature, roll preheating temperature, ambient temperature, reduction amount for each pass, reduction rate, front and rear tension, and coefficient of friction; S103: The flow stress data of mold steel was obtained through thermal simulation isothermal compression experiments. The high-temperature rheological stress constitutive model was obtained by fitting the Arrhenius hyperbolic sine model and a standardized format recognizable by finite element was constructed.

3. The multi-field coupled simulation method for die steel rolling according to claim 1, characterized in that, The S2 feature analysis specifically includes: S201: Data preprocessing, outliers are removed using principle or box plot method, continuous parameters are filled by cubic spline interpolation, and all parameters are normalized by Min-Max. S202: Global sensitivity analysis based on variance, calculate the first-order sensitivity index of each parameter and the total sensitivity index, set the threshold of the total sensitivity index, and eliminate redundant parameters. S203: Integrate the retained process parameters with fixed geometric data and material constitutive parameters to form a dimensionality-reduced input feature set.

4. The multi-field coupled simulation method for die steel rolling according to claim 1, characterized in that, The S3 process specifically includes: S301: Establish a three-dimensional finite element geometric model of the workpiece and the roll, adopt symmetric modeling, and define the face-to-face contact pairs and the modified Coulomb friction model; S302: Mesh discretization, C3D8R elements are used for the workpiece and C3D4 elements are used for the roll, and the mesh quality is locally refined and controlled in the contact area; S303: Initialize the temperature field, stress boundary and thermal boundary conditions, including initial temperature assignment, symmetry constraints, roll motion boundary, convective heat transfer, radiative heat transfer and contact heat conduction boundary; S304: Encapsulates geometry, mesh, material properties, and boundary conditions to generate standard finite element model data files.

5. The multi-field coupled simulation method for die steel rolling according to claim 1, characterized in that, The S4 secondary processing specifically includes: S401: Constructing the transient temperature field control equations containing plastic heat sources and frictional heat sources based on Fourier's heat conduction law; S402: Based on the large deformation elastoplastic theory, the stress-strain field control equations are constructed, and the Arrhenius hyperbolic sinusoidal constitutive model is used to describe the high-temperature flow stress. S403: Construct a JMAK-type dynamic recrystallization microstructure evolution model, coupled with temperature and strain driving mechanisms; S404: Defines the two-way coupling relationship between the temperature field, stress-strain field, and microstructure field; S405: A sequential coupling strategy is adopted to set the iterative solution order of displacement field, heat source term, temperature field, microstructure and material properties within each time increment step; S406: Encapsulates the control equations, coupling relationships, and solution order, generating coupled solution configuration data and user subroutine interfaces that can be recognized by the solver.

6. The multi-field coupled simulation method for die steel rolling according to claim 1, characterized in that, The transient multi-field coupled iterative solution performed in S5 includes: An adaptive time step control strategy is adopted, which automatically reduces the step size when convergence is difficult and increases the step size when convergence is good. After the iteration of each incremental step, the displacement field residual and the temperature field residual are calculated. When the residual is less than the preset threshold, the current incremental step is determined to have converged.

7. The multi-field coupled simulation method for die steel rolling according to claim 1, characterized in that, The S6 decoupling result output specifically includes: S601: Field decoupling, extracting displacement field, stress-strain field, temperature field, recrystallization volume fraction, and grain size data, and using response surface methodology to construct a process parameter-physical field-microstructure mapping spectrum; S602: Export the results to Excel, PLY, and JSON formats, and generate cloud maps, animations, and PDF analysis reports.

8. A multi-field coupled simulation system for die steel rolling, characterized in that, include: Acquisition module: configured to acquire the initial input data required for mold steel rolling simulation, including geometric data of the rolled part and rolls, rolling process parameter data, and material constitutive data; Analysis module: Performs data preprocessing and parameter sensitivity analysis on the initial input data acquired by the acquisition module, removes abnormal data that deviates from the normal distribution range, and filters key process parameters that have a significant impact on the rolling results based on the sensitivity analysis results, and constructs an input feature set after dimensionality reduction. Primary processing module: Based on the geometric data, a three-dimensional finite element geometric model of the rolling process is established, the three-dimensional finite element geometric model is discretized into a mesh, and the temperature field and mechanical boundary conditions of the rolling area are initialized to generate the finite element model data after primary processing. Secondary processing module: Based on the input feature set output by the analysis module and the finite element model data after primary processing generated by the primary processing module, construct multi-physics coupling conditions including temperature field control equation, stress-strain field control equation and microstructure evolution model, define the coupling relationship and iterative solution order between multi-physics fields, and generate coupling solution configuration data; Coupled simulation module: Input the coupled solution configuration data generated by the secondary processing module into the finite element solver, perform transient multi-field coupled iterative solution, solve the displacement field and temperature field in turn in each time increment step, update the material properties according to the real-time temperature field, repeat the iterative solution until the calculation converges, and obtain the multi-physics field coupled simulation results. Decoupling output module: Performs field quantity decoupling processing on the multi-physics coupling simulation results obtained by the coupling simulation module, extracts single physical field distribution data, process parameter prediction values ​​and microstructure distribution data respectively, and constructs the above decoupled data into a mapping spectrum between process parameters, physical field response and microstructure and outputs it.

9. An electronic device, characterized in that, include: Processor, communication bus, user interface, network interface, and memory: The communication bus is used to enable communication between various components; The user interface is used to realize simulation interaction, model visualization, parameter configuration, simulation monitoring and field quantity cloud map display; The network interface is used to interact with material databases, process acquisition systems, high-performance computing clusters, or cloud platforms; The memory stores an executable program; when the executable program is executed by the processor, the electronic device performs the multi-field coupling simulation method for die steel rolling as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images