Anchoring rod body intelligent production method based on multi-agent and anchoring rod body

Abstract

The application discloses a kind of anchor rod body intelligent production method and anchor rod body based on multiple agents, and the target wall thickness and target rolling force determined by intelligent perforating body are used to carry out perforating treatment to base material, and hollow pipe blank is obtained;The power of each region in hollow pipe blank is adjusted by intelligent annealing body to realize isothermal annealing, and the optimal holding time is calculated, and once cold straightening is carried out using straightening machine;The hollow pipe blank after pickling is obtained after once straightening hollow pipe blank is pickled;The rolling force prediction value and the rolling piece outlet thickness prediction value of each pass obtained by intelligent finish rolling body are used to carry out multi-pass finish rolling forming to the hollow pipe blank after pickling, and rolled anchor rod is obtained;The rolled anchor rod is carried out secondary annealing treatment and secondary straightening;The rolled anchor rod after secondary straightening is carried out fixed-length cutting, cold-rolled rib and thread rolling processing, and anchor rod body is obtained.The application adaptively adjusts process parameters, improves the stability of product quality, and reduces energy consumption under the premise of ensuring annealing effect.

Description

Technical Field

[0001] This invention relates to the field of anchor bolt processing technology, and in particular to an intelligent production method for anchor bolt bodies based on multi-agent systems and the anchor bolt body itself. Background Technology

[0002] As a key support component in geotechnical engineering, anchor bolts are widely used in mining, tunneling, and slope protection projects. Their mechanical properties and dimensional accuracy directly affect project safety. Currently, the manufacturing process for anchor bolts includes heating and piercing → primary annealing and straightening → pickling → multi-pass precision rolling → secondary annealing and straightening → length cutting → cold rolling ribs and threads.

[0003] In the aforementioned process, the process parameters for each step (such as heating temperature and holding time) are preset to fixed values ​​based on experience. For example, the reduction rate in the finishing rolling stage is typically set to 10%-15%, and the annealing process uses an electric annealing furnace to hold the material at a fixed temperature range (780-820℃ for primary annealing and 580-620℃ for secondary annealing) for 3.5-6 hours, followed by cooling at a set rate. Consequently, this process lacks real-time perception and response capabilities to fluctuations in the incoming material condition and upstream processes, making it impossible to dynamically adjust process parameters. This easily leads to an increased proportion of defective products, affecting product quality stability. Furthermore, the fixed annealing holding time and cooling rate prevent dynamic optimization based on factors such as the ambient temperature of the workshop after rolling, resulting in over-annealing or ineffective energy consumption, leading to high energy consumption and low energy efficiency. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] To achieve the above objectives, the present invention proposes an intelligent production method for anchor bolts based on multi-agent systems, the method comprising:

[0006] The characteristic parameters of the piercing process are obtained, and the substrate is pierced by an intelligent piercing body based on the target wall thickness and target rolling force determined by the characteristic parameters of the piercing process to obtain a hollow tube blank; The hollow tube blank is fed into the intelligent annealing body. The intelligent annealing body dynamically adjusts the power of each region to achieve uniform temperature annealing by monitoring the surface and core temperature field distribution of the anchor rods in each region of the furnace in real time. The optimal holding time is dynamically calculated based on the feedback of the temperature field. After the optimal holding time ends and the tube is cooled, a straightening machine is used to perform a cold straightening. The hollow tube blank after one straightening is pickled to obtain the pickled hollow tube blank; The rolling force prediction value and the exit thickness prediction value of each pass are obtained through intelligent precision rolling. Based on the rolling force prediction value and the exit thickness prediction value of each pass, the pickled hollow tube blank is subjected to multi-pass precision rolling to obtain a rolled anchor rod with the required dimensions. The rolled anchor rod is fed into the intelligent annealing body for secondary annealing, and the straightening machine is used to straighten the rolled anchor rod after annealing. After secondary straightening, the rolled anchor rods are cut to length using a cutting device. The cut rolled anchor rods are then subjected to cold rib rolling and thread rolling to obtain the produced anchor rod body.

[0007] The intelligent manufacturing method of this invention may also have the following additional technical features: In one embodiment of the present invention, the characteristic parameters of the piercing process include: the actual measured diameter of the incoming material, the roll inclination angle, the roll speed, the real-time rolling force measurement value, the yield stress of the substrate at the current temperature, the working diameter of the roll, the approximate contact area between the roll and the metal, the interface temperature, and the heating temperature.

[0008] In one embodiment of the present invention, the intelligent piercing body includes a hybrid prediction model, which includes a physical mechanism model and a first error compensator; the step of piercing the substrate with the intelligent piercing body based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank includes: Based on the piercing process characteristic parameters, the predicted wall thickness and the first predicted rolling force are obtained through the wall thickness prediction mechanism model and the first rolling force prediction mechanism model in the physical mechanism model. Based on the piercing process characteristic parameters, the wall thickness compensation amount and the first rolling force compensation amount are obtained through the first error compensator; Based on the predicted wall thickness and the wall thickness compensation amount, the target wall thickness is obtained; Based on the first predicted rolling force and the first rolling force compensation, the target rolling force is obtained; The substrate is perforated based on the target wall thickness and the target rolling force to obtain a hollow tube blank.

[0009] In one embodiment of the present invention, the method further includes: real-time acquisition of rolling process parameters, and autonomous fine-tuning of the roll inclination angle and roll speed based on the rolling process parameters by the intelligent piercing body; the autonomous fine-tuning of the roll inclination angle and roll speed based on the rolling process parameters by the intelligent piercing body includes: Construct a first objective function that includes a wall thickness tracking term, a rolling force stabilization term, a control quantity change penalty term, and a constraint violation penalty term; Based on the rolling process parameters, the first objective function is solved using a sequential quadratic programming algorithm through the intelligent piercing body to obtain the optimal control sequence of the roll inclination angle and the roll speed; The intelligent perforating body sends the first control quantity in the optimal control sequence to the execution unit, thereby enabling autonomous fine-tuning of the roll tilt angle and the roll speed.

[0010] In one embodiment of the present invention, the intelligent annealing body includes a multi-zone coupled thermal field mixing prediction model and a phase transformation kinetic model; the intelligent annealing body dynamically adjusts the power of each zone to achieve uniform temperature annealing by real-time monitoring of the surface and core temperature field distribution of the anchor rods in each zone of the furnace, and dynamically calculates the optimal holding time based on the feedback of the temperature field, and performs a cold straightening process after the optimal holding time ends and the body is cooled, including: The intelligent annealing body uses a multispectral thermal imager array deployed inside the furnace to monitor the surface and core temperature field distribution of the anchor rods in various areas of the furnace in real time. Based on the surface and core temperature field distribution of the anchor bolts in each region, the surface temperature of the anchor bolts in each region in the next control cycle is obtained through the multi-region coupled thermal field mixing prediction model. The second objective function is solved based on the anchor surface temperature of each region in the next control cycle to obtain the target input electric power of each region. The power of each region is dynamically adjusted by the target input electric power of each region to achieve uniform temperature annealing. The second objective function includes a temperature tracking term, a power consumption term, and a power change rate term. Based on the feedback of the temperature field, the current phase transition fraction is calculated through the phase transition kinetic model, and the optimal holding time required to reach the target phase transition fraction is predicted. When the optimal holding time is reached and the holding is completed, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after the cooling is completed.

[0011] In one embodiment of the present invention, the multi-zone coupled thermal field mixing prediction model includes a lumped parameter mechanism model and a neural network spatial coupling compensator; the step of obtaining the anchor surface temperature of each region in the next control cycle based on the surface and core temperature field distribution of the anchor bolts in each region through the multi-zone coupled thermal field mixing prediction model includes: Based on the surface and core temperature field distribution of the anchor bolts in each region, the compensation heat flow of each region is obtained through the neural network spatial coupling compensator; Based on the lumped parameter mechanism model and the compensated heat flow of each region, the anchor surface temperature of each region in the next control cycle is obtained.

[0012] In one embodiment of the present invention, the neural network spatial coupling compensator is a graph neural network, wherein the nodes in the graph neural network correspond to each of the regions, and the edges in the graph neural network represent the adjacency relationships of the regions.

[0013] In one embodiment of the present invention, based on the feedback of the temperature field, the current phase transition fraction is calculated through the phase transition kinetic model, and the optimal holding time required to reach the target phase transition fraction is predicted. When the optimal holding time is reached and the holding is completed, a cooling command is triggered, and after cooling is completed, a cold-state straightening is performed using a straightening machine, including: Based on the feedback from the temperature field, the current equivalent time is calculated using the superposition principle; Based on the current equivalent time, the current phase transition fraction is calculated using the Avrami equation; Based on the target phase transition fraction, the temperature setpoint, and the current phase transition fraction, predict the optimal holding time required to reach the target phase transition fraction; When the optimal heat preservation time is less than or equal to zero, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after cooling is completed.

[0014] In one embodiment of the present invention, the intelligent finishing roll includes a mechanistic plasticity model and a second error compensator; obtaining the predicted rolling force and exit thickness for each pass through the intelligent finishing roll includes: The rolling deformation geometry parameters of each pass are obtained, and based on the rolling deformation geometry parameters, the second predicted rolling force and the predicted exit thickness of each pass are obtained through the second rolling force prediction mechanism model and the exit thickness prediction mechanism model in the mechanism plasticity model. Based on the rolling deformation geometry parameters of each pass, the second rolling force compensation amount and exit thickness compensation amount of each pass are obtained through the second error compensator. Based on the second rolling force compensation amount and the second predicted rolling force, the predicted rolling force value for each pass is obtained; Based on the export thickness compensation amount and the predicted export thickness, the predicted export thickness value for each pass is obtained.

[0015] To achieve the above objectives, another aspect of the present invention proposes an anchor rod body, which is obtained by the intelligent production method of anchor rod body based on multi-agent.

[0016] The intelligent production method for anchor rods based on multi-agent technology in this invention obtains characteristic parameters of the piercing process, and uses an intelligent piercing body to pierce the substrate based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank. The hollow tube blank is then fed into an intelligent annealing body, which dynamically adjusts the power of each region to achieve uniform annealing by monitoring the surface and core temperature field distribution of the anchor rod in each area of ​​the furnace in real time. The optimal holding time is dynamically calculated based on the temperature field feedback, and after the optimal holding time ends and cooling occurs, a straightening machine is used for a first-stage cold straightening. The straightened core is then... A hollow tube blank is pickled to obtain a pickled hollow tube blank. The predicted rolling force and exit thickness of the rolled piece for each pass are obtained through an intelligent finishing rolling mill. Based on these predicted values, the pickled hollow tube blank is subjected to multi-pass finishing rolling to obtain a rolled anchor rod with the required dimensions. The rolled anchor rod is then sent to an intelligent annealing mill for secondary annealing, and a straightening machine is used to straighten the annealed anchor rod. The straightened anchor rod is then cut to length using a cutting device, and the cut anchor rod undergoes cold rib rolling and thread rolling to obtain the produced anchor rod body. Therefore, this invention can adaptively adjust process parameters through an intelligent piercing mill, intelligent annealing mill, and intelligent finishing rolling mill, ensuring that each process always operates in an optimal state, eliminating deviations caused by manual adjustments, and improving the stability of product quality. Meanwhile, the power of each area can be independently adjusted in real time according to the temperature field distribution to achieve uniform annealing, avoiding overheating caused by fixed time in traditional processes, and significantly reducing energy consumption while ensuring annealing effect.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating an intelligent production method for anchor rods based on multiple agents according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of an intelligent production device for anchor rods based on multiple agents according to an embodiment of the present invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] The intelligent production method for anchor rods based on multiple agents according to embodiments of the present invention is described below with reference to the accompanying drawings.

[0022] Figure 1 This is a flowchart illustrating the intelligent production method for anchor rods based on multiple agents according to an embodiment of the present invention.

[0023] like Figure 1 As shown, the method may include the following steps: Step 101: Obtain the characteristic parameters of the piercing process, and use the intelligent piercing body to pierce the substrate based on the target wall thickness and target rolling force determined by the characteristic parameters of the piercing process to obtain a hollow tube blank.

[0024] In one embodiment of the present invention, the above-mentioned multi-agent may include an intelligent piercing body, an intelligent annealing body, and an intelligent finishing body.

[0025] In one embodiment of the present invention, during the skew rolling piercing process, the diameter tolerance of the incoming base material (40Cr alloy round steel), microscopic unevenness of heating temperature, and changes in the state of the equipment roll system can lead to fluctuations in the wall thickness of the pierced tube. Furthermore, in another embodiment of the present invention, traditional fixed parameter control or simple PID regulation cannot simultaneously ensure wall thickness accuracy, rolling process stability (force and vibration), and equipment safety, resulting in a high rate of out-of-tolerance wall thickness and significant equipment impact. Therefore, in one embodiment of the present invention, an intelligent piercing body can be used to pierce the base material based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank, thereby improving the stability of product quality.

[0026] In one embodiment of the present invention, the above-mentioned perforation process characteristic parameters may include: the actual measured diameter of the incoming material. Roll tilt angle Roll speed Real-time rolling force measurement value Yield stress of the substrate at the current temperature working diameter of the roll Approximate value of the contact area between the roll and the metal ,and , And related to the preset top position, interface temperature and heating temperature .

[0027] In one embodiment of the present invention, the intelligent piercing body is an intelligent system in the production of anchor rods that can intelligently calculate the target wall thickness and target rolling force based on the characteristic parameters of the piercing process, so as to stably produce high-precision hollow tube blanks in the future.

[0028] Furthermore, in one embodiment of the present invention, the aforementioned smart perforated body may include a hybrid prediction model, which includes a physical mechanism model and a first error compensator.

[0029] Specifically, in one embodiment of the present invention, the method for piercing a substrate using a smart piercing body based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank may include the following steps: Step 1011: Based on the characteristic parameters of the piercing process, the predicted wall thickness and the first predicted rolling force are obtained through the wall thickness prediction mechanism model and the first rolling force prediction mechanism model in the physical mechanism model.

[0030] In one embodiment of the present invention, the above-mentioned wall thickness prediction mechanism model can be:

[0031] in, To predict wall thickness, the above , , These are the coefficients of the mechanistic model.

[0032] In one embodiment of the present invention, the above-mentioned first rolling force prediction mechanism model can be:

[0033] in, The first predicted rolling force, friction coefficient ,in, The above are empirical coefficients calibrated through experiments. and These are the coefficients of the mechanistic model.

[0034] Step 1012: Based on the characteristic parameters of the piercing process, the wall thickness compensation amount and the first rolling force compensation amount are obtained through the first error compensator.

[0035] In one embodiment of the present invention, the first error compensator can be a lightweight neural network. Specifically, in one embodiment, the lightweight neural network may include a two-layer fully connected network with eight hidden neurons and a ReLU activation function.

[0036] Furthermore, in one embodiment of the present invention, the input vector of the first error compensator is... ,in, and From the vibration spectrum vector The two frequency band energy values ​​with the highest correlation to wall thickness deviation are extracted. And, the output vector obtained after inputting the above input vector into the first error compensator is... ,in, This is the wall thickness compensation amount. This is the first rolling force compensation amount.

[0037] Step 1013: Based on the predicted wall thickness and the wall thickness compensation amount, the target wall thickness is obtained.

[0038] In one embodiment of the present invention, the predicted wall thickness is obtained through the above steps. and wall thickness compensation amount Then, the predicted wall thickness and the wall thickness compensation amount can be added together to obtain the target wall thickness. .

[0039] Step 1014: Based on the first predicted rolling force and the first rolling force compensation amount, the target rolling force is obtained.

[0040] In one embodiment of the present invention, the first predicted rolling force is obtained through the above steps. and the first rolling force compensation amount Then, the first predicted rolling force and the first rolling force compensation amount can be added together to obtain the target rolling force. .

[0041] Furthermore, in one embodiment of the present invention, the mechanistic model coefficients in the aforementioned physical mechanism model and the parameters in the first error compensator are obtained through model training. In one embodiment of the present invention, the loss function in the aforementioned model training process is: ,in, , MSE is the mean square error function. This is the actual wall thickness. This is the actual value of the first rolling force.

[0042] Step 1015: The substrate is perforated based on the target wall thickness and target rolling force to obtain a hollow tube blank.

[0043] In one embodiment of the present invention, after obtaining the target wall thickness and target rolling force through the above steps, the substrate can be pierced by a piercing mill based on the target wall thickness and target rolling force to obtain a hollow tube blank.

[0044] Furthermore, in one embodiment of the present invention, the above method may further include: real-time acquisition of rolling process parameters, and autonomous fine-tuning of the roll inclination angle and roll speed based on the rolling process parameters by an intelligent piercing body.

[0045] In one embodiment of the present invention, the method for autonomously fine-tuning the roll inclination angle and roll speed based on rolling process parameters by an intelligent piercing body may include the following steps: Step 1: Construct a first objective function that includes a wall thickness tracking term, a rolling force stabilization term, a control quantity change penalty term, and a constraint violation penalty term.

[0046] In one embodiment of the present invention, the first objective function can be:

[0047] in, To predict the time domain, To control the time domain, and ; , , , , Positive weighting coefficients, such as =1.0, =0.5, =0.1, =0.1, =0.2; These are slack variables; To control the cycle, For period right Predicted wall thickness for the period, The target finished product thickness is... For period right The predicted value of the first rolling force in the cycle, p represents the prediction time domain. The variable i represents the control time domain. Variables.

[0048] as well as, , All are control quantity increments.

[0049] Furthermore, the above For wall thickness tracking, For the rolling force stabilization term, To control the penalty term for changes in the quantity, To constrain violations and penalties.

[0050] Furthermore, the constraints corresponding to the first objective function mentioned above may include: 1. Hard constraint 1: , ; 2. Hard constraint 2: , Within each control cycle; 3. Soft constraints: , , Vibrational energy, from and Regression with vibration characteristics yielded, For the rolling force safety threshold, The vibration energy safety threshold; 4. Slack variable constraints: .

[0051] Step 2: Based on the rolling process parameters, the first objective function is solved using a sequential quadratic programming algorithm through the intelligent piercing body to obtain the optimal control sequence of roll inclination angle and roll speed.

[0052] In one embodiment of the present invention, the intelligent piercing body can be solved in real time using a sequential quadratic programming (SQP) algorithm based on rolling process parameters to obtain the optimal control sequence of roll tilt angle and roll speed within the next Nc control cycles. .

[0053] Step 3: The intelligent perforating body sends the first control quantity in the optimal control sequence to the execution unit to realize autonomous fine-tuning of the roll tilt angle and roll speed.

[0054] In one embodiment of the present invention, after obtaining the optimal control sequence through the above steps, the intelligent perforated body can control the first control quantity in the optimal control sequence. The data is sent to the execution unit (servo motor and hydraulic tilt adjustment mechanism) to achieve autonomous fine-tuning of the roll tilt angle and roll speed. In one embodiment of the invention, the aforementioned intelligent perforating body can periodically perform autonomous fine-tuning of the roll tilt angle and roll speed.

[0055] Step 102: The hollow tube blank is fed into the intelligent annealing body. The intelligent annealing body dynamically adjusts the power of each area to achieve uniform annealing by monitoring the surface and core temperature field distribution of the anchor rods in each area of ​​the furnace in real time. The optimal holding time is dynamically calculated based on the temperature field feedback. After the optimal holding time ends and the tube is cooled, a straightening machine is used to perform a cold straightening.

[0056] In one embodiment of the present invention, after obtaining the hollow tube blank through the above steps, the hollow tube blank can be sent into the intelligent annealing body. The intelligent annealing body dynamically adjusts the power of each area to achieve uniform temperature annealing by real-time monitoring of the surface and core temperature field distribution of the anchor rods in each area of ​​the furnace. It also dynamically calculates the optimal holding time based on the feedback of the temperature field, and performs a cold straightening once after the optimal holding time ends and the tube is cooled.

[0057] In one embodiment of the present invention, the intelligent annealing body is an intelligent system that can monitor the temperature field of the anchor surface and core in each area of ​​the furnace in real time, achieve uniform temperature in the furnace through dynamic adjustment of zoned power, and intelligently calculate the optimal holding time based on temperature field feedback to complete the precise and intelligent annealing of hollow tube billets.

[0058] In one embodiment of the present invention, the aforementioned intelligent annealing body includes a multi-zone coupled thermal field mixing prediction model and a phase transition dynamics model.

[0059] Furthermore, in one embodiment of the present invention, the above-mentioned intelligent annealing body dynamically adjusts the power of each region to achieve uniform annealing by real-time monitoring of the surface and core temperature field distribution of the anchor rods in each region of the furnace, and dynamically calculates the optimal holding time based on the temperature field feedback. The method of performing a cold straightening process using a straightening machine after the optimal holding time ends and the body is cooled may include the following steps: Step 1021: The intelligent annealing body uses a multispectral thermal imager array deployed inside the furnace to monitor the surface and core temperature field distribution of the anchor rods in various areas of the furnace in real time.

[0060] In one embodiment of the present invention, a multispectral thermal imager array deployed inside the furnace can be used to monitor the surface and core temperature field distribution of the anchor bolts in various regions of the furnace in real time. Specifically, in one embodiment of the present invention, regarding the surface and core temperature field distribution... For the region exist The temperature of the anchor core at any given moment. For the region exist The average surface temperature of the anchor bolt at any given time.

[0061] Step 1022: Based on the surface and core temperature field distribution of the anchor bolts in each region, the surface temperature of the anchor bolts in each region in the next control cycle is obtained through a multi-region coupled thermal field mixing prediction model.

[0062] In one embodiment of the present invention, the multi-zone coupled thermal field mixing prediction model may include a lumped parameter mechanism model and a neural network spatial coupling compensator.

[0063] Furthermore, in one embodiment of the present invention, the method for obtaining the surface temperature of the anchor bolts in each region for the next control cycle based on the surface and core temperature field distribution of the anchor bolts in each region through a multi-region coupled thermal field mixing prediction model may include the following steps: Step 10221: Based on the surface and core temperature field distribution of the anchor bolts in each region, the compensation heat flow of each region is obtained through a neural network spatial coupling compensator.

[0064] In one embodiment of the present invention, the above-mentioned neural network spatial coupling compensator can be a graph neural network (GNN), in which nodes correspond to each region and edges represent adjacent relationships.

[0065] Furthermore, in one embodiment of the present invention, the input vector of the aforementioned neural network spatial coupling compensator is [ ],in, It is a region Input electrical power, This indicates the furnace door is in open or closed position. This represents the cooling water flow rate. Furthermore, the output of the aforementioned neural network spatial coupling compensator represents the compensated heat flow for each region. , For the region Compensating heat flow, For the region The set of adjacent regions, For the region The average surface temperature of the anchor bolt.

[0066] Step 10222: Based on the lumped parameter mechanism model and the compensated heat flow of each region, the surface temperature of the anchor bolt in each region is obtained in the next control cycle.

[0067] In one embodiment of the present invention, the compensated heat flux of each region is obtained through the above steps. Then, based on the lumped parameter mechanism model and the compensation heat flow of each region, the surface temperature of the anchor bolt in each region can be obtained in the next control cycle.

[0068] In one embodiment of the present invention, the above-mentioned lumped parameter mechanism model is as follows:

[0069] in, For the region heat capacity, , For the region The load quality, The specific heat capacity at constant pressure of the material. For the region External heat loss function, For ambient temperature, This represents the inter-regional thermal coupling coefficient.

[0070] Furthermore, in one embodiment of the present invention, based on the above-described lumped parameter mechanism model and the compensated heat flow of each region, the anchor surface temperature of each region in the next control cycle is obtained as follows:

[0071] in, The control period is the discrete time step. To control the cycle, and Nominal constant coefficient, For ambient temperature, the nonlinear time-varying component is compensated by heat flux. Compensation will be provided.

[0072] Step 1023: Solve the second objective function based on the anchor surface temperature of each region in the next control cycle to obtain the target input electric power of each region, and dynamically adjust the power of each region to achieve uniform temperature annealing by means of the target input electric power of each region. The second objective function includes a temperature tracking term, a power consumption term, and a power change rate term.

[0073] In one embodiment of the present invention, the second objective function is:

[0074] Where M represents the total number of regions. , , , The positive weighting coefficients are... For temperature setpoint, For period right Predicted average surface temperature of anchor bolts. For region z in periodicity Input electrical power, and These are slack variables. And, For temperature tracking items, For power consumption items and For the power change rate term, This is a penalty item for violating soft constraints.

[0075] Furthermore, in one embodiment of the present invention, the constraints corresponding to the second objective function include: 1. System dynamic constraints: ; 2. Hard constraints: , , ; in, For the region The upper limit of power, T max Maximum temperature protection; 3. Soft constraints: , , in, This is the upper limit of total power. It is designed for maximum temperature protection.

[0076] Furthermore, in one embodiment of the present invention, the second objective function can be solved using an interior-point method solver based on the anchor surface temperature of each region in the next control cycle to obtain the optimal input power allocation sequence. and the first input electrical power sequence The target input electrical power for each region in the current control cycle is determined, and the power of each region is dynamically adjusted based on the target input electrical power to achieve uniform temperature annealing.

[0077] Step 1024: Based on the feedback of the temperature field, calculate the current phase transition fraction through the phase transition kinetic model and predict the optimal holding time required to reach the target phase transition fraction. When the optimal holding time is reached and the holding is completed, trigger the cooling command and perform a cold-state straightening using a straightening machine after cooling is completed.

[0078] In one embodiment of the present invention, the method of calculating the current phase transition fraction and predicting the optimal holding time required to reach the target phase transition fraction based on the feedback of the temperature field, triggering a cooling command when the optimal holding time is reached and the holding is completed, and performing a cold-state straightening using a straightening machine after cooling can include the following steps: Step 10241: Based on the feedback of the temperature field, calculate the current equivalent time using the superposition principle.

[0079] In one embodiment of the present invention, based on the feedback of the temperature field, the superposition principle (Scheil superposition method) can be used to discretize the continuous temperature history into a series of isothermal steps. Specifically, at time step t... n At that time, the superposition principle is used to calculate the current equivalent time. , where Δτ j Let Q be the duration of the j-th time step, Q be the activation energy of the phase transition, and R be the gas constant. For the first Time step temperature, This represents the total number of time steps.

[0080] Step 10242: Calculate the current phase transition fraction using the Avrami equation based on the current equivalent time.

[0081] In one embodiment of the present invention, the current phase transition fraction is calculated using the Avrami equation based on the current equivalent time. , where K and The Avrami coefficient is related to the material composition (e.g., 40Cr) and is determined through preliminary process experiments.

[0082] Step 10243: Based on the target phase change fraction, temperature setpoint, and current phase change fraction, predict the optimal holding time required to reach the target phase change fraction.

[0083] In one embodiment of the present invention, the temperature setpoint is: To achieve the target phase transition fraction The optimal holding time required to achieve the target phase transition fraction is:

[0084] In one embodiment of the present invention, the above-mentioned optimal heat preservation time can be the remaining heat preservation time.

[0085] Step 10244: When the optimal heat preservation time is less than or equal to zero, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after cooling is completed.

[0086] Step 103: Pickling is performed on the hollow tube blank after the first straightening to obtain the pickled hollow tube blank.

[0087] In one embodiment of the present invention, the hollow tube blank after initial straightening is pickled to remove oxide scale from its surface. The pickling stage removes the oxide scale generated on the surface after annealing, preventing oxide inclusions from affecting the subsequent finishing rolling process and avoiding the initiation of microcracks, thus helping to maintain the integrity of the interface structure after processing.

[0088] Step 104: The rolling force prediction value and the exit thickness prediction value of each pass are obtained through intelligent precision rolling. Based on the rolling force prediction value and the exit thickness prediction value of each pass, the pickled hollow tube blank is subjected to multi-pass precision rolling to obtain a rolled anchor rod with the required dimensions.

[0089] In one embodiment of the present invention, after obtaining the pickled hollow tube blank through the above steps, the predicted rolling force value and the predicted exit thickness value of the rolled piece for each pass can be obtained through intelligent precision rolling. Based on the predicted rolling force value and the predicted exit thickness value of the rolled piece for each pass, the pickled hollow tube blank is subjected to multi-pass precision rolling to obtain a rolled anchor rod with the required dimensions.

[0090] In one embodiment of the present invention, the intelligent finishing roll body is an intelligent system that can intelligently predict the rolling force and the thickness of the rolled piece at the exit of each finishing roll, and perform multi-pass precision finishing rolls on the pickled hollow tube blank based on the predicted parameters to form a qualified dimensional anchor rod.

[0091] In one embodiment of the present invention, the aforementioned intelligent precision rolled body may include a mechanistic plasticity model and a second error compensator.

[0092] Furthermore, in one embodiment of the present invention, the method for obtaining the predicted rolling force and exit thickness of each pass through intelligent finishing mill can include the following steps: Step 1041: Obtain the rolling deformation geometric parameters for each pass, and based on the rolling deformation geometric parameters, use the second rolling force prediction mechanism model and the exit thickness prediction mechanism model in the mechanistic plasticity model to obtain the second predicted rolling force and the predicted exit thickness for each pass.

[0093] In one embodiment of the present invention, the above-mentioned second rolling force prediction mechanism model is as follows:

[0094] in, Width of rolled piece, contact arc length ,in, Where is the radius of the roll. For the first Pass-by pressure reduction For the first Stress state coefficient of each pass, average deformation resistance .

[0095] And, inlet deformation resistance Export deformation resistance This path truly responds to change ,in, For the first The inlet temperature of each track, For the first The exit temperature of each track, To accumulate true response, For strain rate, For the first The thickness of the entrance to the passage. For the first The thickness of the exit of each pass.

[0096] In one embodiment of the present invention, the above-mentioned outlet thickness prediction mechanism model is as follows:

[0097] in, For the first The empty roll gap of the track, For the first The rigidity of the rolling mill for each pass.

[0098] Furthermore, in one embodiment of the present invention, the second predicted rolling force for each pass can be obtained through the aforementioned second rolling force prediction mechanism model and exit thickness prediction mechanism model. and predicted export thickness .

[0099] Step 1042: Based on the rolling deformation geometry parameters of each pass, the second rolling force compensation amount and exit thickness compensation amount of each pass are obtained through the second error compensator.

[0100] In one embodiment of the present invention, the second error compensator described above can be a random forest regression.

[0101] In one embodiment of the present invention, the input vector of the second error compensator is: =[ ],in, For the first The surface temperature of the rolls in each pass, For the first The bearing vibration characteristics of each pass, This is due to the deviation of the previous force. For the first The rotational speed of the rolls in each pass, For the first The amount of reduction per pass. And, in one embodiment of the invention, the output of the second error compensator is a second rolling force compensation amount. and export thickness compensation .

[0102] Step 1043: Based on the second rolling force compensation amount and the second predicted rolling force, the predicted rolling force value for each pass is obtained.

[0103] In one embodiment of the present invention, after obtaining the second rolling force compensation amount and the second predicted rolling force through the above steps, the second rolling force compensation amount and the second predicted rolling force can be added together to obtain the predicted rolling force value for each pass. .

[0104] Step 1044: Based on the export thickness compensation amount and the predicted export thickness, obtain the predicted export thickness value for each pass.

[0105] In one embodiment of the present invention, by obtaining the exit thickness compensation amount and the predicted exit thickness through the above steps, the predicted exit thickness value for each pass can be obtained based on the exit thickness compensation amount and the predicted exit thickness. .

[0106] Furthermore, in one embodiment of the present invention, after obtaining the predicted rolling force and the predicted thickness of the rolled piece at the exit of each pass through the above steps, the pickled hollow tube blank can be subjected to multi-pass precision rolling based on the predicted rolling force and the predicted thickness of the rolled piece at the exit of each pass to obtain a rolled anchor rod with the required dimensions.

[0107] Furthermore, in one embodiment of the present invention, the above method may further include: after each rolling pass, calculating and fine-tuning the reduction amount of the next pass using intelligent finishing mill, until all rolling passes are completed.

[0108] Specifically, in one embodiment of the present invention, the method of calculating and fine-tuning the reduction amount of the next pass through intelligent finishing roll after each rolling pass until all rolling passes are completed may include the following steps: S1, construct a third objective function including the final thickness deviation term, the reduction deviation term, and the reduction smoothness term.

[0109] In one embodiment of the present invention, the third objective function is:

[0110] in, The total number of times. This is the predicted thickness of the finished product; This refers to the reduction amount in the original procedure; w1, w2, and w3 are weighting coefficients. And, For the final thickness deviation, The degree of deviation from the original procedure, This refers to the smoothness of the change in compression amount.

[0111] S2, after the current u-th rolling pass is completed, obtain the measured exit thickness of the current pass as the entry thickness of the next pass.

[0112] S3, by optimizing the third objective function under the conditions of satisfying process constraints and variable boundary constraints, the optimal reduction amount of the (u+1)th pass is obtained.

[0113] In one embodiment of the present invention, the above-mentioned process constraints include: Rolling force constraint: ≤ , For the first The rolling force constraint value for each pass.

[0114] In one embodiment of the present invention, the above-mentioned variable boundary constraints include: ≤ ≤ , For the first Minimum reduction per pass For the first The maximum amount of pressure applied per pass.

[0115] Furthermore, in one embodiment of the present invention, the above is a constrained nonlinear programming (NLP) problem, and the intelligent finishing mill can be solved using a sequential quadratic programming (SQP) algorithm to obtain the optimal sequence of subsequent reduction amounts. and will The optimal reduction amount for the (u+1)th pass was determined.

[0116] S4. Apply the optimal reduction amount of the (u+1)th pass to the upcoming (u+1)th pass rolling, and repeat steps S2 to S4 after the (u+1)th pass rolling is completed, until all passes rolling is completed.

[0117] Step 105: The rolled anchor rod is sent into the intelligent annealing body for secondary annealing treatment, and the rolled anchor rod after annealing is straightened for the second time using a straightening machine.

[0118] In one embodiment of the present invention, after obtaining the rolled anchor bolt through the above steps, the rolled anchor bolt can be sent into an intelligent annealing body for secondary annealing treatment, and a straightening machine is used to perform secondary straightening on the annealed rolled anchor bolt. In one embodiment of the present invention, the method for secondary annealing treatment in the intelligent annealing body can be referred to the relevant description in the above embodiments, and will not be repeated here.

[0119] Step 106: The rolled anchor rod after secondary straightening is cut to length using a cutting device, and the cut rolled anchor rod is then cold rolled for ribs and thread rolling to obtain the produced anchor rod body.

[0120] In one embodiment of the present invention, ribs and threads can be formed by cold rolling. Since the rod structure is already stable and has good plasticity, the rolling deformation is more uniform. Microscopically, local dislocation density and work hardening occur in the rib and thread areas, forming a strong and tough bonding zone, improving local tensile strength and connection performance, while maintaining overall plasticity and structural density.

[0121] The intelligent production method for anchor rods based on multi-agent technology in this invention obtains characteristic parameters of the piercing process, and uses an intelligent piercing body to pierce the substrate based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank. The hollow tube blank is then fed into an intelligent annealing body, which dynamically adjusts the power of each region to achieve uniform annealing by monitoring the surface and core temperature field distribution of the anchor rod in each area of ​​the furnace in real time. The optimal holding time is dynamically calculated based on the temperature field feedback, and after the optimal holding time ends and cooling occurs, a straightening machine is used for a first-stage cold straightening. The straightened core is then... A hollow tube blank is pickled to obtain a pickled hollow tube blank. The predicted rolling force and exit thickness of the rolled piece for each pass are obtained through an intelligent finishing rolling mill. Based on these predicted values, the pickled hollow tube blank is subjected to multi-pass finishing rolling to obtain a rolled anchor rod with the required dimensions. The rolled anchor rod is then sent to an intelligent annealing mill for secondary annealing, and a straightening machine is used to straighten the annealed anchor rod. The straightened anchor rod is then cut to length using a cutting device, and the cut anchor rod undergoes cold rib rolling and thread rolling to obtain the produced anchor rod body. Therefore, this invention can adaptively adjust process parameters through an intelligent piercing mill, intelligent annealing mill, and intelligent finishing rolling mill, ensuring that each process always operates in an optimal state, eliminating deviations caused by manual adjustments, and improving the stability of product quality. Meanwhile, the power of each area can be independently adjusted in real time according to the temperature field distribution to achieve uniform annealing, avoiding overheating caused by fixed time in traditional processes, and significantly reducing energy consumption while ensuring annealing effect.

[0122] Figure 2 This is a schematic diagram of the structure of the intelligent production device for anchor rods based on multiple agents according to an embodiment of the present invention.

[0123] like Figure 2 As shown, the device may include: The determination module 201 is used to obtain the characteristic parameters of the piercing process, and to perform piercing treatment on the substrate by the intelligent piercing body based on the target wall thickness and target rolling force determined by the characteristic parameters of the piercing process to obtain a hollow tube blank; The first annealing module 202 is used to feed the hollow tube blank into the intelligent annealing body. The intelligent annealing body dynamically adjusts the power of each area to achieve uniform annealing by monitoring the surface and core temperature field distribution of the anchor rods in each area of ​​the furnace in real time. It also dynamically calculates the optimal holding time based on the temperature field feedback, and performs a cold straightening process after the optimal holding time ends and the tube is cooled. Pickling module 203 is used to pickle the hollow tube blank after one straightening to obtain the pickled hollow tube blank; The prediction module 204 is used to obtain the predicted rolling force and the predicted thickness of the rolled piece at the exit of each pass through the intelligent finishing mill, and to perform multi-pass finishing milling on the pickled hollow tube blank based on the predicted rolling force and the predicted thickness of the rolled piece at the exit of each pass to obtain a rolled anchor rod with the required dimensions. The second annealing module 205 is used to send the rolled anchor rod into the intelligent annealing body for secondary annealing treatment, and to use a straightening machine to perform secondary straightening on the rolled anchor rod after annealing. The cutting module 206 is used to cut the rolled anchor rods to length using a cutting device after secondary straightening, and to perform cold rib rolling and thread rolling on the cut rolled anchor rods to obtain the produced anchor rod body.

[0124] In one embodiment of the present invention, the aforementioned smart perforated body includes a hybrid prediction model, which includes a physical mechanism model and a first error compensator; the aforementioned determining module 201 is specifically used for: Based on the characteristic parameters of the piercing process, the predicted wall thickness and the first predicted rolling force are obtained through the wall thickness prediction mechanism model and the first rolling force prediction mechanism model in the physical mechanism model. Based on the characteristic parameters of the piercing process, the wall thickness compensation amount and the first rolling force compensation amount are obtained through the first error compensator; The target wall thickness is obtained based on the predicted wall thickness and the wall thickness compensation amount; The target rolling force is obtained based on the first predicted rolling force and the first rolling force compensation amount; The substrate is perforated based on the target wall thickness and target rolling force to obtain a hollow tube blank.

[0125] In one embodiment of the present invention, the above-mentioned device is further used for: real-time acquisition of rolling process parameters, and autonomous fine-tuning of the roll inclination angle and roll speed based on the rolling process parameters by an intelligent piercing body; the above-mentioned device is specifically used for: Construct a first objective function that includes a wall thickness tracking term, a rolling force stabilization term, a control quantity change penalty term, and a constraint violation penalty term; Based on the rolling process parameters, the first objective function is solved by a sequential quadratic programming algorithm using an intelligent piercing body, thereby obtaining the optimal control sequence for the roll inclination angle and roll speed. The intelligent perforating body sends the first control quantity in the optimal control sequence to the execution unit, enabling autonomous fine-tuning of the roll tilt angle and roll speed.

[0126] In one embodiment of the present invention, the aforementioned intelligent annealing body includes a multi-zone coupled thermal field mixing prediction model and a phase transition dynamics model; the aforementioned first annealing module 202 is specifically used for: The intelligent annealing body uses a multispectral thermal imager array deployed inside the furnace to monitor the surface and core temperature field distribution of the anchor rods in various areas of the furnace in real time; Based on the surface and core temperature field distribution of anchor bolts in each region, the surface temperature of anchor bolts in each region in the next control cycle is obtained through a multi-region coupled thermal field mixing prediction model. The second objective function is solved based on the anchor surface temperature of each region in the next control cycle to obtain the target input electric power of each region. The power of each region is dynamically adjusted by the target input electric power of each region to achieve uniform temperature annealing. The second objective function includes a temperature tracking term, a power consumption term, and a power change rate term. Based on the feedback of the temperature field, the current phase transition fraction is calculated through the phase transition kinetic model, and the optimal holding time required to reach the target phase transition fraction is predicted. When the optimal holding time is reached and the holding is completed, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after the cooling is completed.

[0127] In one embodiment of the present invention, the multi-zone coupled thermal field mixing prediction model includes a lumped parameter mechanism model and a neural network spatial coupling compensator; the first annealing module 202 is further used for: Based on the surface and core temperature field distribution of anchor bolts in each region, the compensation heat flow of each region is obtained through a neural network spatial coupling compensator; Based on the lumped parameter mechanism model and the compensation heat flow in each region, the surface temperature of the anchor bolt in each region is obtained in the next control cycle.

[0128] In one embodiment of the present invention, the first annealing module 202 is further configured to: Based on the feedback of the temperature field, the current equivalent time is calculated using the superposition principle; Based on the current equivalent time, the current phase transition fraction is calculated using the Avrami equation; Based on the target phase transition fraction, temperature setpoint, and current phase transition fraction, predict the optimal holding time required to reach the target phase transition fraction; When the optimal heat preservation time is less than or equal to zero, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after cooling is completed.

[0129] In one embodiment of the present invention, the aforementioned intelligent precision rolled body includes a mechanistic plasticity model and a second error compensator; the aforementioned prediction module 204 is specifically used for: The rolling deformation geometry parameters of each pass are obtained, and based on the rolling deformation geometry parameters, the second predicted rolling force and the predicted exit thickness of each pass are obtained through the second rolling force prediction mechanism model and the exit thickness prediction mechanism model in the mechanistic plasticity model. Based on the rolling deformation geometry parameters of each pass, the second rolling force compensation amount and exit thickness compensation amount of each pass are obtained through the second error compensator. Based on the second rolling force compensation and the second predicted rolling force, the predicted rolling force value for each pass is obtained. Based on the export thickness compensation amount and the predicted export thickness, the predicted export thickness value for each pass is obtained.

[0130] In one embodiment of the present invention, the above-described apparatus is further configured to: S1, construct a third objective function including the final thickness deviation term, the reduction deviation term, and the reduction smoothness term; S2, after the current u-th rolling pass is completed, obtain the measured exit thickness of the current pass as the entry thickness of the next pass; S3, by optimizing the third objective function under the conditions of satisfying process constraints and variable boundary constraints, the optimal reduction amount of the (u+1)th pass is obtained; S4. Apply the optimal reduction amount of the (u+1)th pass to the upcoming (u+1)th pass rolling, and repeat steps S2 to S4 after the (u+1)th pass rolling is completed, until all passes rolling is completed.

[0131] The intelligent production device for anchor rods based on multi-agent technology in this invention acquires characteristic parameters of the piercing process, and uses an intelligent piercing body to pierce the substrate based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank. The hollow tube blank is then fed into an intelligent annealing body, which dynamically adjusts the power of each region to achieve uniform annealing by real-time monitoring of the surface and core temperature field distribution of the anchor rod in each area of ​​the furnace. The body also dynamically calculates the optimal holding time based on the temperature field feedback, and performs a cold straightening process using a straightening machine after the optimal holding time ends and cooling is completed. The straightened tube blank is then subjected to a cold straightening process. A hollow tube blank is pickled to obtain a pickled hollow tube blank. The predicted rolling force and exit thickness of the rolled piece for each pass are obtained through an intelligent finishing rolling mill. Based on these predicted values, the pickled hollow tube blank is subjected to multi-pass finishing rolling to obtain a rolled anchor rod with the required dimensions. The rolled anchor rod is then sent to an intelligent annealing mill for secondary annealing, and a straightening machine is used to straighten the annealed anchor rod. The straightened anchor rod is then cut to length using a cutting device, and the cut anchor rod undergoes cold rib rolling and thread rolling to obtain the produced anchor rod body. Therefore, this invention can adaptively adjust process parameters through an intelligent piercing mill, intelligent annealing mill, and intelligent finishing rolling mill, ensuring that each process always operates in an optimal state, eliminating deviations caused by manual adjustments, and improving the stability of product quality. Meanwhile, the power of each area can be independently adjusted in real time according to the temperature field distribution to achieve uniform annealing, avoiding overheating caused by fixed time in traditional processes, and significantly reducing energy consumption while ensuring annealing effect.

[0132] This invention proposes an anchor rod body, which is obtained through an intelligent production method for anchor rod bodies based on multi-agent technology.

[0133] In this specification, the use of terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refers to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0134] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for intelligent production of anchor bolts based on multi-agent systems, characterized in that, The method includes: The characteristic parameters of the piercing process are obtained, and the substrate is pierced by an intelligent piercing body based on the target wall thickness and target rolling force determined by the characteristic parameters of the piercing process to obtain a hollow tube blank; The hollow tube blank is fed into the intelligent annealing body. The intelligent annealing body dynamically adjusts the power of each region to achieve uniform temperature annealing by monitoring the surface and core temperature field distribution of the anchor rods in each region of the furnace in real time. The optimal holding time is dynamically calculated based on the feedback of the temperature field. After the optimal holding time ends and the tube is cooled, a straightening machine is used to perform a cold straightening. The hollow tube blank after one straightening is pickled to obtain the pickled hollow tube blank; The rolling force prediction value and the exit thickness prediction value of each pass are obtained through intelligent precision rolling. Based on the rolling force prediction value and the exit thickness prediction value of each pass, the pickled hollow tube blank is subjected to multi-pass precision rolling to obtain a rolled anchor rod with the required dimensions. The rolled anchor rod is fed into the intelligent annealing body for secondary annealing, and the straightening machine is used to straighten the rolled anchor rod after annealing. After secondary straightening, the rolled anchor rods are cut to length using a cutting device. The cut rolled anchor rods are then subjected to cold rib rolling and thread rolling to obtain the produced anchor rod body.

2. The method according to claim 1, characterized in that, The characteristic parameters of the piercing process include: the actual diameter of the incoming material, the roll inclination angle, the roll speed, the real-time rolling force measurement value, the yield stress of the substrate at the current temperature, the working diameter of the roll, the approximate contact area between the roll and the metal, the interface temperature, and the heating temperature.

3. The method according to claim 2, characterized in that, The intelligent piercing body includes a hybrid prediction model, which includes a physical mechanism model and a first error compensator; the process of piercing the substrate using the intelligent piercing body based on the target wall thickness and target rolling force determined by the piercing process characteristic parameters to obtain a hollow tube blank includes: Based on the piercing process characteristic parameters, the predicted wall thickness and the first predicted rolling force are obtained through the wall thickness prediction mechanism model and the first rolling force prediction mechanism model in the physical mechanism model. Based on the piercing process characteristic parameters, the wall thickness compensation amount and the first rolling force compensation amount are obtained through the first error compensator; Based on the predicted wall thickness and the wall thickness compensation amount, the target wall thickness is obtained; Based on the first predicted rolling force and the first rolling force compensation, the target rolling force is obtained; The substrate is perforated based on the target wall thickness and the target rolling force to obtain a hollow tube blank.

4. The method according to claim 3, characterized in that, The method further includes: real-time acquisition of rolling process parameters, and autonomous fine-tuning of the roll inclination angle and roll speed based on the rolling process parameters using the intelligent piercing body; the autonomous fine-tuning of the roll inclination angle and roll speed based on the rolling process parameters using the intelligent piercing body includes: Construct a first objective function that includes a wall thickness tracking term, a rolling force stabilization term, a control quantity change penalty term, and a constraint violation penalty term; Based on the rolling process parameters, the first objective function is solved using a sequential quadratic programming algorithm through the intelligent piercing body to obtain the optimal control sequence of the roll inclination angle and the roll speed; The intelligent perforating body sends the first control quantity in the optimal control sequence to the execution unit, thereby enabling autonomous fine-tuning of the roll tilt angle and the roll speed.

5. The method according to claim 1, characterized in that, The intelligent annealing body includes a multi-zone coupled thermal field mixing prediction model and a phase transformation kinetic model. The intelligent annealing body dynamically adjusts the power of each zone to achieve uniform annealing by real-time monitoring of the surface and core temperature field distribution of the anchor bolts in each zone of the furnace. Based on the feedback from the temperature field, the optimal holding time is dynamically calculated. After the optimal holding time ends and cooling occurs, a cold straightening process is performed using a straightening machine, including: The intelligent annealing body uses a multispectral thermal imager array deployed inside the furnace to monitor the surface and core temperature field distribution of the anchor rods in various areas of the furnace in real time. Based on the surface and core temperature field distribution of the anchor bolts in each region, the surface temperature of the anchor bolts in each region in the next control cycle is obtained through the multi-region coupled thermal field mixing prediction model. The second objective function is solved based on the anchor surface temperature of each region in the next control cycle to obtain the target input electric power of each region. The power of each region is dynamically adjusted by the target input electric power of each region to achieve uniform temperature annealing. The second objective function includes a temperature tracking term, a power consumption term, and a power change rate term. Based on the feedback of the temperature field, the current phase transition fraction is calculated through the phase transition kinetic model, and the optimal holding time required to reach the target phase transition fraction is predicted. When the optimal holding time is reached and the holding is completed, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after the cooling is completed.

6. The method according to claim 5, characterized in that, The multi-zone coupled thermal field hybrid prediction model includes a lumped parameter mechanism model and a neural network spatial coupling compensator; the process of obtaining the anchor surface temperature of each region in the next control cycle based on the surface and core temperature field distribution of the anchor bolts in each region through the multi-zone coupled thermal field hybrid prediction model includes: Based on the surface and core temperature field distribution of the anchor bolts in each region, the compensation heat flow of each region is obtained through the neural network spatial coupling compensator; Based on the lumped parameter mechanism model and the compensated heat flow of each region, the anchor surface temperature of each region in the next control cycle is obtained.

7. The method according to claim 6, characterized in that, The neural network spatial coupling compensator is a graph neural network, wherein the nodes in the graph neural network correspond to each of the regions, and the edges in the graph neural network represent the adjacency relationships of the regions.

8. The method according to claim 5, characterized in that, Based on the feedback from the temperature field, the current phase transition fraction is calculated using the phase transition kinetic model, and the optimal holding time required to reach the target phase transition fraction is predicted. When the optimal holding time is reached and the holding is completed, a cooling command is triggered, and after cooling, a cold-state straightening is performed using a straightening machine, including: Based on the feedback from the temperature field, the current equivalent time is calculated using the superposition principle; Based on the current equivalent time, the current phase transition fraction is calculated using the Avrami equation; Based on the target phase transition fraction, the temperature setpoint, and the current phase transition fraction, predict the optimal holding time required to reach the target phase transition fraction; When the optimal heat preservation time is less than or equal to zero, a cooling command is triggered, and a cold-state straightening is performed using a straightening machine after cooling is completed.

9. The method according to claim 1, characterized in that, The intelligent precision-rolled body includes a mechanistic plasticity model and a second error compensator; The method of obtaining the predicted rolling force and exit thickness for each pass through intelligent finishing mill includes: The rolling deformation geometry parameters of each pass are obtained, and based on the rolling deformation geometry parameters, the second predicted rolling force and the predicted exit thickness of each pass are obtained through the second rolling force prediction mechanism model and the exit thickness prediction mechanism model in the mechanism plasticity model. Based on the rolling deformation geometry parameters of each pass, the second rolling force compensation amount and exit thickness compensation amount of each pass are obtained through the second error compensator. Based on the second rolling force compensation amount and the second predicted rolling force, the predicted rolling force value for each pass is obtained; Based on the export thickness compensation amount and the predicted export thickness, the predicted export thickness value for each pass is obtained.

10. An anchor rod body, characterized in that, The anchor rod body is obtained by the intelligent production method of anchor rod body based on multiple agents as described in any one of claims 1-9.

Images