Method for the controlled annealing of bulletproof steel strips for the processing of ultra-high strength steels

Abstract

The application discloses a bulletproof steel strip cover annealing control method for ultra-high strength steel processing and relates to the technical field of metal material heat treatment. The method comprises the following steps: constructing a multi-physical field coupling model describing the state evolution of the bulletproof steel strip in the cover annealing process; setting a multi-objective optimization function oriented to the predicted final service performance index of the bulletproof steel strip; based on the model and the optimization function, performing adaptive control on the annealing process including the heating, soaking and cooling stages; and in the annealing process, on-line monitoring the measured values of the process parameters, comparing the measured values with the predicted values obtained based on the model, and when the deviation between the measured values and the predicted values exceeds a preset threshold, on-line correcting the model parameters and dynamically adjusting the subsequent control strategy. The application directly links the process control and the final service performance through the establishment of a closed-loop feedback mechanism, thereby improving the annealing uniformity, product performance consistency and batch production stability.

Description

Technical Field

[0001] This invention relates to the field of heat treatment technology for metallic materials, and in particular to a method for controlling the annealing of bulletproof steel with a shield for processing ultra-high strength steel. Background Technology

[0002] As a critical protective material, bulletproof steel strips require extremely high precision and stability in heat treatment processes. Bell-type annealing is a crucial step in the production process, used to eliminate cold-rolling stress and control microstructure. Currently, bell-type annealing of cold-rolled strip steel commonly employs an all-hydrogen bell-type annealing furnace, using strongly convective circulating hydrogen for heating and cooling.

[0003] However, existing bell-type annealing processes rely heavily on fixed process curves and manual experience for control. Firstly, the steel coil is tightly wound in multiple layers within the furnace, resulting in extremely poor radial thermal conductivity. This leads to significant temperature differences between the inner and outer layers during heating and cooling, causing severe inhomogeneity in the microstructure and properties of different locations within the same coil after annealing, directly impacting the consistency of the final product's impact resistance. Secondly, current annealing control primarily aims to ensure sufficient recrystallization or achieve a certain hardness range, without directly optimizing key service indicators such as final impact resistance, leading to a disconnect between process and performance. Finally, fixed process curves cannot cope with variations in actual operating conditions such as fuel calorific value fluctuations, equipment aging, and differences in furnace loading methods, making it difficult to guarantee the stability of mass production. Therefore, existing technologies suffer from inefficient control methods, inappropriate optimization objectives, and poor adaptability, resulting in uneven internal temperature and microstructure of the bulletproof steel strip after bell-type annealing, presenting a pressing technical problem that needs to be addressed. Summary of the Invention

[0004] The present invention aims to solve the technical problems in the prior art, such as uneven internal temperature and microstructure of steel coils caused by crude control methods, inappropriate optimization objectives, and poor adaptability, which in turn affect the consistency of the impact resistance performance of the final product and the stability of mass production.

[0005] To achieve the above objectives, this invention provides a method for controlling the annealing of bulletproof steel strips in ultra-high strength steel processing, comprising the following steps: constructing a multiphysics coupling model describing the state evolution of the bulletproof steel strip during annealing; setting a multi-objective optimization function guided by the predicted final service performance index of the bulletproof steel strip; adaptively controlling the annealing process, including heating, homogenization, and cooling stages, based on the multiphysics coupling model and the multi-objective optimization function; during the annealing process, monitoring the measured value of at least one process parameter online, comparing the measured value of the process parameter with the predicted value of the process parameter at the same measurement location obtained based on the multiphysics coupling model, and when the deviation between the two exceeds a preset threshold, correcting the parameters of the multiphysics coupling model online, and dynamically adjusting the subsequent control strategy based on the corrected model.

[0006] Optionally, the multiphysics coupling model includes at least a temperature field model, a tissue evolution model, and a stress field model.

[0007] Optionally, the multi-objective optimization function includes ballistic impact resistance index, microstructure uniformity index, and annealing process energy consumption index based on the predicted microstructure after annealing.

[0008] Optionally, the step of online monitoring of the measured value of at least one process parameter includes: arranging a temperature sensor in the bell-type annealing furnace to collect the measured temperature value as the measured value of the process parameter in real time; and the comparison step specifically includes: comparing the measured temperature value with the temperature prediction value at the corresponding sensor location, which is predicted in real time by the multiphysics coupling model based on the current process state.

[0009] Optionally, the temperature sensor is a wireless temperature measuring device that is installed in the furnace along with the steel coil.

[0010] Optionally, the step of adaptively controlling the annealing process includes employing a variable heating rate strategy during the heating phase.

[0011] Optionally, the variable heating rate strategy includes: using a heating rate of 60-80℃ / h in the low-temperature range below a preset temperature threshold, and using a heating rate of 30-50℃ / h in the high-temperature range above or equal to the preset temperature threshold.

[0012] Optionally, the step of adaptively controlling the annealing process includes, during the homogenization stage, determining the start and end times of homogenization based on the predictions of the coldest point temperature and recrystallization fraction of the steel coil by the multiphysics coupling model.

[0013] Optionally, homogenization may be terminated when the recrystallization fraction predicted by the model reaches ≥95%.

[0014] Optionally, the homogenization stage is carried out in a full hydrogen protective atmosphere with hydrogen purity ≥ 99.999%, and the furnace pressure is maintained at 8-12 kPa.

[0015] Optionally, the step of adaptively controlling the annealing process includes employing a multi-stage cooling strategy during the cooling phase.

[0016] Optionally, the multi-stage cooling strategy includes three stages: shrouded slow cooling, air cooling, and water cooling.

[0017] Optionally, the multi-objective optimization function includes ballistic impact resistance index, microstructure uniformity index, and annealing process energy consumption index based on the predicted microstructure after annealing; the method further includes: performing a quality evaluation on the ballistic steel strip after annealing, and feeding back the evaluation results to iteratively optimize the weights in the multi-objective optimization function.

[0018] The present invention has the following beneficial effects:

[0019] 1. By constructing a multiphysics coupling model and setting a multi-objective optimization function oriented towards final service performance, the annealing process control is no longer solely aimed at traditional recrystallization or hardness, but is directly linked to key indicators such as final ballistic impact resistance. This solves the problem of disconnect between process and performance, making process control more targeted and purposeful.

[0020] 2. By monitoring key process parameters online and using a multi-physics coupling model to calculate the temperature field and microstructure inside the steel coil in real time, this invention can use the actual state inside the steel coil (such as the coldest point) as the criterion for switching process stages, rather than relying on the lagging furnace temperature. This effectively improves the problem of inaccurate control by "using furnace temperature instead of steel temperature" in traditional processes, significantly improves control accuracy, and thus improves the annealing uniformity of the steel coil.

[0021] 3. The closed-loop feedback mechanism of "online monitoring-model comparison-parameter correction-dynamic control" established by this invention enables the control system to respond in real time and compensate for the impact of actual working conditions (such as fluctuations in fuel calorific value, changes in equipment status, etc.), giving the annealing process good self-adaptability and effectively enhancing the robustness of the process and the stability of mass production.

[0022] 4. Through model-based prediction and optimization, this invention can generate better process curves. For example, by using variable-rate heating and accurately determining the end time of heat soaking, unnecessary overheating and ineffective heat preservation are effectively avoided. Compared with the traditional fixed-curve process, it has significant energy-saving and consumption-reducing effects. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the system structure of an adaptive control device according to an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the overall process of the adaptive control method according to an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram illustrating the temperature changes at each stage of the annealing process in an embodiment of the present invention.

[0027] Figure 4 This is a detailed flowchart of the online correction and closed-loop control section according to an embodiment of the present invention.

[0028] In the diagram: 10. Bell-type annealing furnace; 11. Steel coil; 20. Temperature sensor; 30. Control system; 31. Model module; 32. Target setting module; 33. Adaptive control module; 34. Online correction module. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be considered as limitations on the invention. All other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention. Unless otherwise defined, all technical and scientific terms used in the embodiments of this invention have the same meaning as commonly understood by those skilled in the art. The terminology used in the embodiments of this invention is for the purpose of describing the embodiments of this invention only and is not intended to limit the invention.

[0030] Before providing a further detailed description of the embodiments of the present invention, some of the nouns and terms involved in the embodiments of the present invention will be explained, and the nouns and terms involved in the embodiments of the present invention shall be interpreted as follows.

[0031] (1) Bell-type annealing: refers to a heat treatment process in which coiled strip steel is stacked and placed into an annealing furnace with a sealed furnace lid, and heated, held and cooled under a protective atmosphere. In the embodiments of the present invention, it specifically refers to all-hydrogen bell-type annealing using hydrogen as a protective and heat transfer medium, which achieves efficient heat exchange through strong convection circulation of hydrogen.

[0032] (2) Multiphysics coupling model: refers to a set of mathematical models that can comprehensively describe the internal state evolution of bulletproof steel strip during the annealing process. In the embodiments of the present invention, it couples at least the temperature field model (used to describe the transfer and distribution of heat inside the steel coil), the microstructure evolution model (used to describe the changes in the microstructure of the steel strip, such as recovery, recrystallization and grain growth), and the stress field model (used to describe the distribution of internal stress caused by temperature inhomogeneity and phase transformation). Through the synergistic calculation of these models, accurate prediction of the state parameters inside the steel coil that cannot be directly measured can be achieved.

[0033] (3) Multi-objective optimization function: This refers to a mathematical expression used to guide the optimization of annealing process parameters. It unifies multiple, and even potentially conflicting, performance objectives, such as final product performance, microstructure uniformity, and production costs (e.g., energy consumption), into a comprehensive evaluation system through weighted summation and other mathematical methods. In the embodiments of this invention, a core feature of this function is that it takes the predicted final service performance of the bulletproof steel strip (e.g., ballistic impact resistance) as the key guide, thereby enabling process control to directly serve the product application objectives.

[0034] (4) Adaptive control: refers to an advanced control method that can automatically adjust its control strategy to maintain optimal performance based on the real-time state of the system and changes in the external environment. In the embodiments of the present invention, it is specifically embodied as a closed-loop control system of "online monitoring-model prediction-deviation comparison-parameter correction-dynamic adjustment", which can cope with various disturbances in actual production and ensure the stability of the process and the uniformity of the product.

[0035] Please see Figure 2 This invention provides a method for controlling the annealing of bulletproof steel with a shield for processing ultra-high strength steel. It aims to solve the technical problems in existing technologies, such as uneven microstructure and poor consistency in the bullet impact resistance of the final product due to the crude control methods and lack of precise control over the internal state of the steel coil. This method establishes a model-based closed-loop control system to achieve refined and intelligent control of the entire annealing process.

[0036] In a basic implementation, the method first requires constructing a multiphysics coupled model for the bulletproof steel strip to be processed, capable of describing its state evolution during bell-type annealing. This step is fundamental to the entire adaptive control method, aiming to create a digital twin to simulate and predict complex changes within the steel coil that are difficult to observe directly in the physical world. Traditional processes can only control the furnace temperature, but cannot obtain the true temperature and microstructure of the steel coil's interior, especially the coldest and hottest points. This step, by establishing a precise mathematical model, makes subsequent precise control possible.

[0037] Next, the method sets up a multi-objective optimization function guided by the predicted final service performance indicators of the bulletproof steel strip. This step embodies the core idea of ​​the invention: the goal of process control is no longer the traditional task of ensuring recrystallization or reaching a certain hardness range, but is directly linked to the final performance of the product. By unifying multiple objectives such as predicted ballistic resistance, microstructure uniformity, and process energy consumption into a single function, the control system can find the optimal balance point among multiple dimensions, thereby solving the problem of traditional process objectives being singular and disconnected from the final performance.

[0038] After the model and objective function are established, the method adaptively controls the entire annealing process, including the heating, soaking, and cooling stages, based on the multiphysics coupling model and the multi-objective optimization function. This means that the control system no longer executes a fixed process curve, but instead calculates the optimal control commands (such as heating power and holding time) in real time based on the model's prediction of future states, driving the entire annealing process towards the optimal direction of the multi-objective optimization function. This model-predictive control method makes the entire process proactive, able to actively respond to changes rather than passively reacting.

[0039] To ensure the accuracy of model predictions and cope with fluctuations in real-world operating conditions, this method also includes a crucial closed-loop correction step. During the annealing process, the system monitors the measured value of at least one process parameter online and compares it with the predicted value of the same process parameter at the same measurement location, obtained based on a multiphysics coupling model. When the deviation between the two exceeds a preset threshold, the system corrects the parameters of the multiphysics coupling model online and dynamically adjusts subsequent control strategies based on the corrected model. This step solves the problem that traditional fixed process curves cannot adapt to actual disturbances such as fuel calorific value fluctuations, equipment aging, and differences in furnace loading methods. Through the closed loop of "measurement-prediction-comparison-correction," the entire control system is endowed with strong adaptive capabilities and robustness.

[0040] Furthermore, in a preferred embodiment, the multiphysics coupling model includes at least a temperature field model, a microstructure evolution model, and a stress field model. The temperature field model is used to accurately calculate the temperature distribution and changes at any location within the steel coil and is the foundation of all models. The microstructure evolution model predicts processes such as recrystallization and grain growth of the microstructure based on temperature history. The stress field model is used to assess thermal stress and phase transformation stress to prevent deformation and cracking. By coupling these three key physical fields, the true state evolution of the steel coil during annealing can be reflected more comprehensively and accurately, thereby significantly improving the accuracy and reliability of the model predictions.

[0041] In another preferred embodiment, the multi-objective optimization function specifically includes a ballistic impact resistance index, a microstructure uniformity index, and an annealing process energy consumption index based on the predicted microstructure after annealing. The introduction of the ballistic impact resistance index ensures that the optimization of the annealing process directly serves the core value of the product; the microstructure uniformity index aims to address the industry pain point of large differences in internal and external properties of the steel coil; and the energy consumption index takes into account the economic efficiency of production. Through the combination of these three specific indicators, the optimization objectives become more explicit and practical, achieving comprehensive optimization of product performance, quality, and cost.

[0042] To more clearly illustrate the above ballistic performance indicators ( ) and tissue homogeneity index ( The specific quantification process enables those skilled in the art to construct the multi-objective optimization function without creative effort. The present invention provides the following preferred computational implementation methods:

[0043] I. Ballistic Impact Resistance Indicators Quantitative calculation method

[0044] Ballistic performance indicators This invention utilizes quantitative indicators calculated from annealed microstructure parameters (including average grain size, volume fraction of each phase, grain size distribution, etc.) predicted by a multiphysics coupling model. These indicators characterize the ability of bulletproof steel strips to resist projectile impact. The invention employs the following method to achieve this. Quantitative calculation:

[0045] (1) Establish a correlation model between microstructure parameters and mechanical properties

[0046] The impact resistance of bulletproof steel strips is closely related to their mechanical properties such as hardness, strength, and toughness. Studies have shown that the morphology (lath or plate) and size of martensite are the effective grain sizes that determine its resistance to impact fracture or perforation. Therefore, the following key microstructure parameters are first extracted from the annealed microstructure predicted by a multiphysics coupling model:

[0047] Average grain size (μm);

[0048] Average width of martensitic laths / blocks (μm);

[0049] Residual austenite volume fraction (%)

[0050] Volume fraction of carbide precipitates (%) and average size (μm).

[0051] Based on the above parameters, the hardness and tensile strength of the annealed bulletproof steel strip are calculated using the following empirical formula:

[0052] hardness The calculation uses a Hall-Petch type relation:

[0053]

[0054] in, It is the lattice friction hardness (matrix hardness). The Hall-Petch slope coefficient, and These are the contribution coefficients of retained austenite and carbides to hardness, respectively. The specific values ​​of these coefficients were obtained by conducting experiments on standard samples of the same grade of bulletproof steel strip using different annealing processes, and calibrating them using linear regression analysis.

[0055] tensile strength (MPa) is calculated in the following form:

[0056]

[0057] in, For matrix strength, The intensity-Hall-Petch slope coefficient, The carbide precipitation strengthening coefficient is given.

[0058] (2) Establish ballistic performance indicators computational model

[0059] Based on the correlation between the ballistic impact resistance and mechanical properties of armor steel, this invention uses the following mathematical model to calculate the ballistic impact resistance index:

[0060]

[0061] in:

[0062] , These are the upper and lower limit reference values ​​for the hardness of the bulletproof steel belt of this grade, respectively;

[0063] , These are the upper and lower reference values ​​for tensile strength, respectively.

[0064] The contribution of tissue homogeneity to toughness (see below for specific calculation method). (calculation)

[0065] , , For the weighting coefficients, satisfying The optimal value is determined through iterative optimization using a multi-objective optimization algorithm.

[0066] Alternatively, in another implementation, Alternatively, the ballistic limit velocity can be used directly. The predicted values ​​are used to characterize this. Based on finite element simulation and regression analysis, the following prediction model is established:

[0067]

[0068] in, The regression coefficients were obtained through least-squares regression analysis using standard ballistic impact test data for this grade of bulletproof steel belt. At this point, = ,in This is the baseline ballistic limit velocity for this grade of bulletproof steel belt.

[0069] (3) Calibration of model parameters

[0070] All empirical coefficients in the above model ( All of these are pre-calibrated through the following steps:

[0071] Step 1: For bulletproof steel strips of the same grade, select no less than 5 different annealing process conditions (covering different heating rates, soaking temperatures and soaking times) to prepare standard samples;

[0072] Step 2: Perform metallographic analysis (obtain microstructure parameters such as grain size and phase composition) and mechanical property testing (hardness, tensile strength, and impact toughness) on each sample.

[0073] Step 3: Conduct standard ballistic impact tests (such as V50 ballistic limit test) on some samples to obtain experimental data on ballistic resistance performance;

[0074] Step 4: Use the least squares method or neural network regression method to fit the above experimental data and calibrate the values ​​of each coefficient in the model;

[0075] Step 5: Store the calibrated coefficients in the model module 31 of the control system 30 for online calculation.

[0076] II. Tissue homogeneity index ( Quantitative calculation method

[0077] tissue homogeneity index This invention is used to quantitatively evaluate the consistency of the microstructure at different locations (e.g., radially from the outer to the inner ring, axially from the top to the bottom) of the same steel coil after annealing. The invention employs the following method to achieve this. Quantitative calculation:

[0078] (1) Determine the microstructure parameters of the sampling points inside the steel coil

[0079] The multiphysics coupling model spatially discretizes the interior of the steel coil (dividing it radially into...) Layers, divided along the axial direction into (layer), at each discrete node Predicted microstructure parameters after annealing ,include:

[0080] Average grain size ;

[0081] Volume fraction of each phase (martensite, retained austenite, carbides) ( (representing different phases);

[0082] Standard deviation of grain size distribution .

[0083] (2) Calculate the uniformity sub-index of a single tissue parameter

[0084] For a specific microstructure parameter (e.g., average grain size) Or the volume fraction of a certain phase Its uniformity sub-index Calculated using the reciprocal of the coefficient of variation:

[0085]

[0086] in, Organize parameters on all discrete nodes The average value, Its standard deviation:

[0087]

[0088]

[0089] The range of values ​​is The closer the value is to 1, the more uniform the distribution of this microstructure parameter within the steel coil. hour, , indicating complete uniformity.

[0090] In another implementation, the standard deviation of the grain area can also be used. As a parameter for evaluating uniformity, the uniformity sub-index is defined as follows:

[0091]

[0092] in, For organizational parameters The standard deviation of grain area, This is the reference standard deviation (obtained through calibration using standard samples).

[0093] (3) Calculate the comprehensive tissue uniformity index

[0094] Taking into account the uniformity of multiple key tissue parameters, Defined as a weighted combination of all uniformity sub-indices:

[0095]

[0096] in:

[0097] This is a sub-indicator for the uniformity of average grain size;

[0098] The uniformity index of the volume fraction of retained austenite;

[0099] This is a sub-indicator of the uniformity of carbide volume fraction;

[0100] This is a sub-index of uniformity of the standard deviation of grain size distribution;

[0101] For the weighting coefficients, satisfying .

[0102] The specific values ​​of each weight coefficient are determined by the analytic hierarchy process (AHP) or design of experiments (DOE) based on the specific grade and usage requirements of the bulletproof steel belt.

[0103] In another implementation, tissue homogeneity index The following simplified form can also be used for calculation:

[0104]

[0105] in, The total number of organizational parameters considered ( ), For the first Each microstructure parameter (such as grain size, volume fraction of each phase, etc.) has a physical meaning: the smaller the average value of the coefficient of variation of each microstructure parameter, the better the microstructure uniformity.

[0106] (4) The physical meaning and threshold setting

[0107] The range of values ​​is In practical applications:

[0108] when At that time, the uniformity of the annealed microstructure of the steel coil was judged to be "excellent";

[0109] when At that time, it was judged as "good";

[0110] when At that time, it was judged as "qualified";

[0111] when At that time, it was judged as "unqualified".

[0112] The above thresholds are determined by statistical analysis of the uniformity data of a large number of production batches and the performance of the final product (such as the consistency of ballistic resistance), and can be dynamically adjusted according to the actual production situation.

[0113] In another implementation, an image processing-based tissue uniformity evaluation method can be used: the tissue parameters of each discrete node predicted by the multi-physics coupling model are mapped into a pseudo-color image, and then the image is processed by grayscale, binarized, contour extracted and ellipse fitted to calculate the size uniformity and distribution uniformity of the dispersed phase, and finally the tissue uniformity evaluation result is obtained by combining the results.

[0114] III. In multi-objective optimization functions and synergy

[0115] In a complete multi-objective optimization function and Synergistic effect:

[0116]

[0117] in, These are the weighting coefficients. This ensures the product's core bulletproof performance. This ensures the consistency of performance across different locations within the same steel coil. The combination of these two methods addresses both the issue of "whether performance meets standards" and the issue of "whether performance is uniform," thereby fully realizing the adaptive control objective oriented towards final service performance.

[0118] All empirical and weighting coefficients involved in the above calculation methods were calibrated through offline experiments on standard samples before the system was put into actual production and stored in the model module 31 of the control system 30. During the production process, these coefficients can be iteratively optimized using the quality evaluation results after annealing, so that the model prediction accuracy can be continuously improved with the accumulation of production data.

[0119] In one specific implementation, the online monitoring and comparison step is achieved by arranging temperature sensors inside the bell-type annealing furnace to collect real-time temperature measurements as process parameters. The comparison step specifically involves comparing the measured temperature values ​​with the predicted temperature values ​​at the corresponding sensor locations, predicted in real-time by a multiphysics coupling model based on the current process state. This method directly links the actual measurements in the physical world with the model predictions in the digital world, providing the most direct and reliable basis for online model correction. This approach solves the potential discrepancy between the model and reality, helping to ensure control accuracy.

[0120] In a more preferred embodiment, the temperature sensor is a wireless temperature measuring device installed in the furnace along with the steel coil. Traditional temperature measuring points are usually arranged in the furnace or on the surface of the steel coil, which cannot accurately reflect the temperature inside the steel coil, especially at the radial center. By using a wireless temperature measuring device, the real temperature data of key locations inside the steel coil can be directly obtained, significantly improving the accuracy of model inversion calculations and online corrections, thereby achieving more precise control over the temperatures of the coldest and hottest points of the steel coil.

[0121] In one optional implementation, the step of adaptively controlling the annealing process includes employing a variable heating rate strategy during the heating phase. This strategy aims to balance heating efficiency and temperature uniformity. In the initial heating stage, the overall temperature of the steel coil is low, and the temperature difference between the inside and outside is small, allowing for a higher heating rate to shorten the production cycle. However, in the high-temperature range near the annealing temperature, the heating rate needs to be reduced to minimize the temperature difference caused by uneven heat conduction. By implementing the variable heating rate strategy, production efficiency is maximized while ensuring final temperature uniformity.

[0122] Furthermore, the variable heating rate strategy can specifically include: using a higher heating rate of 60-80℃ / h in the low-temperature range below a preset temperature threshold (e.g., 400℃), and using a lower heating rate of 30-50℃ / h in the high-temperature range above or equal to the preset temperature threshold. This segmented parameter setting is based on optimization results derived from extensive experiments and model simulations, providing a clear and effective implementation plan for the variable heating rate strategy, making it more operable and reliable.

[0123] In another preferred embodiment, the heating stage employs a pulse combustion control method. Compared to the traditional method of continuously adjusting the combustion valve opening, pulse combustion controls the total heat output by adjusting the duty cycle or frequency of the combustion pulse through a high-frequency switching burner. This method can achieve precise and linear control of heating power over a wide adjustment range, maintaining a stable combustion state, especially when low power output is required. This provides a powerful means of achieving variable heating rates and precise temperature control.

[0124] In one optional implementation, the step of adaptively controlling the annealing process includes, during the homogenization stage, determining the start and end times of homogenization based on the predictions of the coldest point temperature and recrystallization fraction of the steel coil by the multiphysics coupling model. This is an effective improvement over the traditional coarse-grained control method that substitutes furnace temperature for steel temperature. Traditional processes start timing when the furnace temperature reaches a set value, but at this point, the inside of the steel coil is far from reaching the target temperature. This invention, through model prediction, ensures that homogenization only begins after the coldest point of the steel coil has truly reached the annealing temperature, and ends promptly after the microstructure transformation (such as recrystallization) is complete, thereby achieving precise control of the homogenization process and avoiding under- or over-hoisting.

[0125] Furthermore, the homogenization process ends when the recrystallization fraction predicted by the model reaches ≥95%. This provides a clear, quantitative standard based on physical metallurgical principles for determining the end of homogenization. Compared to traditional methods that rely on experience or fixed times, this approach can dynamically determine the shortest effective homogenization time based on actual conditions such as different steel coil specifications and furnace loading methods. This helps ensure annealing quality while effectively improving production efficiency and energy utilization.

[0126] In a preferred embodiment, the homogenization stage is conducted in a protective atmosphere of pure hydrogen with a purity ≥99.999%, with the furnace pressure maintained at 8-12 kPa. The high-purity pure hydrogen atmosphere has extremely high thermal conductivity, significantly enhancing convective heat transfer efficiency and helping to reduce the temperature difference between the inside and outside of the steel coil; simultaneously, its strong reducing properties ensure a bright, oxidation-free surface on the steel strip. The specific positive pressure inside the furnace effectively prevents outside air from leaking in. These optimized process environment parameters provide favorable conditions for achieving high-quality, highly uniform annealing results.

[0127] In one alternative implementation, the step of adaptively controlling the annealing process includes employing a multi-stage cooling strategy during the cooling phase. The principle is that different microstructure transformations may occur in the steel strip as it cools at different temperature ranges, requiring different cooling rates. For example, some temperature zones require slow cooling to promote the precipitation of beneficial phases, while others require rapid cooling to suppress the formation of brittle phases. By employing multi-stage cooling, the cooling curve can be finely "tailored" according to the material properties, thereby achieving optimized control over the final microstructure and properties.

[0128] Furthermore, the multi-stage cooling strategy can specifically include three stages: slow cooling with a shroud, air cooling, and water cooling. The slow cooling stage with a shroud utilizes residual heat to achieve gradual cooling, the air cooling stage achieves moderate-rate cooling through forced convection, and the water cooling stage achieves rapid cooling to the furnace exit temperature. The combination of these three stages provides a wide-range, controllable cooling path, capable of meeting the complex process requirements of most steel grades during the cooling process.

[0129] In a preferred embodiment, the bulletproof steel strip is an ultra-high-strength steel with a tensile strength ≥1500MPa. This type of steel typically contains multiple alloying elements, exhibits complex phase transformation behavior, and places extremely stringent requirements on the precision and uniformity of the heat treatment process. The adaptive control method provided by this invention aims to solve the quality control challenges faced by such high-end materials under traditional processes, thus demonstrating its technological advantages and value in this application.

[0130] Furthermore, the chemical composition of the bulletproof steel strip, by mass percentage, may include: C 0.25-0.45%, Si 0.15-0.50%, Mn 0.50-1.50%, Cr 0.80-2.00%, Mo 0.20-0.80%, and Ni 0.50-3.00%. The given specific composition range defines a preferred application of the method of this invention. This composition system is a typical design for achieving a combination of ultra-high strength and good toughness, and its complex phase transformation characteristics also make the modeling and adaptive control method of this invention even more necessary.

[0131] In a preferred embodiment, the multi-objective optimization function includes ballistic impact resistance indicators, microstructure uniformity indicators, and annealing process energy consumption indicators based on post-annealing microstructure prediction. Furthermore, the method further includes: evaluating the quality of the ballistic steel strip after annealing and feeding the evaluation results back to iteratively optimize the weights in the multi-objective optimization function. This step adds a higher-level, cross-batch learning and evolutionary closed loop to the original process closed-loop control. By feeding the final product quality data back to the system, the system can self-learn and optimize its control objectives (i.e., weight coefficients), thereby enabling the entire production system to continuously improve.

[0132] Please see Figure 1This invention also provides an adaptive control device for annealing bulletproof steel strips in a shield-like manner. This device serves as the physical embodiment of the above-described method and typically manifests as a control system 30 integrating hardware and software. The device includes: a model module 31, internally configured with the aforementioned multiphysics coupling model describing the state evolution of the bulletproof steel strip during shield-like annealing; a target setting module 32, used to set a multi-objective optimization function guided by the predicted final service performance indicators of the bulletproof steel strip; an adaptive control module 33, used to adaptively control the annealing process based on the multiphysics coupling model and the multi-objective optimization function; and an online correction module 34, used for online monitoring, comparison, and correction of model parameters during the annealing process.

[0133] In a preferred embodiment, the multiphysics coupling model in the device specifically includes a temperature field model, a tissue evolution model, and a stress field model. This corresponds to the preferred scheme in the aforementioned method, ensuring that the device has the ability to perform high-precision state prediction.

[0134] In another preferred embodiment, such as Figure 1 As shown, the device also includes a temperature sensor 20 connected to the online correction module 34. This temperature sensor is used to collect the measured temperature value, which serves as a process parameter, in real time. Specifically, the online correction module 34 compares the measured temperature value collected by the temperature sensor 20 with the predicted temperature value at the corresponding sensor location, generated in real time by the model module 31 based on the current process state. This structure enables the device to interact with the physical world in real time and provide closed-loop feedback, making it a key hardware configuration for achieving adaptive control.

[0135] This invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor (e.g., a central processing unit within the control system 30), the aforementioned adaptive control method can be implemented. This allows the core technical solution of this invention to be distributed and deployed as a software product, protecting the algorithm and model itself, which are the core of the invention.

[0136] The technical solution of the present invention will now be described in more detail with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to more clearly illustrate the technical solution of the present invention, and are not intended to limit the scope of protection of the present invention.

[0137] In one specific embodiment, the adaptive control method described in this invention is applied to the annealing production of 1500MPa grade bulletproof steel strip. The chemical composition of the bulletproof steel strip, by mass percentage, is: C 0.32%, Si 0.25%, Mn 1.20%, Cr 1.50%, Mo 0.45%, Ni 1.80%. The equipment used is a bell-type annealing furnace 10, equipped with a set of... Figure 1 The control system 30 shown is pre-built in this control system 30 with a multiphysics coupling model for this grade of steel strip. This model describes the temperature field through the following unsteady heat conduction differential equation:

[0138]

[0139] Where ρ is density, denoted by , where T is the specific heat capacity, t is the temperature, t is the time, r is the radial coordinate, z is the axial coordinate, and k is the thermal conductivity. The model also includes recrystallization kinetics and phase transformation kinetics models for this material.

[0140] The calculation process for the tissue evolution model, stress field model, and multiphysics coupling model is explained in detail below.

[0141] (a) Organizational Evolution Model

[0142] The microstructure evolution model is used to describe the evolution of the microstructure of bulletproof steel strips during annealing, and includes at least a recrystallization kinetic model and a grain growth model. The recrystallization kinetic model uses the Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation to describe the relationship between recrystallization volume fraction and annealing time. The grain growth model uses the Sellars model to describe the growth of grain size with temperature and time after recrystallization.

[0143] 1. Recrystallization kinetic model

[0144] The recrystallization kinetic model uses the JMAK equation to describe the recrystallization volume fraction under isothermal annealing conditions. Annealing time Relationship:

[0145]

[0146] in:

[0147] This represents the recrystallization volume fraction, with a value ranging from 0 to 1. The annealing holding time is expressed in seconds (s). This is the recrystallization rate constant, which is related to temperature, material composition, and deformation amount, and its unit is 1. ; The Avrami index is related to nucleation mechanisms and growth patterns, and typically ranges from 1 to 4.

[0148] For ease of engineering application, the above equation can also be rewritten with a 50% recrystallization time. For reference parameter format:

[0149]

[0150] in:

[0151] The time required to achieve a 50% recrystallization volume fraction, expressed in seconds (s). To match the Avrami index Related constants.

[0152] rate constant With temperature The relationship between them follows an Arrhenius-type equation:

[0153]

[0154] in:

[0155] This is the frequency factor, a material-related constant, with units of 1000 ppm. ; The recrystallization activation energy is expressed in joules per mole (J / mol) and reflects the energy barrier required for the recrystallization process. Let be the ideal gas constant, and take the value of . J / (mol·K); Temperature is the thermodynamic temperature, and the unit is Kelvin (K).

[0156] Recrystallization time 50% Alternatively, the following empirical relation can be used to describe it directly:

[0157]

[0158] in:

[0159] The initial grain size is expressed in micrometers (μm). For cold rolling cumulative strain (dimensionless); These are material constants; This refers to the grain size index; The strain index.

[0160] The undetermined parameters in the above model ( , , , , , , (etc.) Isothermal annealing experiments were conducted on standard samples of this grade of bulletproof steel strip under different temperature and time conditions. The recrystallization volume fraction under each condition was determined by metallographic analysis, calibrated using nonlinear regression analysis, and stored in model module 31 of control system 30. For non-isothermal annealing processes (such as heating and cooling stages), the recrystallization kinetic model uses the superposition principle for numerical integration, discretizing the continuous temperature variation process into several small isothermal time steps, and calculating the increment of recrystallization volume fraction step by step.

[0161] 2. Grain growth model

[0162] After recrystallization, the grains will grow during subsequent holding periods. The Sellars model is used to describe the evolution of average grain size with increasing holding temperature and time.

[0163]

[0164] in:

[0165] The average grain size after annealing is expressed in micrometers (μm). The initial average grain size at the time of recrystallization is complete, in micrometers (μm). The holding time after recrystallization is complete is expressed in seconds (s). Temperature is thermodynamic temperature, and the unit is Kelvin (K). This refers to the grain growth constant related to the material. It is the grain growth index, which is related to the material and the grain growth mechanism. The activation energy for grain growth is expressed in joules per mole (J / mol). Let be the ideal gas constant, and take the value of . J / (mol·K).

[0166] When the initial grain size When the size is much smaller than the grown grain size, the above equation can be simplified to:

[0167]

[0168] Taking the logarithm of both sides of the above equation, we get:

[0169]

[0170] This linearized form facilitates the determination of model parameters through regression analysis of experimental data. , and .

[0171] The undetermined parameters in the above model ( , , (etc.) Isothermal grain growth experiments were conducted on standard samples of the bulletproof steel belt under different temperature and time conditions. The average grain size under each condition was determined by metallographic analysis, calibrated by linear regression analysis, and stored in the model module 31 of the control system 30.

[0172] 3. Coupled calculations of organizational evolution models

[0173] During annealing, recrystallization and grain growth occur sequentially. The calculation process of the microstructure evolution model is as follows: First, based on the historical temperature data of each discrete node at different times provided by the temperature field model, the recrystallization volume fraction of each node is calculated using the recrystallization kinetics model. ;when When recrystallization at that node is deemed essentially complete, the grain growth model is then used to calculate the grain size changes during the subsequent heat preservation process. Through the above coupled calculations, the microstructure (recrystallization fraction, grain size, etc.) at any location inside the steel coil can be predicted in real time throughout the entire annealing process.

[0174] (II) Stress Field Model

[0175] The stress field model is used to describe the distribution and evolution of internal stress in bulletproof steel strips caused by temperature inhomogeneity and phase transformation during annealing. The stress inside the steel coil during annealing mainly consists of two parts: thermal stress (caused by uneven thermal expansion due to temperature gradients) and structural stress (caused by volume changes due to phase transformations). The stress field model uses the thermo-elastic-plastic constitutive equation to describe the stress-strain relationship.

[0176] 1. Total strain decomposition

[0177] Based on the small deformation assumption, the total strain increment at any point within the steel strip is... It can be decomposed into the sum of the following components:

[0178]

[0179] in:

[0180] This represents the elastic strain increment; This represents the increment of plastic strain. This is the thermal strain increment, generated by thermal expansion and contraction caused by temperature changes; The phase transformation strain increment is generated by the volume change caused by the microstructure transformation (such as recrystallization, phase transformation); This is the increment of plastic strain during phase transformation, generated by plastic flow during the phase transformation process.

[0181] 2. Thermal strain

[0182] Thermal strain increment Caused by temperature changes, the calculation formula is:

[0183]

[0184] in:

[0185] The instantaneous linear expansion coefficient of the material is given in units of 1000 ppm. It is temperature The function; This represents the temperature increment, expressed in degrees Celsius (°C). Kronecker symbol (when) hour ,when hour ).

[0186] 3. Phase transition strain

[0187] Phase transition strain increment The phase transformation strain arises from the volume change caused by the structural transformation. For the recrystallization process, the phase transformation strain can be expressed as:

[0188]

[0189] in:

[0190] The relative volume change rate (dimensionless) before and after recrystallization was determined by thermal expansion experiments. This represents the increase in recrystallization volume fraction.

[0191] In more general phase transformation cases (such as martensitic phase transformations that may occur during cooling), the phase transformation strain can be expressed as:

[0192]

[0193] in:

[0194] For the first The characteristic strain tensor corresponding to a phase transition; For the first The increase in the volume fraction of the seed phase; This represents the total number of phase transition types.

[0195] 4. Thermo-elastic-plastic constitutive equations

[0196] Stress increment With elastic strain increment The relationship between them follows the generalized Hooke's Law:

[0197]

[0198] in:

[0199] This is the elastic-plastic stiffness tensor, which is related to the material's current temperature, microstructure, and loading history. This represents the total strain increment.

[0200] After the material enters the plastic state, the yield condition adopts the von Mises yield criterion:

[0201]

[0202] in:

[0203] Let be the yield function; For von Mises equivalent stress, , It is the stress deviator tensor; These are internal variables (such as cumulative plastic strain) used to describe work hardening / softening; The yield stress under the current condition is an internal variable. ,temperature and recrystallization fraction The function.

[0204] 5. Equilibrium Equations and Boundary Conditions

[0205] The stress field satisfies the static equilibrium equations (ignoring inertial forces):

[0206]

[0207] in:

[0208] Spatial coordinates; These are volume force components (such as gravity).

[0209] Displacement boundary conditions and force boundary conditions are applied to the contact surface between the steel coil and the inner cover; the free surface of the steel coil satisfies the stress-free boundary condition.

[0210] Material parameters (elastic modulus) in the above stress field model Poisson's ratio Coefficient of linear expansion Yield stress All values ​​(e.g., temperature, microstructure) are functions of temperature and microstructure. Their specific values ​​are obtained through thermophysical parameter testing and mechanical property testing of the bulletproof steel strip at different temperatures, and are stored in the model module 31 of the control system 30. The stress field model, temperature field model, and microstructure evolution model achieve multi-physics collaborative simulation through coupled calculations: the temperature field model provides temperature distribution data to the stress field model, the microstructure evolution model provides recrystallization fraction and phase transformation progress data to the stress field model, and the calculation results (internal stress distribution) of the stress field model can be used to assess the deformation and cracking risks during the annealing process.

[0211] (III) Calculation process of multiphysics coupling model

[0212] In the multiphysics coupling model, the temperature field model, microstructure evolution model, and stress field model are coupled for calculation in the following manner: Within each time step, the temperature field model first calculates the transient temperature distribution inside the steel coil; then, the temperature history of each node is input into the microstructure evolution model to calculate the recrystallization fraction and grain size; finally, the temperature field and microstructure data are input into the stress field model to calculate the thermal stress and microstructure stress distribution. The coupled calculation employs the finite element method for spatial discretization and time integration, with the steel coil divided radially into... Layers, divided along the axial direction Layer, at each discrete node Simultaneously solve the governing equations of each physical field.

[0213] The multi-objective optimization function set in the target setting module 32 of the control system 30 is as follows:

[0214]

[0215] Where F is the comprehensive evaluation value, To predict the ballistic impact resistance index based on the microstructure after annealing, This is an indicator of the uniformity of the microstructure of steel coils after annealing. ω1, ω2, and ω3 are weighting coefficients representing the energy consumption of the annealing process. In this embodiment, performance and uniformity are emphasized, therefore ω1 and ω2 are assigned higher values.

[0216] In terms of the process flow, firstly, four coils 11 of bulletproof steel strip of this grade, each with an outer diameter of 1800mm, an inner diameter of 610mm, a width of 1200mm, and a thickness of 2.5mm, are loaded into the furnace platform of the bell-type annealing furnace 10. Simultaneously, a high-temperature resistant temperature sensor 20 is placed inside one of the steel coils to monitor the internal temperature of the coil in real time. The adaptive control module 33 of the control system 30 automatically generates an initial segmented annealing strategy based on the input steel coil parameters and optimization objectives. Please refer to [link / reference]. Figure 3This strategy employs variable-rate heating during the heating phase. In the low-temperature range from room temperature to 400°C, a heating rate of 70°C / h is used to improve efficiency; in the high-temperature range from 400°C to the target annealing temperature of 680°C, a lower heating rate of 40°C / h is switched to reduce the temperature of the hottest spot, T, of the steel coil. hot With the coldest point temperature T cold The temperature difference between them. The heating method uses pulse combustion control to precisely execute this variable rate curve.

[0217] During operation, when the adaptive control module 33 determines the coldest point temperature T of the steel coil based on the calculation of the multiphysics coupling model... cold Once the target annealing temperature of 680℃ is reached, the system automatically enters the homogenization stage and begins timing. This differs from traditional processes that rely on furnace temperature T. furnace They are different, such as Figure 3 As shown, the furnace temperature T at this time furnace The temperature may have reached 690℃, but the internal temperature of the steel coil has only just reached the target. During the homogenization process, the model continuously predicts the recrystallization fraction inside the steel coil. When the predicted value reaches 97% (this process actually takes 6.2 hours), the adaptive control module 33 automatically issues a command to end the homogenization and enter the cooling stage. Throughout the process, the online correction module 34 continuously compares the measured temperature fed back by the temperature sensor 20 with the model's predicted value (corresponding to...). Figure 4 (S201-S204 process), since the deviation in this embodiment is always within the threshold of ±3℃, large-scale model parameter correction is not triggered (S205).

[0218] The cooling stage employs a three-stage strategy. First, slow cooling with a cooling hood is performed, reducing the temperature from 680℃ to 560℃ in approximately 10 hours. Then, forced air cooling is applied with the cooling hood replaced, reducing the temperature to 280℃ in approximately 10 hours. Finally, water cooling is used, rapidly reducing the temperature to 70℃ before the coil is removed from the furnace. The switching between each cooling stage is automatically triggered by the adaptive control module 33 based on the internal temperature of the steel coil. After annealing, samples were taken from different positions on the steel coil 11 from the outer to the inner ring for testing. The maximum deviation in Brinell hardness was only 6 HB, indicating good annealing uniformity. After subsequent quenching and tempering, the tensile strength of this batch of steel strip reached 1550 MPa, and the impact energy at -40℃ was greater than 27 J, meeting the predetermined Class C ballistic performance requirements and verifying the accuracy and effectiveness of the method of this invention.

[0219] In an alternative implementation, the target can be adjusted and optimized to meet the production requirements of products with higher ballistic protection levels. This embodiment uses the same equipment and steel strip material as the previous embodiments, but in the target setting module 32 of the control system 30, the operator significantly improves the ballistic impact resistance performance index. The weighting coefficient ω1 was adjusted, and the energy consumption index was appropriately reduced. The weight ω3 is applied. Upon receiving the new weight, the adaptive control module 33 recalculates the model prediction and optimization. The model results indicate that a lower annealing temperature and a longer holding time should be used to obtain finer recrystallized grains and improve the final low-temperature toughness. Therefore, the system automatically generates a new set of process parameters: the target annealing temperature is adjusted to 660℃, and the predicted soaking time is extended to 8.5 hours. In actual operation, the soaking stage ends after the model predicts that the recrystallization fraction reaches 98%, and the actual time is 8.8 hours. After annealing, the hardness uniformity of the steel coil is better, and the deviation is controlled within ±4HB. After subsequent quenching and tempering using the same process, the tensile strength of the material remains at 1540MPa, but the impact energy at -40℃ is significantly increased to 35J, exhibiting better low-temperature toughness and meeting the requirements of a higher bulletproof level. This embodiment demonstrates that by adjusting the objective function, the present invention can flexibly customize and optimize the process for products with different performance requirements.

[0220] Furthermore, the method of the present invention is equally effective even under limited monitoring conditions. In another embodiment, assuming that the production site does not have temperature sensors 20 installed with the furnace, the monitoring system can only use conventional thermocouples arranged on the furnace chamber, inner shroud, and outer surface of the steel coil 11. In this case, the role of the multiphysics coupling model becomes more critical. The control system 30 relies entirely on the model, using limited measured boundary temperature data to invert the internal temperature field of the entire steel coil. In the early stages of annealing, because model parameters (such as interlayer contact thermal resistance, convective heat transfer coefficient, etc.) may deviate from reality, the online correction module 34 will frequently trigger the online correction mechanism based on the difference between the measured surface temperature of the steel coil and the surface temperature predicted by the model, such as... Figure 4 As shown, when the deviation exceeds the threshold (S204), step S205 is executed to continuously iterate and update the model parameters, enabling the model prediction results to converge quickly to the actual situation. To ensure the annealing effect, when the model uncertainty is high, the adaptive control module 33 will automatically adopt a more conservative strategy. For example, when judging the end of homogenization, the recrystallization fraction threshold will be increased to 99%, and an additional 10% safety margin time will be added to ensure that the coldest point of the steel coil can also complete sufficient annealing. Finally, the hardness deviation of the annealed steel coil is ±9HB. Although the uniformity is inferior to the embodiment equipped with internal temperature measurement, it is still a significant improvement compared to the traditional control method that relies on manual experience (usually the deviation is above ±15HB), proving that the method of the present invention has good adaptability and robustness.

[0221] In another preferred embodiment, the adaptive control of the present invention can be applied to more finely regulate the cooling stage. This embodiment uses a Cr-Mo-V bulletproof steel sensitive to cooling rate, whose microstructure evolution model includes a kinetic model of secondary hardening and brittle phase precipitation at different cooling rates. The heating and soaking stages are similar to the basic embodiment. Upon entering the cooling stage, the adaptive control module 33 aims to perform refined nonlinear cooling control. In the first cooling stage (slow cooling with a hood), the model aims to control the cooling rate at 10-15°C / h to promote the precipitation of dispersed carbides, providing a favorable initial microstructure for subsequent quenching. The system achieves precise control of the slow cooling rate by finely adjusting the furnace cooling water flow rate. Upon entering the second cooling stage (air cooling), the microstructure evolution model predicts a risk of brittle phase precipitation in the 450-550°C temperature range. Therefore, the adaptive control module 33 dynamically adjusts the rotation speed of the cooling hood fan, allowing the steel coil's cooling curve to quickly cross this dangerous temperature range, increasing the cooling rate in this range to 25-30°C / h. Once the temperature drops below the danger zone, the wind speed is appropriately reduced to decrease thermal stress. Through refined and nonlinear control of the cooling process, the annealed steel strip not only achieves a uniform recrystallized structure but also successfully suppresses the formation of brittle phases and obtains a favorable carbide distribution. This results in a significant improvement in the toughness and fatigue resistance of the final product while ensuring high strength.

Claims

1. A method for controlling the annealing of bulletproof steel with a shield for processing ultra-high strength steel, characterized in that, Includes the following steps: For bulletproof steel strips, a multiphysics coupling model is constructed to describe their state evolution during the shroud annealing process; A multi-objective optimization function is defined, guided by the predicted final service performance indicators of the ballistic steel belt; Based on the multiphysics coupling model and the multi-objective optimization function, adaptive control is performed on the annealing process, which includes heating, homogenization and cooling stages. During the annealing process, the measured value of at least one process parameter is monitored online. The measured value of the process parameter is compared with the predicted value of the process parameter at the same measurement location obtained based on the multiphysics coupling model. When the deviation between the two exceeds a preset threshold, the parameters of the multiphysics coupling model are corrected online, and the subsequent control strategy is dynamically adjusted based on the corrected model.

2. The method according to claim 1, characterized in that, The multiphysics coupling model includes at least a temperature field model, a tissue evolution model, and a stress field model.

3. The method according to claim 1 or 2, characterized in that, The multi-objective optimization function includes ballistic performance index, microstructure uniformity index, and energy consumption index of the annealing process, all based on the predicted microstructure after annealing.

4. The method according to claim 1, characterized in that, The steps for online monitoring of the measured value of at least one process parameter include: Temperature sensors are arranged inside the bell-type annealing furnace to collect the measured temperature values, which are used as the process parameters, in real time. Furthermore, the comparison step specifically involves comparing the measured temperature value with the predicted temperature value at the corresponding sensor location, which is predicted in real time by the multiphysics coupling model based on the current process state.

5. The method according to claim 1, characterized in that, in, The steps for adaptive control of the annealing process include employing a variable heating rate strategy during the heating phase.

6. The method according to claim 5, characterized in that, The variable heating rate strategy includes: using a heating rate of 60-80℃ / h in the low temperature range below a preset temperature threshold, and using a heating rate of 30-50℃ / h in the high temperature range above or equal to the preset temperature threshold.

7. The method according to claim 1, characterized in that, in, The steps for adaptive control of the annealing process include the following during the homogenization stage: determining the start and end times of homogenization based on the predictions of the coldest point temperature and recrystallization fraction of the steel coil by the multiphysics coupling model; and ending homogenization when the recrystallization fraction predicted by the model reaches ≥95%.

8. The method according to claim 7, characterized in that, The homogenization stage is carried out in a full hydrogen protective atmosphere with hydrogen purity ≥99.999%, and the furnace pressure is maintained at 8-12 kPa.

9. The method according to claim 1, characterized in that, in, The adaptive control steps for the annealing process include employing a multi-stage cooling strategy during the cooling stage; the multi-stage cooling strategy includes three stages: slow cooling with a shroud, air cooling, and water cooling.

10. The method according to claim 1, characterized in that, The multi-objective optimization function includes ballistic impact resistance index, microstructure uniformity index, and annealing process energy consumption index based on the predicted microstructure after annealing; the method further includes: performing a quality evaluation on the ballistic steel strip after annealing, and feeding back the evaluation results to iteratively optimize the weights in the multi-objective optimization function.

Images