A production method of hot-rolled wire rod for cold heading steel without annealing

Abstract

The present invention discloses a method for producing annealing-free hot-rolled cold-heading steel wire rod, which relates to the technical field of steel production. The method comprises: selecting a steel billet that meets the production specifications of the cold-heading steel wire rod, and performing grinding, finishing, and preheating on the steel billet to be processed; conveying the steel billet to a high-compression ratio rolling mill, monitoring the hot rolling process in real time, and performing segmented hot-rolling control to achieve multi-pass temperature gradient rolling to obtain hot-rolled cold-heading steel wire rod; conveying the hot-rolled cold-heading steel wire rod to an insulated tunnel furnace, and obtaining a finished cold-heading steel wire rod through segmented controlled cooling and surface treatment, thereby achieving the production of annealing-free hot-rolled cold-heading steel wire rod. The present invention utilizes an advanced high-compression ratio rolling mill to produce cold-heading steel wire rods of different specifications, and utilizes an insulated tunnel furnace to control the wire rod cooling temperature and cooling rate, thereby achieving the production of annealing-free hot-rolled cold-heading steel wire rod. This greatly simplifies the production process, shortens the production cycle, reduces production costs, and improves production efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of steel production, in particular to a production method of annealing-free cold heading steel hot-rolled wire rod. Background Art

[0002] Hot rolling is a crucial process in steel production technology. It involves rolling steel billets at high temperatures to reduce their thickness and increase their length, ultimately producing steel of the desired shape and size. Cold-heading steel is a high-quality carbon structural steel primarily used to manufacture standard parts such as bolts, nuts, and bearing rings. It is characterized by high strength, toughness, and wear resistance. However, the traditional cold-heading steel production process requires annealing to improve its structure and properties, which increases production costs and prolongs the production cycle.

[0003] To address this issue, some companies have adopted a method for producing cold-headed steel without annealing. This method primarily modifies the composition and microstructure of the steel billet, enabling it to achieve the desired properties without the need for annealing. Additionally, some companies employ specialized rolling processes, such as high-reduction-ratio rolling, to enhance the strength and toughness of cold-headed steel.

[0004] However, existing methods for producing annealing-free cold-heading steel still present several challenges. First, these methods often require specialized equipment and complex processes, which increases production costs. Second, these methods impose stringent requirements on the composition and microstructure of the steel billets, which limits their application. Finally, the performance of the cold-heading steel produced by these methods still needs to be improved, particularly in terms of strength and toughness.

[0005] Therefore, how to produce high-performance annealing-free cold heading steel through a simple process without changing the composition and organizational structure of the steel billet remains an urgent problem to be solved. Summary of the Invention

[0006] Based on this, it is necessary to provide a production method for annealing-free cold heading steel hot-rolled wire rod to address the above technical problems.

[0007] The present invention provides a method for producing annealing-free cold heading steel hot-rolled wire rod, comprising:

[0008] S1. Select steel billets that meet the production specifications of cold heading steel wire rods, and perform grinding, finishing and preheating on the steel billets to be processed;

[0009] S2. The steel billet is transported to a high-compression ratio rolling mill, and the hot rolling process is monitored in real time to perform segmented hot rolling control, thereby achieving multi-pass temperature gradient rolling to obtain hot-rolled cold-headed steel wire rod;

[0010] S3. The hot-rolled cold-heading steel wire rod is transported to a heat-insulating tunnel furnace, and the cold-heading steel wire rod is obtained through segmented controlled cooling and surface treatment, thereby realizing the production of annealing-free cold-heading steel hot-rolled wire rod.

[0011] Furthermore, the steel billet that meets the production specification requirements of cold heading steel wire rod is selected, and the steel billet to be processed is subjected to grinding, finishing and preheating treatment, including:

[0012] S11. Preset the production specifications of the cold heading steel wire rod, select a steel billet of corresponding size, and detect the content of each component element in the steel billet using a laser spectrometer;

[0013] S12, using two grinding wheels to polish the billet, the first round uses a 16-20 mesh grinding wheel to remove wrinkles on the corners of the billet, and the second round uses a 24-26 mesh grinding wheel to finely polish the billet surface;

[0014] S13. Place the ground steel billet in a preheating environment at 600-800°C, gradually heat the preheating environment to 1090-1170°C, and introduce a CO2 / N2 mixed gas to increase the surface hardness of the steel billet.

[0015] Furthermore, the steel billet is transported to a high-compression ratio rolling mill, and the hot rolling process is monitored in real time to perform segmented hot rolling control, thereby achieving multi-pass temperature gradient rolling to obtain a hot-rolled cold heading steel wire rod.

[0016] S21. Based on the production specifications of the cold heading steel wire rod and the composition of the steel billet, the rolling process is divided into three stages: rough rolling, intermediate rolling, and finishing rolling. The physical constraint meta-learning algorithm is used to match the optimal rolling parameters.

[0017] S22, real-time monitoring of hot rolling data during the hot rolling process, dynamic compensation of optimal rolling parameters, and dynamic adjustment of the pass interval time according to the monitoring and compensation results;

[0018] S23. After rolling is completed, a nano-Al2O3-MgO composite coating is sprayed on the surface of the cold heading steel wire rod, and a dense oxide film is generated using the waste heat. The actual production specifications of the cold heading steel wire rod are then verified.

[0019] Furthermore, based on the production specifications of cold heading steel wire rod and the composition of the steel billet, the rolling process is divided into three stages: rough rolling, intermediate rolling, and finishing rolling. The physical constraint meta-learning algorithm is used to match the optimal rolling parameters, including:

[0020] S211. Based on the preset production specifications of the cold heading steel wire rod, reversely deduce the cross-sectional dimensions of the steel billet and calculate the target compression ratio of the steel billet to be processed by hot rolling to the cold heading steel wire rod;

[0021] S212, dividing the rolling process of the steel billet into three stages: rough rolling, intermediate rolling, and finishing rolling; establishing a parameter mapping table using a historical database; and outputting the initial compression ratio and rolling temperature range of each stage;

[0022] S213. Establish physical constraints for hot rolling of steel billets based on metallurgical theory. The physical constraints include phase transformation dynamics constraints, deformation resistance and rolling force constraints, and precipitation phase pinning constraints.

[0023] S214. Set input variables and output variables, build and train a parameter recommendation model based on meta-learning, integrate physical metallurgical constraints, and match rolling parameters corresponding to different steel billets;

[0024] S215. Input the size and component content of the steel billet to be processed into the parameter recommendation model, generate rolling parameters for each stage and adjust the logic in real time to obtain the optimal rolling parameters.

[0025] Furthermore, we set input and output variables, build and train a parameter recommendation model based on meta-learning, integrate physical metallurgical constraints, and match rolling parameters corresponding to different steel billets, including:

[0026] S2141. Assign each original task to a type of steel, set the production specifications of cold heading steel wire rod and the content of each component in the steel billet as input variables, and the rolling parameters as output variables to construct a support set;

[0027] S2142. Linking a physical calculation module after the hidden layer output to implement physical information embedding for calculating the temperature and reduction rate of billet rolling at each stage;

[0028] S2143. Setting two physical loss term functions, phase change loss and rolling force loss, and combining the phase change loss and rolling force loss to set a comprehensive loss function, thereby constructing a parameter recommendation model;

[0029] S2144. Pre-train model parameters on all steel grade tasks to initially form a hot rolling knowledge base, and when facing new steel grades, fine-tune the calculation tasks through single-step gradient descent.

[0030] Furthermore, the size and composition of the steel billet to be processed are input into the parameter recommendation model to generate rolling parameters for each stage and adjust the logic in real time. The optimal rolling parameters include:

[0031] S2151. Obtain the carbon content of the steel billet to be processed, match the rough rolling temperature corresponding to the rough rolling stage according to the carbon content value, and calculate the rough rolling reduction rate;

[0032] The calculation formula for rough rolling temperature is:

[0033] ;

[0034] Where, T粗轧 Indicates the rough rolling temperature in the rough rolling stage; T0 indicates the preset initial temperature; [C] indicates the carbon content;

[0035] The calculation formula for the rough rolling reduction rate is:

[0036] ;

[0037] Where r 粗轧 represents the rough rolling reduction rate in the rough rolling stage; r1 represents the initial compression ratio in the rough rolling stage output by the parameter mapping table; R total Indicates the target compression ratio;

[0038] S2152, obtaining the titanium content and manganese content of the steel billet to be processed and the grain size after rough rolling, and calculating the intermediate rolling cooling rate and intermediate rolling reduction rate of the steel billet entering the intermediate rolling stage;

[0039] The calculation formula for the intermediate rolling cooling rate is:

[0040] ;

[0041] Where, v T Indicates the cooling rate of intermediate rolling; Δ T Indicates the difference between rough rolling temperature and medium rolling temperature; Δ t Indicates the time difference; [ Mn ] indicates the manganese content;

[0042] The calculation formula for the intermediate rolling reduction rate is:

[0043] ;

[0044] Where, r 中轧 Indicates the intermediate rolling reduction rate; r 2 represents the initial compression ratio of the intermediate rolling stage output by the parameter mapping table; [ Ti ] indicates titanium content;

[0045] S2153, calculating the Ar3 phase transformation point of the steel billet to be processed, setting the finishing temperature in the finishing rolling stage according to the Ar3 phase transformation point, and calculating the finishing rolling reduction rate in the finishing rolling stage according to the carbon content;

[0046] The calculation formula for finishing rolling reduction is:

[0047] ;

[0048] Where r 精轧 represents the finishing reduction rate; r3 represents the initial compression ratio of the finishing stage output by the parameter mapping table.

[0049] Furthermore, the hot rolling data in the hot rolling process is monitored in real time, the optimal rolling parameters are dynamically compensated, and the pass interval time is dynamically adjusted according to the monitoring and compensation results, including:

[0050] S221, using sensors inside the rolling mill group to collect hot rolling data of the billet rolling process in real time;

[0051] S222, calculating the difference between the actual rolling force and the target rolling force of the rolling mill group during each stage of rolling, and calculating the reduction rate compensation amount through the proportional-integral coefficient;

[0052] S223, calculating the temperature difference between the beginning and the end of the billet at each stage. If the temperature difference between the beginning and the end is greater than a preset threshold, the pass interval is extended to balance the temperature drop and the rolling rhythm.

[0053] S224. Import the real-time hot rolling data into the historical parameter library, set the update cycle, and regularly update the comprehensive loss function. If there is a sudden change in the hot rolling data, an emergency stop signal is triggered.

[0054] Furthermore, the hot-rolled cold-heading steel wire rod is transported to an insulated tunnel furnace, and through segmented controlled cooling and surface treatment, a finished cold-heading steel wire rod is obtained. The production of annealing-free cold-heading steel hot-rolled wire rod includes:

[0055] S31, obtaining and recording the parameters of the cold heading steel wire rod after hot rolling, the wire rod parameters including the final rolling temperature, the grain size after rolling, the predicted ferrite ratio, and the rolling force curve;

[0056] S32. Divide the cooling process of the cold heading steel wire rod into three stages: rapid cooling, slow cooling, and final cooling. Build a physical constraint meta-learning framework to generate cooling parameters to control the cooling of the cold heading steel wire rod in sections.

[0057] S33. Dephosphorize the cooled cold heading steel wire rod using high pressure water, and identify surface defects through online laser detection. After the detection is correct, obtain the finished cold heading steel wire rod.

[0058] Furthermore, the cooling process of cold-heading steel wire rod is divided into three stages: rapid cooling, slow cooling, and final cooling. A physical constraint meta-learning framework is built to generate cooling parameters for segmented control of cold-heading steel wire rod cooling, including:

[0059] S321. Introduce wire rod parameters to expand the input layer, embed phase change dynamics equations and heat conduction equations to build a physical constraint layer, and build an output layer to output segmented cooling rate values and final cooling temperature;

[0060] S322, dividing the cooling process of the cold heading steel wire rod into rapid cooling, slow cooling and final cooling stages, and calculating the cooling rate and cooling time of the rapid cooling stage respectively;

[0061] S323. Based on the preset segmented air cooling strategy, a fan-water mist combination cooling is adopted in the rapid cooling stage, the fan is turned off in the slow cooling stage, and natural cooling is carried out by roller conveyor. In the final cooling stage, natural cooling is adopted and a nitrogen-carbon dioxide mixture is introduced to form a dense oxide film.

[0062] Furthermore, the calculation of the cooling rate and cooling time in the rapid cooling stage includes:

[0063] The calculation formula of the rapid cooling rate in the rapid cooling stage is:

[0064] ;

[0065] Where, v 速冷 Indicates the rapid cooling rate in the rapid cooling stage; T 终轧 Indicates the final rolling temperature; T 速冷终 Indicates the target temperature in the rapid cooling stage; t 速冷 Indicates rapid cooling time;

[0066] The calculation formula for cooling time in the slow cooling stage is:

[0067] ;

[0068] Where, t 缓冷 represents the slow cooling time; k0 represents the phase change rate constant; represents the phase change activation energy; R represents the gas constant; T 缓冷 Indicates slow cooling temperature; X total represents the target ferrite ratio; n represents the Avrami index; and e represents the natural constant.

[0069] The beneficial effects of the present invention are:

[0070] 1. The present invention adopts an advanced rolling mill with a large compression ratio to produce cold heading steel wire rods of different specifications. With the help of an insulated tunnel furnace, the cooling temperature and cooling rate of the wire rods are controlled, thereby realizing the production of annealing-free cold heading steel hot-rolled wire rods, greatly simplifying the production process, shortening the production cycle, reducing production costs, and improving production efficiency; the produced cold heading steel wire rods do not need to undergo annealing treatment, avoiding the changes in organizational structure and performance that may be caused by annealing treatment, thereby improving the quality stability of the product; at the same time, the present invention has no strict requirements on the composition and organizational structure of the steel billet, so it can be applied to more steel grades, expanding its application range; by precisely controlling the cooling temperature and cooling rate, the strength and toughness of the product are significantly improved, and it has better performance.

[0071] 2. Through multi-pass temperature gradient rolling, the temperature changes in each stage are precisely controlled, and the segmented controlled cooling process is combined to achieve the directional phase transformation from austenite to ferrite and the spheroidization of pearlite, breaking through the limitations of traditional processes; in the rolling process, through the synergistic effect of differentiated temperature control and large reduction rate in the three stages of rough, medium and finish rolling, ultrafine grains and uniform structure are directly obtained; in the cooling process, rapid cooling is used to inhibit carbide coarsening, slow cooling is used to promote phase transformation balance, and the final cooling oxide film control technology is used to completely eliminate the need for annealing; ultimately, the integrated performance of high strength, high plasticity and corrosion resistance of cold heading steel wire rod is improved, while significantly reducing energy consumption and production costs, and promoting the upgrading of the steel hot rolling industry towards green and high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0073] Figure 1 1 is a flow chart of a method for producing annealing-free cold heading steel hot-rolled wire rod according to an embodiment of the present invention;

[0074] Figure 2 This is a typical microstructure characterization diagram of SCM435 steel before optimization according to an embodiment of the present invention;

[0075] Figure 3 This is a typical microstructure characterization diagram of SCM435 steel after optimization according to an embodiment of the present invention;

[0076] Figure 4 3. Mechanical properties diagram of SCM435 steel with different shear cycle numbers before optimization according to an embodiment of the present invention;

[0077] Figure 5 3. Graph showing mechanical properties of SCM435 steel after optimization at different shear cycles according to an embodiment of the present invention;

[0078] Figure 6 1 is a typical microstructure diagram of a B7 steel cold heading line before optimization according to an embodiment of the present invention;

[0079] Figure 7 1 is a typical organization diagram of a cold heading line of B7 steel after optimization according to an embodiment of the present invention. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0081] See also Figure 1, provides a method for producing annealing-free cold heading steel hot-rolled wire rod, comprising:

[0082] S1. Select steel billets that meet the production specifications of cold heading steel wire rods, and perform grinding, finishing and preheating on the steel billets to be processed.

[0083] In the description of the present invention, selecting a steel billet that meets the production specification requirements of cold heading steel wire rod, and performing grinding, finishing and preheating treatment on the steel billet to be processed includes:

[0084] S11. Preset the production specifications of the cold heading steel wire rod, select a steel billet of corresponding size, and detect the content of each component element in the steel billet using a laser spectrometer.

[0085] Specifically, before cold heading steel wire rod production, the billet size must be reverse-calculated based on the target wire rod specification (e.g., Φ16mm diameter) to ensure a compression ratio of ≥80%. For example, to produce Φ16mm wire rod, a 160mm×160mm square billet is required. The compression ratio is calculated as the ratio of the original cross-sectional area to the target cross-sectional area to ensure sufficient plastic deformation during the rolling process.

[0086] The laser spectrometer detects the composition of steel billets and analyzes the contents of key elements such as C, Mn, Cr, and Ti in real time. If the composition deviation exceeds the process requirements, unqualified billets are automatically rejected. For example, if a billet is detected to have a C content of 0.35% and a Mn content of 0.75%, which meets the SWRCH35K steel grade standard, it is considered qualified.

[0087] S12, using two-wheel grinding method, the first round uses 16-20 mesh grinding wheel to remove the wrinkles on the corners of the steel billet, and the second round switches to 24-26 mesh grinding wheel to finely polish the surface of the steel billet.

[0088] Specifically, a two-wheel grinding method is adopted. The first round uses a 16-20 mesh grinding wheel (coarser abrasive grain size), with a pressure of 7000-8000N and a feed speed of 0.8-1.2m / s, focusing on removing wrinkles at the corners of the steel billet and surface cracks (defects with a depth of ≥0.5mm need to be completely eliminated).

[0089] For example, a folding defect with a depth of 0.8mm at a corner was reduced to ≤0.2mm after the first round of grinding. A second round of polishing was then performed using a 24-26 grit grinding wheel (fine grit) at a pressure of 4000-5000N and a feed rate of 1.0-1.5m / s, achieving a surface roughness Ra ≤6.3μm while avoiding material loss due to excessive grinding. For example, after grinding, a steel billet had no visible cracks and a roughness of Ra = 5.8μm, meeting the surface quality requirements before rolling.

[0090] S13. Place the ground steel billet in a preheating environment at 600-800°C, gradually heat the preheating environment to 1090-1170°C, and introduce a CO2 / N2 mixed gas to increase the surface hardness of the steel billet.

[0091] Specifically, the reground billet is first placed in a preheating zone at 600-800°C and gradually heated through radiant heating (the time is adjusted according to billet size, such as 30 minutes for a 160mm square billet) to avoid thermal stress cracking caused by direct high-temperature heating. The furnace temperature is then raised to the austenitizing range of 1090-1170°C, and a CO2 / N2 mixture (CO2 content is 5%-8%) is introduced. The activated carbon atoms produced by the decomposition of CO2 at high temperature react with the steel surface, forming a carbon concentration gradient (the carbon content in the surface layer increases by 0.05%-0.10%), enhancing the surface hardness (HRB+3) and suppressing the formation of a decarburized layer.

[0092] For example, after this treatment, the surface hardness of a steel billet is increased from HRB 82 to HRB 85, and the thickness of the decarburized layer is only 0.03mm, meeting the performance requirements of annealing-free cold heading steel.

[0093] S2. The steel billet is transported to the large compression ratio rolling mill group, and the hot rolling process is monitored in real time to perform segmented hot rolling control, so as to realize multi-pass temperature gradient rolling and obtain hot-rolled cold heading steel wire rod.

[0094] In the description of the present invention, the steel billet is conveyed to a high-compression ratio rolling mill, the hot rolling process is monitored in real time, the hot rolling process is controlled in sections, and multiple temperature gradient rolling is achieved to obtain a hot-rolled cold-heading steel wire rod, which includes:

[0095] S21. According to the production specifications of cold heading steel wire rod and the composition content of steel billet, the rolling process is divided into three stages: rough rolling, intermediate rolling and finishing rolling, and the physical constraint meta-learning algorithm is used to match the optimal rolling parameters.

[0096] Physically constrained meta-learning is a hybrid intelligent algorithm that integrates metallurgical physics rules with data-driven meta-learning. Through a "learning to learn" mechanism, it extracts common patterns from historical data on multiple steel grades, enabling rapid adaptation to new tasks (such as parameter matching for new steel grades). During model training and inference, it enforces adherence to fundamental metallurgical laws (such as phase transformation dynamics and thermal-mechanical coupling equations), ensuring that recommended parameters conform to materials science principles. Its core approach is to embed key metallurgical equations used in cold heading steel production (such as phase transformation dynamics, thermal deformation resistance models, and oxide film growth laws) as hard constraints within the meta-learning framework. This ensures that the algorithm-generated process parameters are not only data-driven but also strictly adhere to materials science principles. For example, in rolling parameter optimization, the algorithm enforces that the final rolling temperature must be below the Ar3 phase transition point to prevent microstructure coarsening caused by austenite recrystallization. Furthermore, this meta-learning mechanism enables the model to rapidly generalize from a small amount of data on new steel grades, reducing the cost of traditional trial-and-error experiments.

[0097] The input variables of the physical constraint meta-learning algorithm include steel composition (C, Mn, etc.), production specifications (diameter, surface requirements), and billet parameters (size, initial hardness); the output variables include the temperature of each rolling stage (rough / medium / finishing rolling), compression ratio, and reduction rate.

[0098] In the production method of the present invention, the physical constraint meta-learning algorithm is embodied in the following aspects:

[0099] 1. Dynamic matching of parameters in the rolling stage

[0100] During multi-pass temperature gradient rolling, a physically constrained meta-learning algorithm calculates the temperature windows and reduction ratios for each stage of roughing, intermediate rolling, and finishing in real time, based on the billet composition and target wire rod specifications. For example, for high-carbon steel, the model automatically reduces the initial rolling temperature to 980-1030°C and increases the finishing reduction ratio to ≥60% to trigger deformation-induced ferrite transformation (DIFT) and directly generate an ultrafine grain structure. By embedding the rolling force-temperature coupling equation, the physically constrained meta-learning algorithm dynamically compensates for mill load fluctuations, ensuring uniform rod performance.

[0101] 2. Optimization of segmented controlled cooling process

[0102] During the cooling phase, a physically constrained meta-learning algorithm combines the post-rolling temperature field distribution with a phase transformation kinetics model to generate cooling rate and time parameters for rapid cooling, slow cooling, and final cooling. For example, if an abnormal temperature gradient is detected in a section of the wire rod, the model automatically extends the slow cooling time and adjusts the fan power to ensure a ferrite content of ≥76% and a pearlite spheroidization rate of ≥90%. Furthermore, the oxide film thickness diffusion equation is used to constrain the final cooling temperature and atmosphere control parameters to suppress surface decarburization.

[0103] 3. Cross-process collaborative control

[0104] A physically constrained meta-learning algorithm breaks down the data barriers between rolling and cooling processes, enabling parameter linkage across the entire process. For example, if the cooling stage detects an insufficient ferrite ratio, feedback signals trigger the rolling model to increase the finishing reduction ratio. Conversely, if an abnormal temperature drop occurs during the rolling phase, the cooling model proactively adjusts the fan on / off strategy. This closed-loop optimization mechanism significantly reduces the impact of process fluctuations on final performance.

[0105] In the description of the present invention, according to the production specifications of cold heading steel wire rod and the composition content of steel billet, the rolling process is divided into three stages: rough rolling, intermediate rolling and finishing rolling, and the physical constraint meta-learning algorithm is used to match the optimal rolling parameters, including:

[0106] S211. Based on the preset production specifications of the cold heading steel wire rod, reversely deduce the cross-sectional dimensions of the steel billet and calculate the target compression ratio of the steel billet to be processed when hot-rolled to the cold heading steel wire rod.

[0107] Specifically, based on the principle of constant volume of cold heading steel, assuming that the volume of metal remains basically unchanged during hot rolling (ignoring oxidation loss), that is:

[0108] ;

[0109] Where, A 0 and A f are the cross-sectional areas of the billet and the finished product, respectively. L 0 and L f is the length before and after rolling.

[0110] Target compression ratio is the ratio of the cross-sectional area before and after rolling, that is:

[0111] ;

[0112] Where, D 0 and D f The diameter of the billet and the finished product.

[0113] S212. Divide the rolling process of the steel billet into three stages: rough rolling, intermediate rolling, and finishing rolling. Use the historical database to establish a parameter mapping table to output the initial compression ratio and rolling temperature range of each stage.

[0114] Specifically, based on historical data and steel grade characteristics, the roughing / medium / finishing rolling stages are divided and the initial parameters are output. The following division criteria can be referred to:

[0115] Rough rolling stage: large compression ratio (38-48%), high temperature zone (1050-1100℃), preferentially crushing the original austenite grains;

[0116] Medium rolling stage: medium compression ratio (30-40%), medium temperature zone (950-1000℃), grain refinement and control of precipitation phase;

[0117] Finishing rolling stage: supercritical compression ratio (≥60%), low temperature zone (750-800℃), triggering deformation induced ferrite transformation (DIFT).

[0118] To establish parameter mapping tables using a historical database, the present invention utilizes an automated data acquisition system to collect historical production data, including indicators such as billet composition, initial dimensions, rolling temperatures at each stage, compression ratio, rolling force, and the final product's grain size and mechanical properties. The collected data is cleaned and standardized to remove outliers and unify unit formats.

[0119] The present invention uses a clustering algorithm (such as K-means) to automatically group data according to steel grade composition and billet size to form data sets of different categories; on this basis, a multivariate linear regression or decision tree model is used to analyze the correlation between the rolling parameters (such as rough rolling temperature, compression ratio) at each stage in each data group and the target results (such as grain size, cold heading pass rate) to establish a mathematical model, namely a parameter mapping table.

[0120] When a new billet enters production, the system automatically matches its composition and size to the closest data category. It then uses the corresponding model to calculate the initial reduction ratio (e.g., 5:1 for roughing, 3.5:1 for intermediate rolling, and 2:1 for finishing) and the temperature range (e.g., 1080-1120°C for roughing, 980-1020°C for intermediate rolling, and 880-920°C for finishing) for roughing, intermediate rolling, and finishing. A real-time feedback mechanism is also implemented, transmitting actual production rolling performance data back to the database. The model is regularly retrained with new data to continuously optimize the accuracy of the parameter mapping table. The entire process is completely data-driven, requiring no human intervention.

[0121] It should be noted that step S212 provides statistically based initial recommended values for rolling parameters by converting historical production data into a structured knowledge base. Before the physical constraint meta-learning model intervenes, benchmark parameters that meet the common characteristics of steel grades are quickly generated to avoid starting calculations from scratch and shorten decision-making time. Using historical optimal parameters as initial values can effectively reduce parameter oscillations caused by random initialization of the model. In the subsequent process, the physical constraint meta-learning algorithm can dynamically correct various rolling parameters based on the initial parameters by combining physical constraints with real-time data feedback.

[0122] S213. Physical constraints for hot rolling of steel billets are established based on metallurgical theory. The physical constraints include phase transformation dynamics constraints, deformation resistance and rolling force constraints, and precipitation phase pinning constraints.

[0123] Specifically, the phase transition dynamics constraint formula is:

[0124] ;

[0125] Where X(t) represents the ferrite volume fraction; k(T) represents the temperature-dependent rate constant; t represents time; and n represents the Avrami index (dimensionless, typical value for cold heading steel is 1.5~2.0).

[0126] The deformation resistance and rolling force constraint formula is:

[0127] ;

[0128] Wherein, σ represents deformation resistance; σ0 represents reference stress; Q represents deformation activation energy; R represents gas constant; T represents absolute temperature; ε represents strain rate; ε0 represents reference strain rate; and m represents strain rate sensitivity coefficient.

[0129] The pinning constraint formula for the precipitated phase is:

[0130] ;

[0131] Where, f TiC represents the volume fraction of TiC precipitation.

[0132] Phase transformation kinetic constraints are based on the thermodynamic and kinetic laws of the transformation from austenite to ferrite, controlling the temperature and cooling rate during rolling to ensure that austenite decomposes into ferrite and pearlite within a specific temperature range. The core of this constraint is to regulate the phase transformation rate and final microstructure ratio through the synergistic effect of the temperature window and time parameters to avoid grain coarsening caused by austenite recrystallization. Phase transformation kinetic constraints force the final rolling temperature to be lower than the Ar3 phase transformation point, triggering deformation-induced ferrite transformation (DIFT) and directly generating ultrafine ferrite grains, replacing the traditional spheroidizing annealing process. By controlling the ferrite volume fraction and pearlite spheroidization rate, the cold heading steel is ensured to have uniform plasticity and strength, meeting the forming requirements of high-strength fasteners under annealing-free conditions.

[0133] It's important to note that phase transformation control is fundamental to hot rolling, particularly in controlled rolling and cooling techniques, where it guides the setting of final rolling temperature and cooling rate. In the meta-learning framework, phase transformation dynamics constraints are incorporated into the loss function. For example, if the model's predicted ferrite fraction deviates from the target, backpropagation is used to adjust the recommended rolling temperature to ensure that microstructure and performance meet the target.

[0134] Deformation resistance and rolling force constraints dynamically calculate mill load and energy consumption during rolling based on the flow stress patterns of the material during high-temperature deformation, combined with the billet composition and temperature distribution. Their core objective is to balance rolling efficiency with equipment load capacity, preventing the risk of roll wear or strip breakage caused by excessive deformation resistance. Deformation resistance and rolling force constraints are used to limit the rolling force range during roughing and finishing, preventing equipment downtime due to overload. By adjusting the reduction ratio and rolling speed, ineffective deformation energy consumption is reduced, improving rolling efficiency.

[0135] It's important to note that deformation resistance and rolling force control are industry-standard technologies. Rolling force models and automatic gauge control (AGC) systems are integrated into modern rolling mills to control process parameters in real time. The meta-learning framework uses an upper limit on rolling force as a hard boundary condition. If the predicted rolling force exceeds the equipment limit, the model automatically reduces the recommended reduction or adjusts the rolling temperature to ensure safe production.

[0136] Precipitation pinning utilizes the solid solution and precipitation behavior of microalloying elements (such as Ti and Nb) in austenite to pin grain boundaries through the dispersed distribution of carbonitrides (TiC and NbC), hindering grain growth. Its core approach is to control the size, quantity, and distribution of precipitates by matching composition design with process parameters. Precipitation pinning is used to pin precipitates to austenite grain boundaries during rolling, limiting grain growth and refining the final ferrite grains to ≤5μm. This grain refinement and uniform microstructure significantly reduces the cold forging cracking rate and improves the material's formability.

[0137] It should be noted that the precipitate pinning principle is primarily used in the production of high-strength and microalloyed steels. In the meta-learning framework, precipitate constraints are encoded as temperature range restrictions. For example, the model avoids temperature ranges where precipitates dissolve back, and when recommending rolling parameters, it prioritizes a process window that preserves precipitates without compromising ductility.

[0138] S214. Set input variables and output variables, build and train a parameter recommendation model based on meta-learning, integrate physical metallurgical constraints, and match the rolling parameters corresponding to different steel billets.

[0139] In the description of the present invention, setting input variables and output variables, building and training a parameter recommendation model based on meta-learning, integrating physical metallurgical constraints, and matching rolling parameters corresponding to different steel billets include:

[0140] S2141. Each original task corresponds to a type of steel, the production specifications of the cold heading steel wire rod and the content of each component in the steel billet are set as input variables, and the rolling parameters are set as output variables to construct a support set.

[0141] Specifically, in the meta-learning framework, each meta-task corresponds to a type of steel, and its support set consists of the historical production data of that type of steel.

[0142] The input variables (input layer) include the production specifications of the cold heading steel wire rod and the composition content of the steel billet (mass percentage of elements such as C, Mn, Cr, and Ti), and the output variables (output layer) are the rolling parameters of the corresponding steel grade (temperature, reduction rate, and rolling force in the rough / medium / finishing rolling stages).

[0143] For example, for SWRCH35K steel, the input is C = 0.35%, Mn = 0.75%, and a target diameter of Φ16mm. The output is a roughing temperature of 1030°C, a reduction of 45%, and a finishing temperature of 780°C, with a reduction of 65%. The support set must cover typical operating conditions for this steel grade (such as composition fluctuations and specification changes between heats). Each task typically contains 50-100 data sets, allowing the model to quickly adapt to new steel variants.

[0144] S2142. After the hidden layer output, the physical calculation module is linked to realize physical information embedding, which is used to calculate the temperature and reduction rate of the billet rolling at each stage.

[0145] Specifically, after the neural network's hidden layer output, a physical calculation module is linked to combine the data-driven prediction results with metallurgical equations to implement physical information constraints. For example, if the hidden layer output initially predicts a final rolling temperature of 790°C, but the Ar3 phase transition point formula calculates Ar3 = 750°C for this steel composition, the physical calculation module will correct the final rolling temperature to ≤750°C.

[0146] The hidden layer is the core component of a neural network, located between the input and output layers. It is responsible for performing nonlinear transformations and extracting features from the input data. Through multi-level weighted calculations and activation function mapping, the hidden layer transforms the raw input into high-dimensional abstract features, providing an intermediate representation for the final prediction.

[0147] In the cold heading steel rolling parameter recommendation model, the specific implementation of the hidden layer includes the following steps:

[0148] 1. Input reception: Receive standardized billet parameters.

[0149] 2. Weighted calculation: Through linear transformation of weight matrix and bias vector, the formula is simplified to:

[0150] Hidden layer output = activation function (weight × input + bias);

[0151] 3. Non-linear activation: Use ReLU function or Sigmoid function to introduce non-linearity, for example:

[0152] ReLU(x)=max(0,x);

[0153] Through nonlinear activation, the hidden layer can learn complex process laws.

[0154] 4. Feature transfer: The processed features are transferred to the output layer to generate preliminary prediction values.

[0155] It should be noted that the physical calculation module includes a phase change dynamics module, a rolling force calculation module and a precipitation phase control module. The specific contents are shown below.

[0156] 1. Phase transformation kinetics module: Input the predicted temperature and time, and calculate the ferrite volume fraction using the Avrami equation. If the predicted value is lower than the target, the rolling temperature and reduction rate are adjusted in the opposite direction.

[0157] This module predicts the austenite to ferrite phase transformation ratio based on the steel grade composition and rolling temperature-time history. The goal is to control the final rolling temperature and cooling rate to ensure that the ferrite ratio in the rolled material reaches the target value, thereby meeting the plasticity and uniformity required for cold heading. This is achieved through:

[0158] 1.1. Input parameters: predicted temperature output by the hidden layer of the neural network and rolling time at each stage;

[0159] 1.2. Dynamic correction: If the module calculation finds that the predicted temperature is higher than the Ar3 phase transformation point (critical temperature for the transformation of austenite to ferrite) of the steel grade, the finishing temperature will be forced to be lowered to below the Ar3 point;

[0160] 1.3. Reverse control: If the ferrite ratio is lower than the target value, the reduction rate in the finishing rolling stage is automatically increased to promote deformation-induced phase transformation.

[0161] 2. Rolling force calculation module: Based on the Sellars deformation resistance model, the predicted reduction rate and temperature are substituted to calculate whether the rolling force exceeds the limit. If it exceeds the limit, the reduction rate is reduced by 3%-5%.

[0162] This module calculates the rolling force per pass based on the Sellars deformation resistance model, taking into account roll dimensions and material high-temperature strength characteristics. The goal is to prevent mill overload or strip breakage caused by excessive reduction, while balancing production efficiency and equipment safety. This is achieved through:

[0163] 2.1. Input parameters: predicted reduction rate and rolling temperature output by the hidden layer;

[0164] 2.2. Dynamic verification: The module calculates the actual rolling force. If it exceeds the equipment safety threshold, it immediately triggers the reduction rate compensation;

[0165] 2.3. Temperature coordination: If the reduction rate cannot be reduced, increase the rolling temperature to reduce the material's deformation resistance.

[0166] 3. Precipitation phase control module: For Ti-containing steel grades, the volume fraction formula of TiC precipitation phase is used to check whether the rolling reduction rate meets the grain refinement requirements. If not, the reduction rate incremental compensation is triggered.

[0167] This module is designed for microalloyed steels containing titanium (Ti) and niobium (Nb). It calculates the pinning strength of precipitates on austenite grain boundaries through the Zener pinning effect. The goal is to ensure that the combination of rolling temperature and reduction rate promotes the dispersion of precipitates while preventing their re-dissolution or excessive coarsening. This is achieved by:

[0168] 3.1. Input parameters: steel grade composition, intermediate rolling temperature;

[0169] 3.2. Dynamic control: If the module detects that the current temperature is higher than the TiC dissolution temperature, the intermediate rolling temperature is reduced to 980°C to retain the precipitated phase;

[0170] 3.3. Reduction rate compensation: If the amount of precipitated phase is insufficient, the reduction rate in the intermediate rolling stage is increased to supplement the precipitated phase through the strain-induced precipitation mechanism.

[0171] S2143. Set two physical loss term functions, namely phase change loss and rolling force loss, and set a comprehensive loss function by combining phase change loss and rolling force loss to construct a parameter recommendation model.

[0172] Specifically, when training the parameter recommendation model, in addition to the traditional data loss (the mean square error between the predicted parameters and the measured values), a physical loss term needs to be added to force the model output to conform to the laws of metallurgy.

[0173] Among them, the phase transformation loss function is used to ensure the core constraints of material performance. During the hot rolling process, the austenite of cold heading steel needs to be transformed into ferrite / pearlite structure through phase transformation, and its ratio directly affects the cold heading properties of the material (such as plasticity and crack resistance). The phase transformation loss function quantifies the deviation between the phase transformation results predicted by the model (such as ferrite ratio) and the target value, and forces the parameters such as rolling temperature and reduction rate recommended by the model to meet the phase transformation dynamics law. For example, if the model initially recommends a final rolling temperature of 800℃ in pursuit of production efficiency, but the Ar3 phase transformation point (critical temperature for the transformation of austenite to ferrite) calculated based on the steel composition is 750℃, the phase transformation loss function will determine that the final rolling temperature is too high (austenite is not fully transformed) and force the model to correct the temperature to below 750℃ through back propagation to ensure that the ferrite ratio is ≥76%.

[0174] The rolling force loss function calculates the theoretical rolling force corresponding to parameters such as the reduction rate and rolling speed recommended by the model, compares it with the rated load capacity of the rolling mill, and imposes penalties for exceeding the limit parameters. Its core purpose is to prevent accidents such as strip breakage and roll fracture caused by equipment overload. For example, if the parameter recommendation model is for grain refinement and recommends a 20% reduction rate in the intermediate rolling stage, but the material's high-temperature deformation resistance model calculates that this reduction rate corresponds to a rolling force of 2800 MPa, exceeding the equipment's safety threshold (2600 MPa). At this point, the rolling force loss function will increase significantly, forcing the model to reduce the reduction rate to below 18% or increase the rolling temperature to reduce the material's resistance.

[0175] The comprehensive loss function combines phase transformation loss and rolling force loss according to assigned weights to form a single optimization objective. During model training, a gradient descent algorithm automatically searches for the parameter combination that minimizes the comprehensive loss, achieving a balance between performance and safety. For example, when producing titanium-containing microalloyed steel, the model may face a dilemma: on the one hand, increasing the temperature promotes the dissolution of TiC precipitates, reducing rolling force;

[0176] On the other hand, lowering the temperature can retain TiC particles, refine the grain size, but increase rolling force. In this case, the comprehensive loss function will select a compromise solution of 960°C and 17% reduction based on the weight distribution. This solution retains some TiC grain refinement while keeping the rolling force within safe limits.

[0177] Specifically, the essence of constructing a comprehensive loss function is to integrate material performance requirements and production safety constraints into a single optimization goal through mathematical means, thereby generating process parameters that conform to both metallurgical laws and equipment carrying capacity. The weight distribution of phase change loss and rolling force loss needs to be dynamically adjusted in combination with actual production - if the current production line is more concerned with the cold heading pass rate, then the phase change loss will have a higher weight in the comprehensive loss, so that the model prioritizes ensuring that the ferrite ratio meets the standard; if the equipment is under high load, the weight of the rolling force loss will be increased to ensure that the parameter recommendation does not trigger the equipment protection mechanism. In the actual training process, the parameter space is continuously explored through gradient descent, and the process parameter combination that can minimize the comprehensive loss is automatically found. Through the coupling of physical rules and data laws, the optimal balance point is found in the contradiction between phase change control and rolling force safety, and finally a parameter solution is generated that can meet the ferrite ratio requirements and control the rolling force within the safety threshold.

[0178] S2144. Pre-train model parameters on all steel grade tasks to initially form a hot rolling knowledge base, and when facing new steel grades, fine-tune the calculation tasks through single-step gradient descent.

[0179] Specifically, S2144 includes the following aspects:

[0180] 1. Global Pre-training: The model is pre-trained on a historical database to learn common patterns across steel grades, such as the need for lower finish rolling temperatures for high-carbon steel and higher intermediate rolling reductions for Ti-containing steels. This pre-training creates a hot-rolling knowledge base, allowing the model to generalize to new steel grades with similar compositions.

[0181] 2. Single-step gradient fine-tuning: For new steel grades, only 50 sets of support data are required to update the model parameters through single-step gradient descent. For example, for 10B21 containing boron (B = 0.002%), the model quickly adjusts based on pre-trained knowledge to generate parameters for a finishing temperature of 760°C and a reduction ratio of 68%, meeting the requirements for improved hardenability.

[0182] S215. Input the size and component content of the steel billet to be processed into the parameter recommendation model, generate rolling parameters for each stage and adjust the logic in real time to obtain the optimal rolling parameters.

[0183] In the description of the present invention, the size and component content of the steel billet to be processed are input into the parameter recommendation model, the rolling parameters of each stage are generated and the logic is adjusted in real time to obtain the optimal rolling parameters including:

[0184] S2151. Obtain the carbon content of the steel billet to be processed, match the rough rolling temperature corresponding to the rough rolling stage according to the carbon content value, and calculate the rough rolling reduction rate.

[0185] The calculation formula for rough rolling temperature is:

[0186] ;

[0187] Where, T 粗轧 Indicates the rough rolling temperature in the rough rolling stage; T0 indicates the preset initial temperature; [C] indicates the carbon content;

[0188] The calculation formula for the rough rolling reduction rate is:

[0189] ;

[0190] Where, r 粗轧 Indicates the rough rolling reduction rate in the rough rolling stage; r 1 represents the initial compression ratio of the rough rolling stage output by the parameter mapping table; R total Indicates the target compression ratio.

[0191] It should be noted that the preset initial temperature T 0 and initial compression ratio r 1 are all obtained from the output of the parameter mapping table established based on the historical database, that is, when a new steel billet is put into production, its composition and size are automatically matched to the closest data category, and the initial compression ratio and initial temperature of the rough rolling, intermediate rolling and finishing rolling stages are automatically calculated.

[0192] Specifically, the preset initial temperature T The determination of 0 is based on the comprehensive setting of steel grade characteristics, equipment capabilities and process objectives: For low carbon steel, T 0 is usually set to 1150-1200℃ to promote dynamic recrystallization; the recrystallization temperature of medium and high carbon steels will decrease due to the increase of carbon content. T 0 needs to be lowered to 1100-1150℃ to avoid grain coarsening; at the same time, it is finally determined through historical data statistics or experimental verification in combination with the maximum load of the rolling mill, the efficiency of the heating furnace and the final performance requirements of the product.

[0193] The roughing temperature calculation formula dynamically adjusts the rolling temperature based on carbon content to precisely match process parameters with material properties. This formula inversely adjusts the heating temperature during the roughing stage based on the carbon content of the steel: the higher the carbon content, the lower the roughing temperature setting. This suppresses excessive austenite grain growth at high temperatures, refining the steel's original grain structure and improving microstructure uniformity during subsequent deformation. Furthermore, lowering the rolling temperature of high-carbon steel effectively controls surface oxide scale formation and reduces material loss due to oxidation. Furthermore, the negative correlation between temperature and carbon content balances rolling force with equipment load capacity, preventing insufficient recrystallization of low-carbon steel due to insufficient temperature and excessive rolling force at high temperatures for high-carbon steel. The practical significance of the roughing temperature calculation formula lies in converting compositional differences into controllable process parameters, enabling flexible production of different steel grades on the same production line, significantly improving yield and product quality consistency.

[0194] The calculation formula of rough rolling reduction ratio can realize dynamic adaptation of target compression ratio: if the target compression ratio of a batch is R total =85%, r 1 is taken from the parameter mapping table as 30% (the optimal value for similar steel grades in historical data), and the calculated reduction ratio is 30% + 0.1 × (85 - 80) = 30.5%. This design ensures that the actual reduction ratio approaches the target value (with an error of ≤ ±1.5%), avoiding excessively high reduction ratios in a single pass that could overload the mill (a reduction exceeding 35% could trigger an alarm), or excessively low reduction ratios that could increase the number of rolling passes (reducing one or two passes). Combined with temperature control, this refines the initial austenite grains and reduces the risk of cold heading cracking.

[0195] S2152. Obtain the titanium content and manganese content of the steel billet to be processed and the grain size after rough rolling, and calculate the intermediate rolling cooling rate and intermediate rolling reduction rate of the steel billet when it enters the intermediate rolling stage.

[0196] The calculation formula for the intermediate rolling cooling rate is:

[0197] ;

[0198] Where,v T Indicates the cooling rate of intermediate rolling; Δ T Indicates the difference between rough rolling temperature and medium rolling temperature; Δ t Indicates the time difference; [ Mn ] indicates the manganese content.

[0199] The calculation formula for the intermediate rolling reduction rate is:

[0200] ;

[0201] Where, r 中轧 Indicates the intermediate rolling reduction rate; r 2 represents the initial compression ratio of the intermediate rolling stage output by the parameter mapping table; [ Ti ] indicates the titanium content.

[0202] It should be noted that the intermediate rolling cooling rate is dynamically adjusted based on the manganese content: for every 0.1% increase in manganese content, the cooling rate increases by 0.05°C / s. This ensures that the intermediate rolling temperature remains stable within the 950-1000°C range, avoiding austenite grain coarsening caused by insufficient cooling or sudden changes in rolling force caused by excessive cooling. This control reduces the proportion of banded structure to ≤8%, improving the uniformity of cold-headed steel.

[0203] The intermediate rolling reduction ratio is calculated based on the titanium content-corrected reduction ratio: for every 0.01% increase in titanium content, the reduction ratio increases by 1%. The precipitation strengthening effect of titanium compensates for the impact of increased reduction on the mill load, while also refining the ferrite grains, increasing the tensile strength after intermediate rolling by 20-30 MPa and reducing the incidence of surface cracks.

[0204] S2153. Calculate the Ar3 phase transformation point of the steel billet to be processed, set the finishing temperature in the finishing rolling stage according to the Ar3 phase transformation point, and calculate the finishing reduction rate in the finishing rolling stage according to the carbon content.

[0205] The calculation formula for finishing rolling reduction is:

[0206] ;

[0207] Where r 精轧 represents the finishing reduction rate; r3 represents the initial compression ratio of the finishing stage output by the parameter mapping table.

[0208] It should be noted that the finishing reduction ratio is calculated by inversely adjusting the compression amount based on the carbon content: when the carbon content drops from 0.35% to 0.25%, the compression ratio increases by 2.9%. The finishing temperature set at the Ar3 phase transformation point can not only suppress the brittleness caused by excessive carbon content, but also optimize cold heading formability through deformation-induced ferrite phase transformation, effectively improving dimensional accuracy control.

[0209] S22. Real-time monitoring of hot rolling data during the hot rolling process, dynamic compensation of optimal rolling parameters, and dynamic adjustment of the pass interval time according to the monitoring and compensation results.

[0210] In the description of the present invention, real-time monitoring of hot rolling data during hot rolling, dynamic compensation of optimal rolling parameters, and dynamic adjustment of pass intervals according to the monitoring and compensation results include:

[0211] S221. Use sensors inside the rolling mill to collect hot rolling data of the billet rolling process in real time.

[0212] Specifically, high-precision sensors are deployed at key locations of the rolling mill's roll bearings, guides, and roller tables to collect the following data in real time:

[0213] 1. Temperature data: The billet surface temperature is monitored using an infrared thermal imager, with data collected every 0.5 seconds to ensure that the temperature gradient is controlled within ±20°C. For example, during the rough rolling stage, the head temperature of a Φ16mm wire rod is 1050°C and the tail temperature is 1030°C, with a temperature difference of 20°C, which is considered normal fluctuation.

[0214] 2. Rolling force data: A rolling force sensor (range 0-3000 MPa, error ≤ ±1%) records the mill load at each pass in real time, transmitting the data to the control center via industrial Ethernet. For example, the measured rolling force during the finishing stage is 2450 MPa, a 2% deviation from the target value of 2500 MPa, triggering dynamic compensation.

[0215] 3. Dimension data: Laser diameter measuring instrument detects the diameter and ovality of rolled pieces and provides immediate feedback when out of tolerance.

[0216] S222. Calculate the difference between the actual rolling force and the target rolling force of the rolling mill group during each stage of rolling, and calculate the reduction rate compensation amount through the proportional-integral coefficient.

[0217] Specifically, the formula for calculating the reduction rate compensation using the proportional-integral coefficient is:

[0218] ;

[0219] Where Δr represents the compensation amount; k p 、k i Respectively represent the proportional and integral coefficients; F 目标 Indicates target suppression force; F 实际 Indicates the actual suppressive force.

[0220] In the production process of cold heading steel hot rolled wire rod, the determination of the proportional-integral (PI) coefficient is closely related to the physical properties of the rolling mill and the deformation behavior of the material. Its essence is to convert the dynamic response characteristics of the rolling mill system into executable control parameters through mathematical means.k p and the integral coefficient k i The value is calibrated based on the mechanical properties of the rolling mill itself, the inertia characteristics of the drive system and the high-temperature rheological law of the material.

[0221] Integration coefficient k i The setting of is related to the long-term fluctuation characteristics of the rolling force, such as the progressive drift of the rolling force caused by roll wear, which requires a higher k i The proportional-integral coefficient is usually obtained by combining offline experiments with online self-tuning: During the equipment commissioning phase, engineers will apply a step-down command to the rolling mill and calculate the initial k p and k i In actual production, the coefficient is dynamically fine-tuned according to real-time data (such as rolling force fluctuation frequency and error convergence speed).

[0222] Proportional coefficient k p Determines the system's immediate response intensity to the current deviation. When the measured rolling force deviates from the target value, k p Adjust the reduction rate compensation directly according to the deviation ratio. For example, if the rolling force deviation is large, the higher k p It will quickly increase the compensation amount and shorten the adjustment time. k i Used to eliminate the cumulative effects of historical deviations, when a deviation persists for a long time (e.g., due to persistently low rolling force caused by roll wear), the integral term continuously adjusts the compensation by accumulating historical deviations until the steady-state error is eliminated. For example, if the rolling force is 5% below the target for three consecutive seconds, the integral term will gradually increase the compensation amount until the deviation is zero.

[0223] S223. Calculate the temperature difference between the beginning and the end of the billet at each stage. If the temperature difference between the beginning and the end is greater than a preset threshold, the pass interval extension adjustment is triggered to balance the temperature drop and the rolling rhythm.

[0224] Specifically, after each pass, the temperature difference between the head and tail of the billet is calculated, and the interval is extended by 0.5 seconds for every 1°C difference. For example, if the measured temperature difference is 25°C (5°C over the limit), the interval is extended by 2.5 seconds. The reduction rate is reduced by 0.2% for every 1°C difference. For example, if the temperature difference is 25°C, the reduction rate is reduced by 1%.

[0225] S224. Import the real-time hot rolling data into the historical parameter library, set the update cycle, and regularly update the comprehensive loss function. If there is a sudden change in the hot rolling data, an emergency stop signal is triggered.

[0226] Specifically, real-time data (temperature, rolling force, and dimensions) are stored by steel type, triggering a model parameter update every 30 minutes.

[0227] Mutation detection and emergency response include the following aspects:

[0228] 1. Sudden change in rolling force: If the rolling force variation between adjacent passes is greater than 20%, it is determined to be a belt break risk and triggers an emergency stop signal;

[0229] 2. Temperature drop: If the temperature drop rate in a certain area is greater than 10℃ / s (normal range 1-3℃ / s), the heater will be started to compensate for the temperature drop;

[0230] 3. Model rollback: If the error is still exceeded after three consecutive compensations, the model will automatically switch to the historical optimal parameter template.

[0231] S23. After rolling is completed, a nano-Al2O3-MgO composite coating is sprayed on the surface of the cold heading steel wire rod, and a dense oxide film is generated using the waste heat. The actual production specifications of the cold heading steel wire rod are then verified.

[0232] Specifically, after cold-heading steel wire rod is rolled, a nano-Al2O3-MgO composite powder mixed with an ethanol suspension is sprayed onto the wire rod surface using high-pressure spraying equipment. Utilizing the residual heat of the wire rod (no additional heating required), the powder particles melt and diffuse at high temperatures, forming a dense and uniform composite coating. During this process, the Al2O3 provides high hardness and wear resistance, while the MgO strengthens the coating's bond to the steel substrate. It also induces oxidation of the surface Fe element to form an Fe3O4 oxide film, inhibiting the formation of FeO red rust.

[0233] S3. The hot-rolled cold-heading steel wire rod is transported to a heat-insulating tunnel furnace, and the cold-heading steel wire rod is obtained through segmented controlled cooling and surface treatment, thereby realizing the production of annealing-free cold-heading steel hot-rolled wire rod.

[0234] In the description of the present invention, the hot-rolled cold-heading steel wire rod is transported to an insulated tunnel furnace, and the cold-heading steel wire rod is obtained through segmented controlled cooling and surface treatment. The production of annealing-free cold-heading steel hot-rolled wire rod includes:

[0235] S31. Obtain and record the parameters of the cold heading steel wire rod after hot rolling, the wire rod parameters including the final rolling temperature, the grain size after rolling, the predicted ferrite ratio, and the rolling force curve.

[0236] S32. The cooling process of cold heading steel wire rod is divided into three stages: rapid cooling, slow cooling and final cooling. A physical constraint meta-learning framework is built to generate cooling parameters to control the cooling of cold heading steel wire rod in segments.

[0237] In the present invention, the cooling process of cold-heading steel wire rod is divided into three stages: rapid cooling, slow cooling, and final cooling. A physical constraint meta-learning framework is constructed to generate cooling parameters for segmented control of cold-heading steel wire rod cooling, including:

[0238] S321. Introduce wire rod parameters to expand the input layer, embed phase change dynamics equations and heat conduction equations to build a physical constraint layer, and build an output layer that outputs segmented cooling rate values and final cooling temperatures.

[0239] Specifically, in the input layer of the neural network, in addition to the production specifications of the cold heading steel wire rod and the composition of the steel billet, the real-time temperature field distribution and phase transformation progress (ferrite ratio) of the wire rod are introduced as extended input variables.

[0240] The physical constraint layer embeds the phase transformation kinetics equation (Avrami equation) and the heat conduction equation. For example, the ferrite volume fraction is calculated using the Avrami equation and compared with the target value (≥76%). If the predicted value deviates, the network weights are adjusted in the opposite direction. The output layer is divided into three channels: the cooling rate of the rapid cooling phase (3-5°C / s), the cooling time of the slow cooling phase (300-500 seconds), and the final cooling temperature (500-600°C).

[0241] It should be noted that the cooling process of cold heading steel wire rods shares the same core logic as the physical constraint meta-learning framework established during the hot rolling stage. Both achieve intelligent optimization of process parameters by integrating metallurgical principles with data-driven models. During the hot rolling stage, the physical constraint meta-learning framework dynamically adjusts the rolling temperature and reduction rate by embedding rules such as the rolling force model and the phase transformation kinetic equation. During the cooling stage, the framework is extended to the heat conduction equation to control the cooling rate and final cooling temperature. The underlying data closed-loop mechanism of the two (real-time monitoring → dynamic compensation → model update) is exactly the same, and both are centered on "physical rule constraints + meta-learning generalization" to ensure that parameter recommendations are both scientifically sound and adaptable to multiple working conditions.

[0242] In a physically constrained meta-learning framework for the cooling phase of cold-heading steel wire rod, the temperature field evolution and microstructural evolution mechanisms during the cooling process are transformed into mathematical constraints through the dual embedding of heat conduction equations and phase transformation kinetics equations, driving a neural network model to generate cooling parameters that conform to metallurgical laws. Specifically, the physically constrained meta-learning framework first takes the wire rod's real-time temperature field distribution, steel grade composition, and target microstructure properties as input variables. The neural network's hidden layer extracts features and preliminarily predicts cooling parameters for the rapid, slow, and final cooling stages. Subsequently, in the physical constraint layer, the Fourier heat conduction equation is used to calculate the theoretical relationship between temperature gradient and cooling rate. The Avrami equation is then used to infer the ferrite formation progress at the current cooling rate. If the predicted parameters result in a theoretical ferrite ratio lower than the target value or the temperature field distribution violates heat conduction laws (e.g., excessive surface cooling leading to excessive core thermal stress), the neural network's weight parameters are forcibly modified through a backpropagation mechanism to ensure that the output cooling rate, cooling time, and final cooling temperature conform to physical laws. For example, when a cooling rate of 5°C / s was initially recommended for the rapid cooling stage, the physical constraint layer detected that the core heat could not be discharged in time at this rate (the temperature drop rate exceeded the allowable range of the Fourier equation), and immediately triggered parameter correction, reducing the rapid cooling rate to 4.2°C / s and extending the slow cooling time to 380 seconds, thereby balancing tissue performance and thermal stress risks.

[0243] The physical constraint meta-learning framework for the cooling stage adopts a channel-by-channel output structure. The cooling rate in the rapid cooling stage is achieved by the coordinated control of the fan speed and the water mist flow rate. The slow cooling time is dynamically adjusted by the roller speed. The final cooling temperature is precisely maintained by the closed-loop temperature control system. The three are linked through real-time verification of the heat conduction equation.

[0244] S322. Divide the cooling process of the cold heading steel wire rod into rapid cooling, slow cooling and final cooling stages, and calculate the cooling rate and cooling time of the rapid cooling stage respectively.

[0245] In the description of the present invention, calculating the cooling rate and cooling time of the rapid cooling stage includes:

[0246] The calculation formula of the rapid cooling rate in the rapid cooling stage is:

[0247] ;

[0248] Where, v 速冷 Indicates the rapid cooling rate in the rapid cooling stage; T 终轧 Indicates the final rolling temperature; T 速冷终 Indicates the target temperature in the rapid cooling stage; t 速冷 Indicates rapid cooling time;

[0249] The calculation formula for cooling time in the slow cooling stage is:

[0250] ;

[0251] Where, t 缓冷 represents the slow cooling time; k0 represents the phase change rate constant; represents the phase change activation energy; R represents the gas constant; T 缓冷 Indicates slow cooling temperature; X total represents the target ferrite ratio; n represents the Avrami index; and e represents the natural constant.

[0252] S323. Based on the preset segmented air cooling strategy, a fan-water mist combination cooling is adopted in the rapid cooling stage, the fan is turned off in the slow cooling stage, and natural cooling is carried out by roller conveyor. In the final cooling stage, natural cooling is adopted and a nitrogen-carbon dioxide mixture is introduced to form a dense oxide film.

[0253] Specifically, the rapid cooling stage utilizes a combination of fans (80%-100% power) and water mist (flow rate 15-20L / min). The fan speed is adjusted based on the real-time temperature drop rate. For example, if the temperature drop rate in a certain section is detected to be less than 3°C / s, the fan speed is increased by 5%. During the slow cooling stage, the fan is turned off, and the roller speed is reduced to 0.1-0.2m / min. Natural cooling (at a rate of 0.5-1°C / s) is used to promote ferrite growth and pearlite spheroidization. During the final cooling stage, a nitrogen-carbon dioxide mixture (5%-8% CO2) is introduced to form a dense Fe3O4 oxide film through gas reaction. Simultaneously, the roller speed is restored to 0.5m / min to ensure the uniformity of the oxide film.

[0254] S33. Dephosphorize the cooled cold heading steel wire rod using high pressure water, and identify surface defects through online laser detection. After the detection is correct, obtain the finished cold heading steel wire rod.

[0255] Specifically, high-pressure water at 15-25 MPa is used to impact the surface of the cooled wire rod, removing the oxide scale through the water hammer effect. The water pressure is achieved by a multi-stage booster pump, and the nozzle adopts a fan-shaped design (covering an angle of 30-45 degrees) to ensure that the water flow evenly covers the surface of the wire rod.

[0256] A laser 3D scanner is used to scan at high speed along the axial direction of the wire rod to generate a three-dimensional surface topography map in real time, and defects such as scratches, folds, and pits are identified through AI algorithms (convolutional neural networks).

[0257] Detection logic:

[0258] 1. Defect classification: scratches, folds, and oxide scale residues;

[0259] 2. Dynamic sorting: When a defect is detected, the pneumatic marker is triggered to mark the position, and the sorting mechanism is controlled by PLC to guide the defective section into the rework line.

[0260] The present invention is described in detail below with reference to specific embodiments.

[0261] Example

[0262] Cold heading steel material selection: SCM435, including the following mass percentage components (unit %):

[0263] C: 0.355; S: 0.25; Mn: 0.75; S: 0.02; P: 0.01; Cr: 1.05; Ni: 0.01; Cu: 0.15; Mo: 0.225.

[0264] In addition, the present invention is applicable to medium and low carbon cold heading steels (C=0.10-0.45%) and microalloyed cold heading steels (containing B / Ti / Nb≤0.05%), including SWRCH series (such as SWRCH35K), ML series (such as ML08Al), 10B21, etc., covering more than 90% of cold heading steel grades in JIS, ASTM and GB standards.

[0265] The following is an explanation of the operating steps of the production process.

[0266] Step 1: Grinding and preheating

[0267] 1.1. Grinding parameters: Use 18-grit grinding wheel to remove corner wrinkles in the first round (grinding depth 0.3mm), and switch to 25-grit grinding wheel for polishing in the second round (Ra≤6.3μm).

[0268] 1.2. Preheating parameters: Initial preheating is 600℃ (to prevent thermal stress cracks). After heating to 1120℃, a CO2 / N2 (volume ratio 3:7) mixed gas is introduced to increase the surface hardness to 250HV (originally 200HV).

[0269] Step 2: Multi-stage temperature controlled rolling

[0270] 2.1 Rough rolling stage: temperature 1100℃→1000℃ (cooling rate 2℃ / s), compression ratio 6:1 (Φ200mm→Φ80mm), single-pass reduction rate 18%.

[0271] 2.2 Intermediate rolling stage: temperature 950℃→900℃ (cooling rate 1.5℃ / s), compression ratio 4:1 (Φ80mm→Φ38mm), reduction rate 15%.

[0272] 2.3 Finishing rolling stage: temperature 880℃→850℃ (cooling rate 1℃ / s), compression ratio 2:1 (Φ38mm→Φ18mm), reduction rate 12%.

[0273] Step 3: Segmented cooling

[0274] 3.1 Rapid cooling stage: fan + water mist cooling (rate 4℃ / s), final cooling to 700℃, austenite grain size ≤20μm (traditional process ≥35μm).

[0275] 3.2 Slow cooling stage: natural cooling for 300 seconds, ferrite ratio 76% (Avrami equation prediction error ±2%).

[0276] 3.3 Final Cooling Stage: A CO2 mixture (5% concentration) was introduced, achieving an oxide film thickness of 18μm (compared to ≥30μm in conventional processes). The average tensile strength and reduction of area (after clean shearing) of the SCM435 steel using this production process significantly decreased by 209MPa (from 863MPa to 654MPa), while the reduction of area increased by 17% (from 33% to 50%). The martensite content in the microstructure was also completely eliminated.

[0277] The properties of the cold heading steel wire rods obtained before and after the production process optimization of SCM435 are shown in Table 1. The data of the cold heading steel wire rods were measured in accordance with the national standard GB / T 28906-2012.

[0278] Table 1: Performance of SCM435 before and after production process optimization

[0279]

[0280] Typical tissue characterization maps before optimization Figure 2 As shown, the typical organization after optimization is as follows Figure 3 shown.

[0281] From the changes in the mechanical properties of SCM435 under different shearing cycles, it can be seen that the average mechanical properties and area shrinkage of SCM435 produced before optimization are relatively small under different shearing cycles, with the fluctuation of tensile strength being only 32MPa and the fluctuation of cross-sectional shrinkage being only 5%. Figure 4 As shown, the mechanical properties of 22MM SCM435 after optimization with different shear cycles are as follows Figure 5 shown.

[0282] In addition, the metallographic conditions of SCM435 under different cooling conditions are as follows. As shown in Table 2 (measured according to the national standard GB / T 28906-2012), the average hardness and grain size after optimization are 10HRB lower than before optimization, the iron oxide scale is 5 microns thicker than that of the air-cooled line, and the thickness of the decarburized layer is consistent.

[0283] Table 2: Metallographic structure of SCM435 under different cooling conditions

[0284]

[0285] After passing through the insulation channel, the average tensile strength and cross-sectional shrinkage rate of SCM435 (after shearing the head and tail) decreased significantly compared to the existing ones, decreasing by 209MPa (from 863MPa to 654MPa), and the surface shrinkage increased by 17% (from 33% to 50%). The martensite in the organization was also completely eliminated, and the hardness decreased by 10HRB (from 99HRB to 89HRB).

[0286] In another embodiment, B7 steel is used for cold steel wire rod processing, and its performance parameters before and after optimization are shown in Table 3, wherein the composition of B7 steel includes the following components in mass percentage: C: 0.395; S: 0.225; Mn: 0.85; S: 0.011; P: 0.012; Cr: 1.05; Mo: 0.25, and the performance data is measured in accordance with the national standard GB / T 28906-2012.

[0287] Table 3: B7 steel performance before and after optimization

[0288]

[0289] The average tensile strength of B7 after optimization (after the head and tail are sheared cleanly) is significantly lower than that before optimization. The tensile strength is reduced by about 160MPa (from 859MPa to 699MPa), the surface shrinkage is increased by 19% (from 29% to 48%), the martensite in the structure is completely eliminated, and the hardness is reduced by 7HRB (from 100HRB to 93HRB). In addition, the typical structure of the cold heading wire before optimization is as follows: Figure 6 As shown, the typical organization of the cold heading line after optimization is as follows Figure 7 shown.

[0290] To sum up, with the help of the above-mentioned technical scheme of the present invention, the present invention produces cold heading steel wire rods of different specifications by adopting a large compression ratio advanced rolling mill group, and controls the cooling temperature and cooling rate of the wire rods with the help of an insulated tunnel furnace, thereby realizing the production of annealing-free cold heading steel hot-rolled wire rods, greatly simplifying the production process, shortening the production cycle, reducing production costs, and improving production efficiency; the produced cold heading steel wire rods do not need to undergo annealing treatment, avoiding the changes in organizational structure and performance that may be brought about by annealing treatment, thereby improving the quality stability of the product; at the same time, the present invention has no strict requirements on the composition and organizational structure of the steel billet, so it can be applied to more steel grades, expanding its application range; by precisely controlling the cooling temperature and cooling rate, the strength and toughness of the product are significantly improved, and it has better performance.

[0291] Through multi-pass temperature gradient rolling, the temperature changes in each stage are precisely controlled, and the segmented controlled cooling process is combined to achieve the directional phase transformation from austenite to ferrite and the spheroidization of pearlite, breaking through the limitations of traditional processes; in the rolling process, through the synergistic effect of differentiated temperature control and large reduction rate in the three stages of rough, medium and finish rolling, ultrafine grains and uniform structure are directly obtained; in the cooling process, rapid cooling is used to inhibit carbide coarsening, slow cooling is used to promote phase transformation balance, and the final cooling oxide film control technology is used to completely eliminate the need for annealing; ultimately, the integrated performance of high strength, high plasticity and corrosion resistance of cold heading steel wire rod is improved, while significantly reducing energy consumption and production costs, and promoting the upgrading of the steel hot rolling industry towards green and high efficiency.

[0292] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

Claims

1. A method for producing annealing-free cold heading steel hot-rolled wire rod, characterized in that: include: S1. Select steel billets that meet the production specifications of cold heading steel wire rods, and perform grinding, finishing and preheating on the steel billets to be processed; S2. The steel billet is transported to a high-compression ratio rolling mill, and the hot rolling process is monitored in real time to perform segmented hot rolling control, thereby achieving multi-pass temperature gradient rolling to obtain hot-rolled cold-headed steel wire rod; S3, conveying the hot-rolled cold-heading steel wire rod to a heat-insulating tunnel furnace, and obtaining a finished cold-heading steel wire rod through segmented controlled cooling and surface treatment, thereby realizing the production of annealing-free cold-heading steel hot-rolled wire rod; The process of conveying the steel billet to a high-compression ratio rolling mill, monitoring the hot rolling process in real time, performing segmented hot rolling control, and implementing multi-pass temperature gradient rolling to obtain a hot-rolled cold-heading steel wire rod comprises: S21. Based on the production specifications of the cold heading steel wire rod and the composition of the steel billet, the rolling process is divided into three stages: rough rolling, intermediate rolling, and finishing rolling. The physical constraint meta-learning algorithm is used to match the optimal rolling parameters. S22, real-time monitoring of hot rolling data during the hot rolling process, dynamic compensation of optimal rolling parameters, and dynamic adjustment of the pass interval time according to the monitoring and compensation results; S23. After rolling is completed, a nano-Al2O3-MgO composite coating is sprayed on the surface of the cold heading steel wire rod to generate a dense oxide film using waste heat, and the actual production specifications of the cold heading steel wire rod are verified; According to the production specifications of the cold heading steel wire rod and the composition content of the steel billet, the rolling process is divided into three stages: rough rolling, intermediate rolling, and finishing rolling, and the physical constraint meta-learning algorithm is used to match the optimal rolling parameters, including: S211. Based on the preset production specifications of the cold heading steel wire rod, reversely deduce the cross-sectional dimensions of the steel billet and calculate the target compression ratio of the steel billet to be processed by hot rolling to the cold heading steel wire rod; S212, dividing the rolling process of the steel billet into three stages: rough rolling, intermediate rolling, and finishing rolling; establishing a parameter mapping table using a historical database; and outputting the initial compression ratio and rolling temperature range of each stage; S213. Establishing physical constraints for hot rolling of steel billets based on metallurgical theory, wherein the physical constraints include phase transformation dynamics constraints, deformation resistance and rolling force constraints, and precipitation phase pinning constraints; S214. Set input variables and output variables, build and train a parameter recommendation model based on meta-learning, integrate physical metallurgical constraints, and match rolling parameters corresponding to different steel billets; S215, inputting the size and component content of the steel billet to be processed into the parameter recommendation model, generating rolling parameters for each stage and adjusting the logic in real time to obtain the optimal rolling parameters; The size and component content of the steel billet to be processed are input into the parameter recommendation model to generate rolling parameters for each stage and adjust the logic in real time to obtain the optimal rolling parameters, including: S2151. Obtain the carbon content of the steel billet to be processed, match the rough rolling temperature corresponding to the rough rolling stage according to the carbon content value, and calculate the rough rolling reduction rate; The calculation formula of the rough rolling temperature is: T 粗轧 =T0-50·[C]; Where, T 粗轧 Indicates the rough rolling temperature in the rough rolling stage; T0 indicates the preset initial temperature; [C] indicates the carbon content; The calculation formula of the rough rolling reduction rate is: r 粗轧 ＝r1+0.1·(R total -80％)； Where r 粗轧 represents the rough rolling reduction rate in the rough rolling stage; r1 represents the initial compression ratio in the rough rolling stage output by the parameter mapping table; R total Indicates the target compression ratio; S2152, obtaining the titanium content and manganese content of the steel billet to be processed and the grain size after rough rolling, and calculating the intermediate rolling cooling rate and intermediate rolling reduction rate of the steel billet entering the intermediate rolling stage; The calculation formula of the intermediate rolling cooling rate is: Where, v T represents the cooling rate of intermediate rolling; ΔT represents the difference between the rough rolling temperature and the intermediate rolling temperature; Δt represents the time difference; [Mn] represents the manganese content; The calculation formula of the intermediate rolling reduction rate is: Where r 中轧 represents the intermediate rolling reduction rate; r2 represents the initial compression ratio of the intermediate rolling stage output by the parameter mapping table; [Ti] represents the titanium content; S2153, calculating the Ar3 phase transformation point of the steel billet to be processed, setting the finishing temperature in the finishing rolling stage according to the Ar3 phase transformation point, and calculating the finishing rolling reduction rate in the finishing rolling stage according to the carbon content; The calculation formula of the finishing rolling reduction rate is: Where r 精轧 represents the finishing reduction rate; r3 represents the initial compression ratio of the finishing stage output by the parameter mapping table.

2. The method for producing annealing-free cold heading steel hot rolled wire rod according to claim 1, characterized in that: The process of selecting a steel billet that meets the production specifications of cold heading steel wire rod and performing grinding, finishing and preheating on the steel billet to be processed includes: S11. Preset the production specifications of the cold heading steel wire rod, select a steel billet of corresponding size, and detect the content of each component element in the steel billet using a laser spectrometer; S12, using two grinding wheels to polish the billet, the first round uses a 16-20 mesh grinding wheel to remove wrinkles on the corners of the billet, and the second round uses a 24-26 mesh grinding wheel to finely polish the billet surface; S13. Place the ground steel billet in a preheating environment at 600-800°C, gradually heat the preheating environment to 1090-1170°C, and introduce a CO2 / N2 mixed gas to increase the surface hardness of the steel billet.

3. The method for producing annealing-free cold heading steel hot rolled wire rod according to claim 2, characterized in that: The setting of input variables and output variables, construction and training of a parameter recommendation model based on meta-learning, integration of physical metallurgical constraints, and matching of rolling parameters corresponding to different steel billets include: S2141. Assign each original task to a type of steel, set the production specifications of cold heading steel wire rod and the content of each component in the steel billet as input variables, and the rolling parameters as output variables to construct a support set; S2142. Linking a physical calculation module after the hidden layer output to implement physical information embedding for calculating the temperature and reduction rate of billet rolling at each stage; S2143. Setting two physical loss term functions, phase change loss and rolling force loss, and combining the phase change loss and rolling force loss to set a comprehensive loss function, thereby constructing a parameter recommendation model; S2144. Pre-train model parameters on all steel grade tasks to initially form a hot rolling knowledge base, and when facing new steel grades, fine-tune the calculation tasks through single-step gradient descent.

4. The method for producing annealing-free cold heading steel hot rolled wire rod according to claim 1, characterized in that: The real-time monitoring of hot rolling data during the hot rolling process, dynamic compensation of optimal rolling parameters, and dynamic adjustment of the pass interval time according to the monitoring and compensation results include: S221, using sensors inside the rolling mill group to collect hot rolling data of the billet rolling process in real time; S222, calculating the difference between the actual rolling force and the target rolling force of the rolling mill group during each stage of rolling, and calculating the reduction rate compensation amount through the proportional-integral coefficient; S223, calculating the temperature difference between the beginning and the end of the billet at each stage. If the temperature difference between the beginning and the end is greater than a preset threshold, the pass interval is extended to balance the temperature drop and the rolling rhythm. S224. Import the real-time hot rolling data into the historical parameter library, set the update cycle, and regularly update the comprehensive loss function. If there is a sudden change in the hot rolling data, an emergency stop signal is triggered.

5. The method for producing annealing-free cold heading steel hot rolled wire rod according to claim 1, characterized in that: The hot-rolled cold-heading steel wire rod is transported to a heat-insulating tunnel furnace, and the cold-heading steel wire rod is obtained through segmented controlled cooling and surface treatment, thereby realizing the production of annealing-free cold-heading steel hot-rolled wire rod. S31, obtaining and recording the parameters of the cold heading steel wire rod after hot rolling, wherein the wire rod parameters include the final rolling temperature, the grain size after rolling, the predicted ferrite ratio, and the rolling force curve; S32. Divide the cooling process of the cold heading steel wire rod into three stages: rapid cooling, slow cooling, and final cooling. Build a physical constraint meta-learning framework to generate cooling parameters to control the cooling of the cold heading steel wire rod in sections. S33. Dephosphorize the cooled cold heading steel wire rod using high pressure water, and identify surface defects through online laser detection. After the detection is correct, obtain the finished cold heading steel wire rod.

6. The method for producing annealing-free cold heading steel hot rolled wire rod according to claim 5, characterized in that: The cooling process of the cold heading steel wire rod is divided into three stages: rapid cooling, slow cooling, and final cooling. A physical constraint meta-learning framework is constructed to generate cooling parameters for segmented control of the cooling of the cold heading steel wire rod. The method includes: S321. Introduce wire rod parameters to expand the input layer, embed phase change dynamics equations and heat conduction equations to build a physical constraint layer, and build an output layer to output segmented cooling rate values and final cooling temperature; S322, dividing the cooling process of the cold heading steel wire rod into rapid cooling, slow cooling and final cooling stages, and calculating the cooling rate and cooling time of the rapid cooling stage respectively; S323. Based on the preset segmented air cooling strategy, a fan-water mist combination cooling is adopted in the rapid cooling stage, the fan is turned off in the slow cooling stage, and natural cooling is carried out by roller conveyor. In the final cooling stage, natural cooling is adopted and a nitrogen-carbon dioxide mixture is introduced to form a dense oxide film.

7. The method for producing annealing-free cold heading steel hot rolled wire rod according to claim 6, characterized in that: The calculation of the cooling rate and cooling time in the rapid cooling stage includes: The calculation formula of the rapid cooling rate in the rapid cooling stage is: Where, v 速冷 Indicates the rapid cooling rate in the rapid cooling stage; T 终轧 Indicates the final rolling temperature; T 速冷终 Indicates the target temperature in the rapid cooling stage; t 速冷 Indicates rapid cooling time; The calculation formula for cooling time in the slow cooling stage is: Where, t 缓冷 represents the slow cooling time; k0 represents the phase transition rate constant; E a represents the phase transition activation energy; R represents the gas constant; T 缓冷 Indicates slow cooling temperature; X total represents the target ferrite ratio; n represents the Avrami index; and e represents the natural constant.

Images