Online high-precision rod wire organization performance prediction and production method

Abstract

An online high-precision rod wire organization performance prediction and production method, comprising: selecting historical production data matched with the current rod wire production process, determining the optimal prediction model based on the matched historical production data, simulating the production rod wire based on the prediction model, and simulating the temperature field, organization and performance parameters; judge whether the error of the simulated temperature field, organization and performance parameters and the measured value of the production rod wire is within the allowable range, if yes, backup the corresponding production data and prediction model into the database, if not, optimize the model parameters combined with the current production data until the error is within the allowable range; based on the optimized prediction model, simulate the production data of the current rod wire, simulate the temperature field, organization and performance parameters, and judge whether the simulated organization performance meets the national production standard and the expected value of the factory, if yes, roll.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hot rolling, and in particular to an online high-precision rod wire microstructure and performance prediction and production method. BACKGROUND

[0002] In the current global manufacturing industry, the steel industry as the core of the basic industry, its production technology and quality control level directly affects the development of the entire manufacturing industry and the industrial security of the country. With the continuous change and improvement of market demand, the performance standards of steel products are also becoming more and more strict, especially the microstructure and performance of rod wire, which is directly related to the final mechanical properties and application range of the product. The traditional production method of rod wire often relies on experience adjustment and post-detection, which is not only low in efficiency, but also difficult to meet the rapidly changing market demand and strict quality standards.

[0003] In recent years, the rapid development of information technology and intelligent manufacturing has brought new opportunities for the steel production, providing new ideas and methods for improving production efficiency and product quality. Through these technologies, real-time data monitoring and analysis during the production process can be realized, so as to predict and adjust the production parameters and optimize the product performance. Under such background, it is particularly important and urgent to develop an online high-precision rod wire microstructure and performance prediction technology and its production process. Based on the real-time collected production data, the advanced prediction model can accurately predict the microstructure and performance of the rod wire under different production data. Such prediction not only can improve the transparency of the production process, but also can foresee potential quality problems in the early stage of production, so as to make adjustments in advance and ensure product quality.

[0004] The commonly used microstructure and performance prediction technology is mainly based on physical metallurgical model or big data model, and is mainly an offline prediction system, and there is no online prediction technology considering the state of the billet (heating furnace out of billet or direct rolling), rod and wire rolling. SUMMARY

[0005] In view of the technical defects and technical disadvantages in the prior art, the embodiments of the present application provide an online high-precision rod wire microstructure and performance prediction and production method to overcome the above problems or at least partially solve the above problems, and the specific scheme is as follows:

[0006] An online high-precision rod wire microstructure and performance prediction and production method, the method comprises:

[0007] Step 1, selecting historical production data with high matching degree for the production process of rod wire, determining the optimal prediction model based on the historical production data, and performing simulation calculation based on the prediction model to simulate the temperature field, microstructure and performance parameters;

[0008] Step 2, judge whether the simulated temperature field, organization and performance parameters are within the allowable range of the measured values of the production, if yes, consider the prediction model as a matched model, backup the corresponding production data and the prediction model into the database, if not, optimize the model parameters combined with the current production data until the error is within the allowable range;

[0009] Step 3, based on the optimized prediction model, simulate the production data of the current bar and wire, simulate the temperature field, organization and performance parameters, and judge whether the simulated organization performance meets the national production standards and the expected value of the factory, if yes, then roll.

[0010] Further, the history production data determines the optimal prediction model, which is to obtain the adjustment factor of the prediction model, and when the adjustment factor of the prediction model and the history production data reach a preset matching degree, the corresponding prediction model is considered as the optimal prediction model.

[0011] The adjustment factor of the prediction model includes the chemical composition of the material, the rolling specification, the rolling program table, the rolling process temperature, the cooling water tank parameter, the air cooling line parameter, etc.; the production data also includes the chemical composition, the rolling specification, the rolling program table, the rolling process temperature, the cooling water tank parameter, the air cooling line parameter, etc., for example, the chemical composition error is within ±3%, the rolling specification is the same, the rolling program table hole type system error is within ±3%, the rolling process temperature error is within ±5℃, the cooling water tank parameter error is within ±3%, and the air cooling line parameter error is within ±5%, then the adjustment factor of the prediction model and the history production data are considered to reach a preset matching degree.

[0012] Further, the prediction model is based on a physical metallurgical model.

[0013] Further, the prediction model includes a temperature field model, an organization evolution model, and an organization and mechanical property relationship model.

[0014] Further, based on the temperature field model, the temperature field parameters are simulated, which specifically include:

[0015] In the simulation of the whole process temperature field, the temperature field model first performs billet modeling and two-dimensional grid division, defaults the rolling direction, grid number, radial and grid number, and adjusts the grid division according to the actual iteration before simulation to improve the simulation calculation efficiency; in the simulation, the appropriate time step is determined through calculation to ensure the convergence of the calculation, and the corresponding material thermal physical property parameters are imported from the database, the current heat transfer model, initial condition and boundary condition are selected based on the process position of the incremental step, the temperature field from the heating furnace or continuous casting outlet to the finished product collection area is calculated, after completion, the temperature field simulation result is saved and output.

[0016] Further, based on the organization evolution model, simulation calculation is carried out to simulate the organization parameters, specifically including:

[0017] The simulated temperature field parameters are taken as initial conditions, rolling process parameters are introduced, and according to the rolling process parameters, the strain ε, strain rate rolling temperature T, action time t of each stage and initial size d0 of austenite before rolling, at a certain pass rolling, the organization evolution model first calculates the critical strain value ε c and the current cumulative strain value ε * When ε * ≥ ε c , dynamic recrystallization occurs, and the dynamic recrystallization grain size d d and the recrystallization percentage X d When ε * ≤ ε c , static recrystallization occurs, and the static recrystallization grain size d s and the recrystallization percentage X s At the interpass gap, the organization evolution model first judges whether the recrystallization percentage X d or X s is ≥ 0.95, if yes, the static recrystallization grain growth d SG and the sub-dynamic recrystallization grain growth d MG after the interpass recrystallization is calculated, if not, the static recrystallization grain growth d SG and the sub-dynamic recrystallization grain growth d MG after the incomplete interpass recrystallization is calculated, after the grain growth calculation is completed, the organization evolution model calculates the average grain size, and outputs and saves the recrystallization occurrence conditions;

[0018] The organization evolution model will calculate the first pass to the last mill pass in a loop, and after the loop is completed, the average grain size of the austenite before cooling and phase change, the temperature field and the residual strain are taken as the initial conditions of the phase change and are output and saved.

[0019] Further, the method further comprises: in the cooling stage after rolling is completed, if the average temperature of the intermediate blank is higher than the phase equilibrium temperature Ae3, the austenite grain will continue to grow, at this time, the organization evolution model first calculates the grain size growth based on the whole process temperature field by using the superposition method, after the average temperature of the intermediate blank reaches the Ae3 temperature, the austenite isocold decomposition starts, the phase transition incubation period model is used to determine the starting temperature of the start of each phase transition, the volume fraction of each phase transition is calculated by using the continuous isothermal superposition method based on the phase transition kinetics equation, when the volume fraction of the transition reaches 98% of the theoretical value of the maximum transition amount calculation model, it is considered that the phase transition is completed, and the final grain size is calculated.

[0020] Further, based on the organization mechanical property relationship model, simulation calculation is carried out, and performance parameters are simulated, specifically including:

[0021] The chemical composition, average grain size of each phase and phase transformation volume fraction of the production bar wire are imported into the organization mechanical property relationship model, and the organization mechanical property relationship model predicts the yield strength, tensile strength, elongation and maximum force total elongation of the production bar wire based on the imported chemical composition, average grain size of each phase and phase transformation volume fraction.

[0022] Further, in step 3, if the simulated organization performance does not meet the national production standard and the expected value of the factory, the controlled cooling process is adjusted, and the simulation is continued until the simulated organization performance meets the national production standard and the expected value of the factory, wherein the adjustment of the controlled cooling process includes the rolling temperature, the water cooling parameter and the air cooling parameter.

[0023] The present application has the following beneficial effects:

[0024] 1. High precision and real-time prediction model: The present application establishes an online high-precision bar and wire organization performance prediction technology by integrating physical metallurgical models and data-driven methods. This technology can match the best simulation model parameters according to the real-time data received on the production line, and can predict the organization evolution and final performance of the bar and wire in real time. In the actual application process of the prediction model, the system can adjust and optimize itself according to the error between the actual measured value and the predicted value of the current production. Through continuous iterative calculation and parameter adjustment, the system keeps the prediction model in the best state at all times, ensuring the prediction accuracy and reliability. This real-time prediction and self-optimization mechanism greatly reduces the need for human intervention, making production adjustments more rapid and accurate to respond to various deviations in production.

[0025] 2. Improved production quality control: By monitoring and predicting the organization performance of the bar and wire in real time, the present application can foresee and identify potential quality problems at an early stage of production. By timely adjusting production parameters such as controlled cooling process, rolling temperature, etc., it can ensure that the final product meets or exceeds the quality standard, meets the factory expectations, significantly reduces the scrap rate, and improves the overall efficiency and economic benefits of the production line.

[0026] 3. Flexibility and adaptability: The technical solution of the present application considers different production conditions and needs, and can adapt to various production lines and material characteristics, whether it is a heating furnace or direct rolling, whether it is bar or wire rolling, the system can provide accurate temperature field and organization performance prediction, thus adapting to diversified production needs. This flexibility makes the present technology. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1A flow chart of an online high-precision rod wire structure performance prediction and production method is provided for the embodiment of the present application.

[0028] Figure 2 A flow chart of a whole-process temperature field model operation is provided for the embodiment of the present application.

[0029] Figure 3 A flow chart of a whole-process structure evolution model operation is provided for the embodiment of the present application.

[0030] Figure 4 A structure and mechanical property relationship model schematic diagram is provided for the embodiment of the present application. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of, but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the protection scope of the present application.

[0032] Embodiment 1

[0033] Referring to Figure 1 The online high-precision rod wire structure performance prediction and production method provided by the embodiment of the present application mainly includes two parts of model optimization and process adjustment.

[0034] The model optimization specifically implements the following steps:

[0035] The historical production data with higher matching degree are selected to perform simulation calculation with the prediction model, so as to simulate the temperature field, structure and performance parameters.

[0036] It is judged whether the errors of the simulated temperature field, structure and performance and the measured values of the current production are within the allowable range. If yes, the current production data and the prediction model are backed up into the database. If no, the model parameter optimization is performed in combination with the current production data until the errors are within the allowable range.

[0037] Module two: process adjustment. The specific implementation steps are as follows:

[0038] Based on the optimized high-precision prediction model, it is judged whether the simulated structure performance meets the national production standards and the expected values of the factory. If yes, rolling is performed. If no, the controlled cooling process (hereinafter referred to as the controlled cooling process) is adjusted to continue simulation until the structure performance meets the standards and the expected values.

[0039] In the embodiment of the application, the whole simulation is calculated by model formula. In order to ensure the reliability of the whole calculation accuracy, the initial model parameters are determined by fitting the experimental data and actual production conditions of typical low alloy steel. However, because of the differences in the environment and other production conditions of various steel plants, the model parameters need to be properly adjusted to maintain the accuracy of the simulation. Therefore, when calculating the temperature field, each heat exchange model, the grain size model when calculating the microstructure evolution, and the performance prediction model will have an adjustment factor. For a new or application case without historical data as a reference, the adjustment factor will be iteratively optimized according to the default value until the simulation accuracy meets the expectation, and then this set of optimized adjustment factors and production data (material chemical composition, rolling specification, rolling program table, rolling process temperature, cooling water tank parameters, air cooling line parameters, etc.) will be recorded as empirical knowledge. When there is a large amount of empirical data stored in the database, if the production data of the new simulation is similar to the empirical data (the system judgment rule is that the chemical composition error is within ±3%, the rolling specification is the same, the rolling program table hole type system error is within ±3%, the rolling process temperature error is within ±5℃, the cooling water tank parameter error is within ±3%, and the air cooling line parameter error is within ±5%), the adjustment factor is directly used, improving the calculation efficiency. The optimal prediction model is the adjustment factor matched with the current production data, which can simulate the results with little deviation from the actual production. The expected deviation value can be adjusted according to the demand. The system defaults ±10℃ for the temperature field, ±8% for the grain size, and 10MPa for the performance prediction. Among them, the historical production data is used to confirm the adjustment factor of the prediction model, that is, to confirm the final prediction model. Unless the production data is exactly the same, the results of the historical data can be directly used as the simulation results this time. However, the actual production data will always be different, so it needs to be simulated again. Although there are many historical data, they are classified and can be retrieved quickly. The memory of the stored data is very small and does not occupy space.

[0040] The prediction model is based on physical metallurgical model, including temperature field model, microstructure evolution model and microstructure-mechanical property relationship model, and the model parameter optimization specifically refers to the optimization of heat exchange model parameter, microstructure evolution model parameter and microstructure-mechanical property relationship model parameter.

[0041] The adjustment of the controlled cooling process is divided into two cases. For bar production line, the controlled cooling process includes the adjustment of the opening rolling temperature and the water cooling parameter. For wire production line, the controlled cooling process is Stelmor controlled cooling process, including the adjustment of the opening rolling temperature, the water cooling parameter and the air cooling parameter.

[0042] The full-process temperature field model of the prediction model is refined according to the actual physical process of the rolled piece, and is specifically divided into: a heat exchange model during conveying or between passes, a descaling water tank heat exchange model, a temperature rise model during rolling, a roll piece contact heat conduction model, a water cooling heat exchange model, a cooling bed heat exchange model, and a wire rod Stelmor heat exchange model. The model considers the following rolling process parameters: material quality and its thermal physical property parameters, rolling specifications, equipment process layout parameters, pre-rolling state (furnace temperature or continuous casting outlet temperature, whether to use a conveying roller cover), rolling program table, room temperature, descaling water tank state, cooling water temperature and pressure, water tank each nozzle state, Stelmor air cooling line fan on / off state and air volume, air cooling line each roller speed, whether to use a heat preservation cover in the air cooling line. The calculation of the temperature field adopts the finite difference method, in which the bar adopts a one-dimensional unsteady heat conduction equation, and the wire rod combines a one-dimensional unsteady heat conduction equation (before the drawing machine) and a two-dimensional unsteady heat conduction equation (after the drawing machine) to improve the calculation efficiency.

[0043] The full-process microstructure evolution model of the prediction model is subdivided into a pre-rolling grain growth model, a recrystallization model during rolling, and a phase transformation model in the collection area. The pre-rolling grain growth model is further divided into an austenite grain growth model in the heating process and the holding process. The recrystallization model during rolling is specifically divided into dynamic recrystallization, subdynamic recrystallization, static recrystallization, grain growth after recrystallization, incomplete recrystallization, and residual strain model. The phase transformation model in the collection area is refined into a grain growth before phase transformation, an austenite isothermal decomposition, a phase transformation start temperature / phase transformation incubation period, a phase transformation superposition, a maximum transformation amount of phase transformation, a calculation model of the volume fraction of each phase, and the grain size of each phase. The full-process microstructure evolution model is developed based on the calculated full-process temperature field data, and uses the same incremental step as the temperature field model.

[0044] The microstructure and mechanical property relationship model of the prediction model is specifically divided into yield strength, tensile strength, and elongation models in the present application according to the following indexes. The above models are calculated by importing the chemical composition of the material, rolling specifications, and other parameters, while combining the grain size of each phase and the volume fraction of each phase obtained by the previous simulation, to ensure that the mechanical properties of the simulated material meet the physical metallurgical law.

[0045] Example 2:

[0046] When performing temperature field parameter simulation calculation, the prediction model is a full-process temperature field model, as shown in Figure 2The whole-process temperature field model running flow chart provided by the embodiment of the present application, when simulating the whole-process temperature field, the prediction model firstly performs modeling of the blank and two-dimensional grid division, the grid number in the rolling direction (X axis) is 20 and the grid number in the radial direction (Y axis) is 4 by default, the grid division is adjusted according to the actual iteration condition before simulation to improve the simulation calculation efficiency. In the simulation, the system determines the appropriate time step through calculation to ensure the convergence of the calculation, and imports the corresponding material thermal physical property parameters from the database, selects the current heat exchange model, initial condition and boundary condition based on the process position condition of the incremental step, calculates the temperature field from the heating furnace or the continuous casting outlet to the finished product collection area, after completion, the system saves and outputs the temperature field simulation result. The prediction model then simulates the whole-process organization evolution based on the whole-process temperature field obtained by simulation.

[0047] Embodiment 3:

[0048] When performing the organization parameter simulation calculation, the prediction model is a whole-process organization evolution model, see Figure 3 The whole-process organization evolution model running flow chart provided by the embodiment of the present application is based on the temperature field simulation result, further simulates the whole-process organization evolution, this model takes the simulated temperature field as the initial condition, imports the rolling process parameters such as including the steel grade, rolling specification, blank size information, rolling program table, controlled cooling parameters and the like. The system calculates the strain ε, strain rate rolling temperature T, action time t of each stage, pre-rolling austenite initial size d0 of each stage according to these parameters. When calculating the pre-rolling austenite initial size d0, the model considers the heating and holding two cases, and the influence of the conveying roller way temperature drop process on the grain size is ignored. When rolling in a pass, the model firstly calculates the critical strain value ε c and the current cumulative strain value ε * When ε * ≥ ε c , dynamic recrystallization occurs, the dynamic recrystallization grain size d d and the recrystallization percentage X d are calculated. When ε * ≤ ε c , static recrystallization occurs, the static recrystallization grain size d s and the recrystallization percentage X s are calculated. The model assumes that once a certain grain has dynamic recrystallization, most of the deformation amount is released, and the remaining deformation energy is insufficient to drive the grain to occur static recrystallization again, that is, the grain which has dynamic recrystallization calculates its grain growth process as sub-dynamic recrystallization. According to this idea, during the interpass interval, the model firstly judges the recrystallization fraction X d or X swhether ≥ 0.95, if yes, calculate the static recrystallization grain growth d after the pass interval recrystallization is completed SG and sub-dynamic recrystallization grain growth d MG , if no, calculate the static recrystallization grain growth d of incomplete static recrystallization that does not occur pass interval recrystallization SG and sub-dynamic recrystallization grain growth d MG , both cases are different in the time of grain growth, the former is equal to the interval time of the current pass minus the time required for recrystallization to be completed, and the latter is equal to the interval time of the current pass minus the time required for the fraction of recrystallization to have occurred. After the grain growth calculation is completed, the model will calculate the average grain size and output and save the occurrence of various recrystallizations. The model will loop from the first pass to the last pass of the mill, and after the loop is completed, the average grain size of the austenite before cooling, the temperature field and the residual strain will be output and saved as the initial conditions of the phase transition.

[0049] In the cooling stage after rolling is completed, if the average temperature of the intermediate blank is higher than the phase equilibrium temperature Ae3, the austenite grains will continue to grow. At this time, the model first uses the superposition method to calculate the growth of the grain size based on the whole process temperature field. After reaching the Ae3 temperature, the isothermal decomposition of austenite begins, first the phase transition incubation period model is used to determine the starting temperature of the occurrence of each phase transition, and then the continuous isothermal superposition method is used to calculate the volume fraction of each phase transition based on the phase transition kinetics equation. When the volume fraction of the transition reaches 98% of the theoretical value of the maximum transition amount of the model, it is considered that the phase transition is completed, and the final grain size is calculated, including the average grain size of ferrite and the lamellar distance of pearlite.

[0050] Example 4:

[0051] When performing performance parameter simulation calculation, the prediction model is a microstructure and mechanical property relationship model, see Figure 3 , the microstructure and mechanical property relationship model running flowchart provided by the embodiment of the application, the model comprehensively considers the factors such as chemical composition, average grain size of each phase and volume fraction of each phase transition, and predicts the yield strength, tensile strength, elongation and maximum force total elongation of the product, wherein the influence of chemical composition considers the influence of multiple elements such as C, Mn, Si, V and N, and the average grain size and volume fraction of each phase consider the influence of crystal structures such as austenite, ferrite, pearlite and bainite.

[0052] As Figure 1As shown, after the simulation is completed, the system will verify whether the error of the simulated temperature field-tissue-performance data and the actual measured value is within the allowable range. If not, the relevant model parameters are adjusted until the error of the simulated temperature field-tissue-performance data and the actual measured value is within the allowable range; if so, after data cleaning, it is archived together with the prediction model, enriching the historical data. Then, the system judges whether the simulated tissue performance meets the production standards and the factory expected value, if so, rolling, if not, adjusting the controlled cooling process and continuing to simulate until the requirements are met. The adjustment of the controlled cooling process includes the rolling temperature, water cooling parameters and air cooling parameters.

[0053] Since the implementation of the present scheme, the prediction accuracy is high, see Tables 1-3 below for the comparison between the actual measured value and the predicted value of the present scheme in different production lines:

[0054]

[0055] Table 1 Table 2

[0056]

[0057]

[0058] Table 3

[0059] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An online high-precision rod wire organizational performance prediction and production method, characterized in that, The method comprises: Step 1, selecting historical production data matched with the production of bar and wire rod production process, determining an optimal prediction model based on the matched historical production data, simulating the production of bar and wire rod based on the prediction model, and simulating a temperature field, a structure and a performance parameter; Step 2, judging whether the error of the simulated temperature field, structure and performance parameter and the measured value of the produced bar and wire rod is within an allowable range, if yes, backing up the corresponding production data and the prediction model into a database, and if no, optimizing the model parameters in combination with the current production data until the error is within the allowable range; Step 3, simulating the production data of the current bar and wire rod based on the optimized prediction model, simulating a temperature field, a structure and a performance parameter, and judging whether the simulated structure and performance meet the national production standard and the expected value of the factory, and if yes, rolling. The optimal prediction model is determined based on the matched historical production data, specifically, an adjustment factor of the prediction model is obtained, and when the adjustment factor of the prediction model and the historical production data reach a preset matching degree, the corresponding prediction model is considered as the optimal prediction model.

2. The online high-precision rod wire microstructure and property prediction and production method according to claim 1, characterized in that, The prediction model is developed based on a physical metallurgical model.

3. The online high-precision rod wire microstructure and property prediction and production method according to claim 2, characterized in that, The prediction model comprises a temperature field model, a structure evolution model and a structure and mechanical property relationship model.

4. The online high-precision rod wire organizational performance prediction and production method according to claim 3, characterized in that, The temperature field model is used for simulating a temperature field parameter, specifically including: In the simulation of the whole temperature field, the temperature field model firstly performs modeling of a blank and two-dimensional grid division, with a default rolling direction, grid number, radial direction and grid number, and the grid division is adjusted according to the actual iteration before simulation to improve the simulation calculation efficiency; in the simulation, a suitable time step is determined to ensure the convergence of the calculation, and corresponding material thermal physical property parameters are imported from a database based on the process position of the incremental step, a current heat exchange model, initial conditions and boundary conditions are selected, the temperature field from a heating furnace or a continuous casting outlet to a finished product collection area is calculated, and after completion, the temperature field simulation result is saved and output.

5. The online high-precision rod wire microstructure and property prediction and production method according to claim 3, characterized in that, The structure evolution model is used for simulating a structure parameter, specifically including: The simulated temperature field parameters are used as initial conditions and imported into the rolling process parameters. Based on the rolling process parameters, the strain for each pass is calculated. strain rate Rolling temperature Duration of each stage Initial dimensions of austenite before rolling During a certain rolling pass, the microstructure evolution model first calculates the critical strain value for dynamic recrystallization. and the current cumulative strain value ,when When dynamic recrystallization occurs, calculate the dynamic recrystallized grain size. and recrystallization percentage ,when When static recrystallization occurs, calculate the static crystal grain size. and recrystallization percentage During the inter-pass interval, the microstructure evolution model first determines the percentage of recrystallization that occurs. or If the value is ≥0.95, then calculate the static recrystallized grain growth after all recrystallization passes are completed. and subdynamic recrystallization grain growth If not, then calculate the static recrystallization grain growth that did not occur during the pass interval recrystallization. and subdynamic recrystallization grain growth After the grain growth calculation is completed, the microstructure evolution model calculates the average grain size and outputs and saves various recrystallization occurrences. The structure evolution model performs cyclic calculation from the first pass to the last stand rolling mill pass, and after the cycle is completed, the average grain size, temperature field and residual strain of the austenite before cooling and phase change are output and saved as initial conditions of the phase change.

6. The online high-precision rod wire organizational performance prediction and production method according to claim 5, characterized in that, The method further comprises: in the cooling stage after rolling, if the average temperature of the intermediate blank is higher than the phase equilibrium temperature Ae3, the austenite grain will continue to grow, at this time, the structure evolution model firstly calculates the growth of the grain size based on the whole process temperature field by using the superposition method, after the average temperature of the intermediate blank reaches the Ae3 temperature, the isothermal decomposition of the austenite starts, the start temperature of the transformation of each phase is determined by using a phase transformation incubation period model, the volume fraction of each phase transformation is calculated by using a continuous isothermal superposition method based on a phase transformation kinetics equation, and when the volume fraction of the transformation reaches 98% of the theoretical value of the maximum transformation amount calculation model of each phase, it is considered that the phase transformation is completed, and the final grain size is calculated.

7. The online high-precision rod wire microstructure and property prediction and production method according to claim 3, characterized in that, The structure and mechanical property relationship model is used for simulating a performance parameter, specifically including: The chemical composition, the average grain size of each phase, and the volume fraction of each phase transformation of the produced bar wire are imported into a microstructure-mechanical property relationship model, and the microstructure-mechanical property relationship model predicts the yield strength, the tensile strength, the elongation, and the maximum force total elongation of the produced bar wire based on the imported chemical composition, the average grain size of each phase, and the volume fraction of each phase transformation.

8. The online high-precision rod wire microstructure and property prediction and production method according to claim 1, characterized in that, In Step 3, if the simulated microstructure properties do not meet the national production standards and the expected values of the factory, the controlled cooling process is adjusted and the simulation is continued until the simulated microstructure properties meet the national production standards and the expected values of the factory, wherein the adjustment of the controlled cooling process includes the opening rolling temperature, the water cooling parameters, and the air cooling parameters.

Images