Cold heading forming machine and automatic control method thereof

Abstract

The invention relates to the technical field of cold heading machines, and provides a cold heading forming machine and an automatic control method thereof.The method comprises the steps that material characteristic parameters, equipment operation parameters and forming quality parameters in the cold heading forming process are synchronously collected, and the collected original parameter data are preprocessed; the pre-processed parameter data are analyzed in real time, the current material adaptability working condition, the equipment operation health working condition and the forming quality standard reaching working condition are recognized, and a parameter adjusting instruction is output based on the current material adaptability working condition or a fault early warning instruction is output based on the equipment operation health working condition; and based on the parameter adjusting instruction, the feeding speed of the feeding driving mechanism and the cold heading pressure, the stroke frequency and the mold closing gap of the cold heading forming driving mechanism are adjusted in real time, and closed-loop self-adaptive control over the forming process is achieved. According to the invention, accurate collaborative identification and judgment of multi-dimensional working conditions can be realized, and the product quality stability and the qualified rate can be improved.

Description

Technical Field

[0001] This invention relates to the field of cold heading machine technology, and more specifically, to a cold heading forming machine and its automatic control method. Background Technology

[0002] Cold heading machines are specialized equipment used primarily for the mass production of fasteners such as nuts and bolts. They are an important stamping processing equipment in the field of mechanical manufacturing. Their main task is to shape raw materials (such as metal wire) into specific shapes, usually stepped shapes. The working principle of cold heading machines is to shape the raw materials by mechanical force without changing their properties (such as softening and hardening treatment).

[0003] Currently, the automation control technology of existing cold heading machines has initially achieved mechanized linkage of processes such as feeding, cold heading, and discharging. It mainly relies on a programmable logic controller (PLC) as the core control unit, combined with simple sensors (such as pressure sensors and displacement sensors) to achieve basic parameter acquisition and closed-loop control. For example, Chinese utility model patent CN208787459U discloses a feeding device for a multi-station cold heading machine. The detection signal from the feeding detection sensor is input to the PLC control system, which outputs a motor start signal to automatically control the feeding length. Chinese invention patent CN117282900A discloses a high-precision cold heading machine that uses multiple electric telescopic cylinders to produce high-strength, high-hardness products. It also incorporates pressure sensors and an alarm system that automatically alarms when the machine malfunctions. An adjustment device automatically adjusts the feeding port and feeding size to ensure that the product's precision and hardness simultaneously meet the expected standards.

[0004] Existing technologies only monitor and control single parameters (such as pressure, displacement, etc.), and cannot simultaneously identify three core working conditions: material compatibility, equipment operating health status, and molding quality compliance. They lack the ability to make multi-dimensional collaborative judgments. Summary of the Invention

[0005] Based on this, in order to improve the multi-dimensional collaborative judgment capability of cold heading machines, the present invention provides a cold heading forming machine and its automatic control method, the specific technical solution of which is as follows: A cold heading forming machine includes a feeding drive mechanism and a cold heading forming drive mechanism; the cold heading forming machine further includes: The multi-dimensional parameter acquisition module is used to simultaneously acquire material characteristic parameters, equipment operating parameters, and forming quality parameters during the cold heading process. The acquired raw parameter data is preprocessed and transmitted to the edge computing module. The edge computing module is used to perform real-time analysis of the preprocessed parameter data, identify the current material compatibility conditions, equipment operating health conditions, and molding quality compliance conditions, and output parameter adjustment instructions based on the current material compatibility conditions or fault warning instructions based on the equipment operating health conditions. The main controller, based on parameter adjustment instructions, adjusts the feeding speed of the feeding drive mechanism and the cold heading pressure, stroke frequency, and mold closing clearance of the cold heading forming drive mechanism in real time to achieve closed-loop adaptive control of the forming process. Among them, the feeding drive mechanism and the cold heading drive mechanism are electrically connected to the main controller, the multi-dimensional parameter acquisition module is electrically connected to the edge computing module, and the edge computing module is bidirectionally electrically connected to the main controller.

[0006] The cold heading forming machine synchronously collects three core parameters—material characteristics, equipment operation, and forming quality—through a multi-dimensional parameter acquisition module, overcoming the shortcomings of existing technologies that only monitor a single parameter and have limited working condition identification.

[0007] Specifically, by collecting and analyzing material characteristic parameters (such as hardness and diameter deviation), the compatibility between materials and process parameters can be accurately determined, avoiding molding defects caused by batch fluctuations in materials; by real-time monitoring of equipment operating parameters (such as pressure fluctuations and vibration amplitude), potential faults such as mold wear and motor overload can be identified in advance, preventing quality problems caused by the expansion of faults; by real-time detection of molding quality parameters (such as dimensional deviations and surface defects), the quality of finished workpieces can be directly controlled, preventing the outflow of unqualified products.

[0008] In summary, the cold heading forming machine of the present invention can achieve accurate collaborative identification and judgment of multi-dimensional working conditions, which is conducive to improving product quality stability and pass rate. It solves the problem that the existing technology only monitors and controls a single parameter (such as pressure, displacement, etc.), and cannot simultaneously identify the three core working conditions of material compatibility, equipment operating health status and forming quality compliance, and lacks multi-dimensional collaborative judgment capability.

[0009] Preferably, the cold heading machine further includes: The remote operation and maintenance platform is used to receive, store, and process operating condition information, parameter adjustment information, and fault warning information from the edge computing module, enabling remote monitoring and predictive maintenance of equipment.

[0010] Preferably, the cold heading machine further includes: The IoT communication module, using 5G industrial modules or WiFi-6 communication protocol, is used to enable bidirectional data interaction between the edge computing module and the remote operation and maintenance platform.

[0011] Preferably, the edge computing module includes: The strain hardening term acquisition unit is used to acquire the real-time strain rate, plastic strain, and strain rate reference value of the material, and to acquire the strain hardening term representing the material hardening effect based on the real-time strain rate, plastic strain, and strain rate reference value. The thermal softening term acquisition unit is used to acquire the real-time temperature of the mold and the real-time temperature of the material, acquire the mold temperature difference value based on the real-time temperature of the mold and the real-time temperature of the material, and acquire the thermal softening term to represent the thermal softening effect based on the mold temperature difference. The micro-damage term acquisition unit is used to acquire the root mean square signal of acoustic emission and to acquire micro-damage terms for quantifying the degree of material damage based on the root mean square signal of acoustic emission. The adaptability condition acquisition unit is used to acquire real-time deformation resistance based on strain hardening, thermal softening and micro-damage terms, and to identify the current material adaptability condition based on the real-time deformation resistance.

[0012] Preferably, the microscopic damage acquisition unit includes: The threshold judgment subunit is used to obtain the acoustic emission saturation threshold and the acoustic emission damage threshold, and to determine whether the acoustic emission root mean square signal is greater than the acoustic emission damage threshold and less than the acoustic emission saturation threshold. The softening factor acquisition subunit is used to acquire the micro-damage softening factor, which is used to characterize the resistance softening ratio caused by micro-damage, based on the acoustic emission damage difference between the acoustic emission root mean square signal and the acoustic emission damage threshold. The micro-damage acquisition subunit is used to acquire micro-damage terms based on the micro-damage softening factor.

[0013] An automatic control method for a cold heading machine, applied to the cold heading machine, includes the following steps: Simultaneously collect material characteristic parameters, equipment operating parameters, and forming quality parameters during the cold heading process, and preprocess the collected raw parameter data; The pre-processed parameter data is analyzed in real time to identify the current material compatibility conditions, equipment operating health conditions, and molding quality compliance conditions. Based on the current material compatibility conditions, parameter adjustment instructions are output, or based on the equipment operating health conditions, fault warning instructions are output. Based on parameter adjustment commands, the feeding speed of the feeding drive mechanism and the cold heading pressure, stroke frequency, and mold closing clearance of the cold heading forming drive mechanism are adjusted in real time to achieve closed-loop adaptive control of the forming process.

[0014] Preferably, the automatic control method for the cold heading machine includes the following steps: The IoT communication module uploads the operating condition information, parameter adjustment information and fault warning information processed by the edge computing module to the remote operation and maintenance platform in real time, and at the same time receives parameter configuration instructions or model update instructions issued by the remote operation and maintenance platform. The remote operation and maintenance platform receives, stores, and processes operating condition information, parameter adjustment information, and fault warning information from the edge computing module, and issues parameter configuration instructions and model update instructions to achieve remote monitoring and predictive maintenance of equipment.

[0015] Preferably, the specific method for real-time analysis of the preprocessed parameter data includes the following steps: Obtain the real-time strain rate, plastic strain, and strain rate reference value of the material, and obtain the strain hardening term to represent the material hardening effect based on the real-time strain rate, plastic strain, and strain rate reference value. Obtain the real-time temperature of the mold and the real-time temperature of the material. Obtain the mold temperature difference value based on the real-time temperature of the mold and the real-time temperature of the material. Obtain the heat softening term to represent the heat softening effect based on the mold temperature difference value. Acquire the root mean square (RMS) acoustic emission signal, and obtain the microscopic damage term for quantifying the degree of material damage based on the RMS acoustic emission signal. The real-time deformation resistance is obtained based on the strain hardening term, thermal softening term, and micro-damage term, and the current material compatibility condition is identified based on the real-time deformation resistance.

[0016] Preferably, the specific method for obtaining microscopic damage terms includes the following steps: Obtain the acoustic emission saturation threshold and the acoustic emission damage threshold, and determine whether the root mean square signal of acoustic emission is greater than the acoustic emission damage threshold and less than the acoustic emission saturation threshold. If so, the micro-damage softening factor, used to characterize the resistance softening ratio caused by micro-damage, is obtained based on the acoustic emission damage difference between the acoustic emission root mean square signal and the acoustic emission damage threshold. The micro-damage term is obtained based on the micro-damage softening factor.

[0017] Preferably, the specific method for adjusting the output parameter includes the following steps: Obtain the total drive power of the system and the output of qualified products per unit time. Based on the total drive power of the system and the output of qualified products per unit time, obtain the energy consumption economic item used to quantify the comprehensive energy consumption of a single product. Obtain 3D point cloud data of the ideal product and 3D scanning data of the actual product. Calculate the structural similarity index based on the 3D point cloud data of the ideal product and the 3D scanning data of the actual product. Obtain the quality loss item for evaluating product quality based on the structural similarity index. Obtain the theoretical processing time per unit and the product size deviation value, and obtain the efficiency penalty item for evaluating production efficiency based on the theoretical processing time per unit and the product size deviation value. A multi-objective optimization function is constructed based on the energy consumption economic term, the quality loss term, and the efficiency penalty term, and parameter adjustment instructions are obtained based on the multi-objective optimization function. Attached Figure Description

[0018] The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0019] Figure 1 This is a schematic diagram of the overall structure of a cold heading forming machine according to one embodiment of the present invention. Figure 1 ; Figure 2 This is a schematic diagram of the overall structure of a cold heading forming machine according to one embodiment of the present invention. Figure 2 ; Figure 3 This is a functional structure diagram of an edge computing module in one embodiment of the present invention; Figure 4 This is a functional structural diagram of a micro-damage acquisition unit in one embodiment of the present invention.

[0020] Figure 5 This is a schematic diagram of the overall process of an automatic control method for a cold heading forming machine according to an embodiment of the present invention; Figure 6 This is a flowchart illustrating a specific method for real-time analysis of preprocessed parameter data in one embodiment of the present invention. Figure 7 This is a flowchart illustrating a specific method for obtaining microscopic damage terms in one embodiment of the present invention; Figure 8 This is a flowchart illustrating a specific method for output parameter adjustment commands in one embodiment of the present invention; Figure 9 This is a schematic diagram of the overall process of the automatic control method for a cold heading forming machine in another embodiment of the present invention.

[0021] Explanation of reference numerals in the attached drawings: 1. Feeding drive mechanism; 2. Cold heading drive mechanism. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to its embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of the invention.

[0023] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0025] In this invention, "first" and "second" do not represent a specific quantity or order, but are merely used to distinguish names.

[0026] Before describing the specific embodiments of the present invention, a brief introduction to the prior art will be given first.

[0027] Cold heading machines, as core equipment in the production of metal products such as fasteners and automotive parts, work by applying instantaneous high pressure to metal wire through a mold, causing the wire to undergo plastic deformation at room temperature, thus achieving rapid workpiece forming. As the manufacturing industry transforms towards intelligent and efficient production, the level of automation control in cold heading machines directly determines production efficiency, product quality stability, and equipment maintenance costs. This is especially true in large-scale, multi-specification workpiece production scenarios, where extremely high demands are placed on the accuracy, real-time performance, and adaptability of automation control.

[0028] Currently, the automation control technology of existing cold heading machines has initially achieved mechanized linkage of processes such as feeding, cold heading, and discharging. This mainly relies on a programmable logic controller (PLC) as the core control unit, combined with simple sensors (such as pressure sensors and displacement sensors) to achieve basic parameter acquisition and closed-loop control. In some existing automatic feeding control devices for cold heading machines, displacement sensors can detect the feeding position, and the PLC can adjust the feeding speed according to preset parameters to automate the feeding process. Alternatively, pressure sensors can collect cold heading pressure in real time, compare it with preset thresholds, and adjust the hydraulic system pressure to ensure stable pressure during the forming process. For example, Chinese utility model patent with authorization announcement number CN208787459U discloses a feeding device for a multi-station cold heading machine. The detection signal from the feeding detection sensor is input to the PLC control system, and the PLC control system outputs a motor start signal to automatically control the feeding length. Chinese invention patent with authorization announcement number CN117282900A discloses a high-precision cold heading machine. By designing multiple electric telescopic cylinders, it produces high-strength hardness products. It is also equipped with a pressure sensor and an alarm. When the machine malfunctions, it can automatically alarm. The automatic feeding port is adjusted by an adjustment device, and the feeding size device is adjusted so that the product's precision and hardness can simultaneously meet the expected standards.

[0029] In actual industrial production scenarios, the existing automatic control technology for cold heading machines still has many technical shortcomings, which are specifically manifested in the following aspects: First, the identification of operating conditions is singular and lacks accuracy, failing to provide multi-dimensional collaborative judgment capabilities. Existing technologies mostly monitor and control only a single parameter (such as pressure or displacement), unable to simultaneously identify three core operating conditions: material compatibility, equipment operating health status, and molding quality compliance. For example, when batch changes in wire lead to fluctuations in material hardness and diameter, existing systems cannot quickly identify material compatibility deviations and continue production according to preset parameters, easily resulting in workpiece molding defects (such as dimensional deviations and surface cracks). Simultaneously, the identification of potential equipment faults such as mold wear and motor overload is delayed, often triggering shutdowns only after a fault occurs, causing production interruptions and equipment damage.

[0030] Secondly, the control strategy is fixed and lacks adaptive adjustment and intelligent decision-making capabilities. The control parameters of existing cold heading forming machines are mostly preset manually, which can only achieve fixed process control for specific materials and workpieces of specific specifications. When the production conditions change dynamically (such as fluctuations in material properties or aging of equipment parts), the control parameters cannot be automatically optimized and adjusted. The parameters need to be calibrated by human experience, which not only results in a delayed response but also makes it easy to affect the consistency of product quality due to human operation errors. Especially in multi-specification, small-batch production scenarios, frequent manual parameter adjustments lead to a significant reduction in production efficiency.

[0031] Third, data processing and transmission modes have limitations, resulting in insufficient real-time performance and remote operation and maintenance capabilities. Current technologies largely rely on local PLCs or cloud servers for parameter data analysis and processing. If relying solely on local PLCs, their data processing capabilities are limited, making real-time analysis and AI decision-making for complex operating conditions impossible. If relying on cloud servers, data transmission latency is high (especially in industrial environments with unstable networks), easily leading to delayed control command execution and failing to meet the real-time control requirements of high-frequency cold heading (10-50 times / second). Furthermore, existing remote monitoring systems mostly only provide data visualization, lacking intelligent operation and maintenance functions such as data-driven predictive maintenance and remote model updates. Equipment maintenance remains primarily manual inspection, resulting in high maintenance costs and low fault diagnosis efficiency.

[0032] Fourth, the safety interlock mechanism is imperfect, and the fault response classification is unclear. Existing safety interlock technologies mostly only trigger a single emergency shutdown action when parameters exceed limits (such as excessive pressure or displacement), without designing a graded response mechanism based on the severity of the fault (such as minor warnings, potential faults, and sudden faults). This easily leads to problems of "excessive shutdown" (shutting down due to minor abnormalities, affecting production) or "insufficient response" (failure to shut down in time for serious faults, resulting in safety hazards). Furthermore, there is a lack of linkage between fault warnings and troubleshooting guidelines, requiring time-consuming manual troubleshooting after a fault occurs, further prolonging production interruption time.

[0033] One of the objectives of this invention is to improve the multi-dimensional collaborative judgment capability of cold heading machines, thereby achieving accurate collaborative identification and judgment of multi-dimensional working conditions and improving product quality stability and pass rate.

[0034] To achieve the above objectives, such as Figure 1 As shown, an embodiment of the present invention provides a cold heading machine, including a feeding drive mechanism 1 and a cold heading drive mechanism 2.

[0035] Specifically, the feeding drive mechanism generally includes a servo drive motor, a ball screw transmission assembly, an electromagnetic brake device, a feeding clutch, and a laser positioning calibration module. The servo drive motor is electrically connected to the pulse width modulation output terminal of the main controller. The ball screw transmission assembly is used to achieve precise transmission of feeding speed and feed amount. The electromagnetic brake device is adapted to the emergency braking requirements of the safety interlock mechanism.

[0036] The cold heading driving mechanism generally consists of a high-pressure hydraulic system, a servo punch assembly, a stroke adjustment module, a pressure feedback unit, and an emergency unloading valve. The high-pressure hydraulic system provides power output for cold heading. The servo punch assembly is electrically connected to the servo control module and is used to adjust the cold heading stroke and impact rate. The pressure feedback unit collects cold heading pressure data in real time and feeds it back to the main controller. The emergency unloading valve is adapted to the emergency pressure relief requirements of the safety interlock mechanism to ensure rapid unloading when the pressure is overloaded.

[0037] Of course, cold heading forming machines generally also include a cutting mechanism, a multi-station forming mold, a mold closing gap adjustment mechanism, a mold temperature monitoring module, and a mechanical locking device. The multi-station forming mold is used to adapt to the continuous forming requirements of workpieces of different specifications. The mold closing gap adjustment mechanism is electrically connected to the main controller to achieve precise adjustment of the gap. The mold temperature monitoring module is used to collect mold temperature data in real time. The mechanical locking device is used to adapt to the emergency locking requirements of the safety interlock mechanism to prevent mold displacement in case of failure.

[0038] Since the structures in the cold heading machine, including the feeding drive mechanism and the cold heading drive mechanism, are all conventional technologies in this field, they will not be described in detail here.

[0039] like Figure 2 As shown, the cold heading forming machine also includes a multi-dimensional parameter acquisition module, an edge computing module, and a main controller.

[0040] The multi-dimensional parameter acquisition module is used to synchronously acquire material characteristic parameters, equipment operating parameters, and forming quality parameters during the cold heading process. The acquired raw parameter data is preprocessed and transmitted to the edge computing module.

[0041] Specifically, the multi-dimensional parameter acquisition module can standardize the acquired raw parameter data and transmit it to the edge computing module in real time. The multi-dimensional parameter acquisition module includes, but is not limited to, vision sensors, pressure sensors, temperature sensors, vibration sensors, displacement sensors, and hardness detection units.

[0042] The edge computing module is used to perform real-time analysis of the preprocessed parameter data, identify the current material compatibility conditions, equipment operating health conditions, and molding quality compliance conditions, and output parameter adjustment instructions based on the current material compatibility conditions or fault warning instructions based on the equipment operating health conditions.

[0043] The edge computing module has local data caching and offline processing capabilities. It performs real-time data analysis and adaptive decision-making, calls a pre-trained working condition identification and parameter optimization model, analyzes the received standardized parameter data in real time, identifies the current material compatibility working condition, equipment operating health condition, and forming quality compliance condition, and outputs parameter adjustment instructions or fault warning instructions. The pre-trained working condition identification and parameter optimization model can be trained and generated based on historical production data of cold heading forming machine, material characteristic database, and quality defect case library, and has dynamic self-learning and update capabilities.

[0044] Specifically, after the standardized parameter data is transmitted to the edge computing module through the multi-dimensional parameter acquisition module, it first undergoes hierarchical preprocessing to remove invalid data and extract key features, including data cleaning and noise reduction, feature hierarchical extraction, and feature normalization.

[0045] In data cleaning and noise reduction, a combination algorithm of 3σ criterion and median filtering can be used to remove abnormal fluctuation data from sensors (such as instantaneous spikes from pressure sensors and blurred imaging data from vision sensors), and to complete missing data using linear interpolation. Among them, time-series parameters such as vibration and pressure are filtered using median filtering, with a window size of 5ms, which is suitable for high-frequency operation scenarios of cold heading forming at 10-50 times / second. Static parameters such as vision and hardness can be filtered using the 3σ criterion to select effective data.

[0046] In feature layering extraction, core features can be extracted according to three dimensions: material characteristics, equipment operation, and molding quality, forming a structured feature vector. Material characteristics are obtained by extracting four types of core features: mean / variance of hardness, diameter deviation, surface roughness, and material density. Equipment operation parameters are obtained by extracting five types of core features: peak / fluctuation value of cold heading pressure, effective value of main motor current, mold vibration amplitude, mold temperature, and displacement deviation of the feeding mechanism. Molding quality parameters are obtained by extracting four types of core features: workpiece head size deviation, rod straightness, number of surface defects, and hardness compliance rate.

[0047] The Min-Max normalization algorithm can be used to map the feature values ​​of each dimension to the [0,1] interval to eliminate the influence of the units and facilitate subsequent calculations.

[0048] The edge computing module can accurately identify three types of working conditions by calling a pre-trained neural network model and combining it with a historical database. The historical database includes at least multiple sets of working condition data and multiple sets of defect case data.

[0049] In material compatibility identification, the main approach is to determine the degree of compatibility between the current material characteristics and the cold heading process parameters (preset initial parameters) to avoid forming defects or equipment damage caused by material mismatch. Specifically, normalized material hardness mean / variance, diameter deviation, and surface roughness can be used as input features to a neural network. A CNN sub-model extracts the correlation between material features, and an LSTM sub-model analyzes the temporal stability of material characteristics (such as hardness fluctuation trends during continuous feeding). A compatibility score (0-100 points) is output, and a compatibility threshold is set. The operating condition is then determined based on the compatibility score and the threshold. For example, if the compatibility score is between 80 and 100, the operating condition is considered compatible, indicating stable material characteristics and a high degree of matching with the current process parameters, requiring no parameter adjustment. If the score is between 60 and 80, it is considered basically compatible, indicating slight fluctuations in material characteristics, requiring fine-tuning of process parameters. Otherwise, it is considered incompatible, indicating that the material characteristics exceed the process compatibility range, easily leading to forming defects, requiring a warning and suspension of feeding.

[0050] The identification of equipment operating health status is mainly achieved by real-time monitoring of the operating status of key components (main motor, mold, feeding mechanism) to identify potential faults in advance (such as mold wear, motor overload) and avoid sudden shutdowns. It typically uses normalized peak / fluctuation values ​​of cold heading pressure, effective values ​​of main motor current, mold vibration amplitude, mold temperature, and feeding mechanism displacement deviation as input features. First, an autoencoder detects the degree to which parameters deviate from the normal range. Then, an XGBoost classifier (with a pre-trained fault type library) locates the faulty component and type. Finally, the operating condition is determined based on abnormal values ​​of equipment operating health. For example, an abnormal value of less than 20% is considered healthy, indicating stable operating parameters of all components and no potential faults. An abnormal value between 20% and 40% is considered sub-healthy, indicating slight deviations in some parameters and early wear of components (such as slight mold wear), requiring early warning and planned maintenance. Other situations are considered faults, indicating that parameters deviate significantly from the normal range, posing a risk of sudden failure (such as motor overload, mold jamming), requiring immediate triggering of a safety response.

[0051] The core objective of the molding quality compliance condition identification is to determine in real time whether the quality of the currently molded workpiece meets the standard, thus preventing defective products from being shipped out. It uses the workpiece head size deviation, rod straightness, number of surface defects, and hardness compliance rate (after feature normalization) as input features. Based on a preset neural network algorithm, it outputs the quality compliance rate (0-100%). For example, when the quality compliance rate is between 95% and 100%, it is judged as excellent, indicating that the workpiece quality fully meets the standard, and no adjustment of process parameters is needed. When the quality compliance rate is between 85% and 95%, it is judged as qualified, indicating that the workpiece has minor defects (such as small surface scratches), and the process parameters need to be fine-tuned to improve the qualification rate. Other situations are judged as unqualified, indicating that the workpiece has serious defects (such as dimensional deviation or insufficient hardness), and the machine needs to be stopped immediately for investigation.

[0052] After identifying the current material compatibility condition, equipment health condition, and molding quality compliance condition, the compatibility score, abnormal equipment health value, and quality compliance rate can be normalized first, and then weighted and fused. Based on the confidence level of the fusion result, a multi-condition fusion decision is determined to avoid misjudgment of a single condition and eliminate the uncertainty of single condition identification.

[0053] Specifically, parameter adjustment instructions or fault warning instructions are output based on the confidence level of the fusion result. For example, if the confidence level is ≥80%, the current parameters are maintained; if the confidence level is between 60% and 80%, parameter adjustment instructions are output, adjusting the feed speed by ±0.5 r / min, cold heading pressure by ±1 MPa, and mold closing clearance by ±0.02 mm; if the confidence level is <60%, a fault warning instruction is output. Of course, different parameter adjustment instructions or fault warning instructions can be pre-mapped based on different fusion result confidence levels, and the final parameter adjustment instruction or fault warning instruction is determined based on the real-time fusion result confidence level.

[0054] Based on different levels of fault warning commands, different control strategies can be executed. For example, a level 1 warning will remind maintenance through the on-site alarm unit and push the warning information to the remote operation and maintenance platform through the IoT communication module. A level 2 warning will immediately control the feeding drive mechanism to stop feeding, the on-site alarm unit will continue to sound and light alarms, and the remote platform will push the warning information and material inspection guidance. A level 3 warning will immediately trigger the safety interlock mechanism, control the feeding drive mechanism and the cold heading drive mechanism to stop urgently, and cut off the main power supply of the equipment.

[0055] Compared to existing control methods that rely on manually preset parameters and can only adapt to single materials / specifications of workpieces, the neural network model pre-trained by the edge computing module can automatically generate optimal control parameter adjustment instructions based on dynamic operating conditions such as material characteristic fluctuations, equipment status changes, and quality feedback, without the need for manual intervention.

[0056] Based on parameter adjustment commands, the main controller adjusts the feeding speed of the feeding drive mechanism and the cold heading pressure, stroke frequency, and mold closing clearance of the cold heading forming drive mechanism in real time, realizing closed-loop adaptive control of the forming process. The feeding drive mechanism and the cold heading forming drive mechanism are electrically connected to the main controller, the multi-dimensional parameter acquisition module is electrically connected to the edge computing module, and the edge computing module is bidirectionally electrically connected to the main controller.

[0057] As a preferred technical solution, the cold heading forming machine also includes a remote operation and maintenance platform and an Internet of Things (IoT) communication module.

[0058] The remote operation and maintenance platform receives, stores, and processes operational information, parameter adjustment information, and fault warning information from the edge computing module, enabling remote equipment monitoring and predictive maintenance. The IoT communication module uses a 5G industrial module or WiFi-6 communication protocol to achieve bidirectional data interaction between the edge computing module and the remote operation and maintenance platform.

[0059] The remote operation and maintenance platform can receive equipment operating data, control parameters, and fault warning information in real time, enabling remote visual monitoring of equipment operating status. Maintenance personnel can grasp the equipment operating status without being physically present at the production site. For equipment in a sub-healthy state, the system can generate predictive maintenance reminders in advance based on operating data to avoid production interruptions caused by sudden failures. At the same time, when a failure occurs, the system can accurately locate the faulty component and push fault diagnosis guidance and emergency handling solutions, which can reduce the workload of maintenance personnel and maintenance costs.

[0060] By coordinating the design of edge computing modules and IoT communication modules, the shortcomings of existing technologies, such as insufficient data processing capabilities, limited remote monitoring functions, reliance on manual inspections for maintenance, and low efficiency in troubleshooting, can be effectively addressed. The edge computing module possesses real-time local data analysis, caching, and offline processing capabilities, avoiding latency issues caused by reliance on cloud server transmission and ensuring the real-time nature of control commands. Simultaneously, the IoT communication module (5G industrial module / WiFi-6) enables bidirectional data interaction between the device and the remote maintenance platform, breaking through the geographical limitations of traditional local monitoring.

[0061] The cold heading forming machine synchronously collects three core parameters—material characteristics, equipment operation, and forming quality—through a multi-dimensional parameter acquisition module, overcoming the shortcomings of existing technologies that only monitor a single parameter and have limited working condition identification.

[0062] Specifically, by collecting and analyzing material characteristic parameters (such as hardness and diameter deviation), the compatibility between materials and process parameters can be accurately determined, avoiding molding defects caused by batch fluctuations in materials; by real-time monitoring of equipment operating parameters (such as pressure fluctuations and vibration amplitude), potential faults such as mold wear and motor overload can be identified in advance, preventing quality problems caused by the expansion of faults; by real-time detection of molding quality parameters (such as dimensional deviations and surface defects), the quality of finished workpieces can be directly controlled, preventing the outflow of unqualified products.

[0063] In summary, the cold heading forming machine of the present invention can achieve accurate collaborative identification and judgment of multi-dimensional working conditions, which greatly improves the stability and pass rate of product quality. It solves the problem that the existing technology only monitors and controls a single parameter (such as pressure, displacement, etc.), and cannot simultaneously identify the three core working conditions of material compatibility, equipment operating health status and forming quality compliance, and lacks multi-dimensional collaborative judgment capability.

[0064] In one embodiment, such as Figure 3 As shown, the edge computing module includes a strain hardening term acquisition unit, a thermal softening term acquisition unit, a micro-damage term acquisition unit, and an adaptability condition acquisition unit.

[0065] The strain hardening term acquisition unit is used to acquire the real-time strain rate, plastic strain, and strain rate reference value of the material, and to acquire the strain hardening term representing the material hardening effect based on the real-time strain rate, plastic strain, and strain rate reference value.

[0066] Specifically, real-time strain rate is the instantaneous deformation rate of the material, the ratio of punch speed to die clearance. Plastic strain is the equivalent plastic strain, which is the amount of plastic deformation accumulated in the material during cold heading. It reflects the accumulation of dislocation density and grain refinement, and is calculated from the deformation geometry, i.e., plastic strain equals ln(initial cross-sectional area of ​​billet / instantaneous cross-sectional area). The strain rate benchmark value can be understood as the quasi-static strain rate, mainly used to make the real-time strain rate dimensionless.

[0067] For example, the strain hardening term consists of a strain hardening factor and a strain rate sensitivity factor, i.e., strain hardening term = strain hardening factor × strain rate sensitivity factor. The strain hardening factor is expressed as the product of the material strength coefficient and the nth power of the plastic strain, and is used to describe the increase in deformation resistance caused by dislocation multiplication and grain refinement due to the accumulation of plastic strain. Among them, the material strength coefficient K, i.e., the reference deformation resistance, reflects the strength of the material under standard conditions and is the core quantitative indicator of the material's inherent strength. n represents the strain hardening index, which describes the ability of plastic strain ε to strengthen the deformation resistance. The larger n is, the more significant the increase in resistance caused by strain accumulation, reflecting the strength of dislocation multiplication and interaction in the material.

[0068] The strain rate sensitivity factor is expressed as (1 + B × lg(real-time strain rate / strain rate baseline value)), which describes the strengthening effect of deformation rate on the material's resistance. Here, B represents the strain rate sensitivity index, used to describe the material's sensitivity to strain rate. When B > 0, an increase in strain rate leads to an increase in resistance; the larger the value of B, the stronger the rate sensitivity effect. It can be calibrated through material tensile testing, with a default value of 0.05.

[0069] Here, the strain rate sensitivity factor describes the hardening effect of a material during plastic deformation due to increased strain rate. Specifically, as the strain rate increases, the rate of dislocation movement within the material accelerates, leading to an increase in deformation resistance. The use of a logarithmic relationship shows that this increase is gradual, consistent with the strain rate hardening law of most metallic materials.

[0070] The thermal softening term acquisition unit is used to acquire the real-time temperature of the mold and the real-time temperature of the material, acquire the mold temperature difference value based on the real-time temperature of the mold and the real-time temperature of the material, and acquire the thermal softening term to represent the thermal softening effect based on the mold temperature difference value.

[0071] During plastic deformation, the material experiences a temperature rise due to energy dissipation, leading to softening. This can be described using an exponential function to represent the attenuation effect of temperature increase on deformation resistance. For example, the thermal softening term is expressed as exp(β × mold temperature difference). Here, β < 0 (physical constraint) represents the temperature rise softening coefficient, used to describe the degree of influence of temperature on material softening, such as dynamic recovery caused by deformation temperature rise and the reduction in resistance due to recrystallization, which can be calibrated through thermal simulation experiments.

[0072] The micro-damage term acquisition unit is used to acquire the root mean square signal of acoustic emission and to acquire micro-damage terms for quantifying the degree of material damage based on the root mean square signal of acoustic emission.

[0073] As a preferred technical solution, such as Figure 4 As shown, the micro-damage acquisition unit includes a threshold judgment subunit, a softening factor acquisition subunit, and a micro-damage acquisition subunit.

[0074] The threshold judgment subunit is used to obtain the acoustic emission saturation threshold and the acoustic emission damage threshold, and to determine whether the root mean square signal of acoustic emission is greater than the acoustic emission damage threshold and less than the acoustic emission saturation threshold; the softening factor acquisition subunit is used to obtain the micro-damage softening factor, which is used to characterize the resistance softening ratio caused by micro-damage, based on the acoustic emission damage difference between the root mean square signal of acoustic emission and the acoustic emission damage threshold; the micro-damage acquisition subunit is used to obtain the micro-damage term based on the micro-damage softening factor.

[0075] Specifically, the root mean square signal of acoustic emission can be obtained by collecting the effective value of acoustic emission signal in the 100kHz~1MHz frequency band through a piezoelectric sensor. It reflects the energy intensity of microcracks, dislocation slip, and pore initiation inside the material and is the core characterization index of micro-damage.

[0076] The micro-damage term is expressed as .in, These are represented, in order, as acoustic emission damage threshold, acoustic emission root mean square signal, and acoustic emission saturation threshold. The acoustic emission damage threshold is mainly used to distinguish the material damage signal from the critical value of equipment vibration and mold friction noise. The acoustic emission saturation threshold corresponds to the AE signal value of the material in the critical damage state, indicating that the damage softening effect no longer increases linearly after exceeding this value. This represents the damage mapping coefficient, which is between 0 and 0.9 and is used to control the growth rate of the damage factor. The usual value is 0.8-0.9.

[0077] The adaptability condition acquisition unit is used to acquire real-time deformation resistance based on strain hardening, thermal softening and micro-damage terms, and to identify the current material adaptability condition based on the real-time deformation resistance.

[0078] Traditional cold heading resistance prediction models only consider strain hardening or temperature compensation, often neglecting real-time damage feedback. This can easily lead to insufficient pressure due to internal material damage, resulting in under-forming. Here, real-time deformation resistance can be expressed as the product of strain hardening, thermal softening, and microscopic damage terms. By integrating multi-dimensional data including acoustic emission signals, temperature difference information, and strain rate, it quantifies the multi-field coupling effect of force, heat, and sound into a dynamic resistance value. This overcomes the limitations of traditional static material models and provides a benchmark for real-time process control of cold heading machines.

[0079] In one embodiment, such as Figure 5 As shown, the present invention also provides an automatic control method for a cold heading machine, applied to the aforementioned cold heading machine, comprising the following steps: S1 synchronously collects material characteristic parameters, equipment operating parameters, and forming quality parameters during the cold heading process, and preprocesses the collected raw parameter data.

[0080] Here, the material characteristic parameters include, but are not limited to, the mean / variance of material hardness, diameter deviation, and surface roughness; the equipment operating parameters include, but are not limited to, the peak / fluctuation value of cold heading pressure, the effective value of the main motor current, the mold vibration amplitude, the mold temperature, and the displacement deviation of the feeding mechanism; and the forming quality parameters include, but are not limited to, the deviation of the workpiece head size, the straightness of the rod, the number of surface defects, and the hardness compliance rate.

[0081] As a preferred technical solution, such as Figure 6As shown, the specific method for real-time analysis of preprocessed parameter data includes the following steps: S11, obtain the real-time strain rate, plastic strain and strain rate reference value of the material, and obtain the strain hardening term to represent the material hardening effect based on the real-time strain rate, plastic strain and strain rate reference value. S12, obtain the real-time temperature of the mold and the real-time temperature of the material, obtain the mold temperature difference value based on the real-time temperature of the mold and the real-time temperature of the material, and obtain the heat softening term to represent the heat softening effect based on the mold temperature difference value.

[0082] S13, acquire the root mean square signal of acoustic emission, and obtain the micro-damage term for quantifying the degree of material damage based on the root mean square signal of acoustic emission.

[0083] S14 obtains real-time deformation resistance based on strain hardening, thermal softening and micro-damage terms, and identifies the current material compatibility condition based on the real-time deformation resistance.

[0084] In step S13, as follows Figure 7 As shown, the specific method for obtaining the microscopic damage term includes the following steps: S131, obtain the acoustic emission saturation threshold and the acoustic emission damage threshold, and determine whether the acoustic emission root mean square signal is greater than the acoustic emission damage threshold and less than the acoustic emission saturation threshold.

[0085] S132, if so, then obtain the micro-damage softening factor, which is used to characterize the resistance softening ratio caused by micro-damage, based on the acoustic emission damage difference between the acoustic emission root mean square signal and the acoustic emission damage threshold.

[0086] S133, obtain the micro-damage term based on the micro-damage softening factor.

[0087] For example, real-time deformation resistance = strain hardening term × thermal softening term × micro-damage term. The strain hardening term is expressed as... ,in, These represent the real-time strain rate and the reference strain rate, respectively. The thermal softening term is expressed as... ,in, This represents the temperature difference value of the mold. The microscopic damage term is represented as... .

[0088] S2 performs real-time analysis of the pre-processed parameter data, identifies the current material compatibility conditions, equipment operating health conditions, and molding quality compliance conditions, and outputs parameter adjustment commands based on the current material compatibility conditions or fault warning commands based on the equipment operating health conditions.

[0089] As a preferred technical solution, in step S2, such as Figure 8As shown, the specific method for adjusting output parameters includes the following steps: S21, obtain the total drive power of the system and the output of qualified products per unit time, and obtain the energy consumption economic item for quantifying the comprehensive energy consumption of a single product based on the total drive power of the system and the output of qualified products per unit time.

[0090] For example, the energy consumption economic item is represented as .in, These represent the total system drive power, the output of qualified products per unit time, the heat loss coefficient, and the mold temperature rise rate, respectively. The heat loss coefficient is calculated as: mold material specific heat capacity × mold material density × effective mold volume / cooling system efficiency. The cooling system efficiency is expressed as the ratio of the actual heat absorption capacity of the cooling medium to its theoretical maximum value.

[0091] S22, acquire the 3D point cloud of the ideal product and the 3D scanning data of the actual product, calculate the structural similarity index based on the 3D point cloud of the ideal product and the 3D scanning data of the actual product, and obtain the quality loss item for evaluating product quality based on the structural similarity index.

[0092] For example, the quality loss item is represented as .in, These represent the 3D point cloud of the ideal product and the 3D scan data of the actual product, respectively. The structural similarity index is mainly used to evaluate the geometric / texture similarity between the actual product and the ideal model. Its value range is [0,1] and the optimal value is 0.

[0093] S23, obtain the theoretical processing time per unit and the product size deviation value, and obtain the efficiency penalty item for evaluating production efficiency based on the theoretical processing time per unit and the product size deviation value.

[0094] For example, .in, These represent, in order, the theoretical processing time per piece, the dimensional deviation penalty coefficient, the product dimensional deviation value, and the preset penalty constant.

[0095] The theoretical processing time for a single piece can be understood as the standard cycle of a cold heading machine completing one feeding-stamping-demolding operation under rated conditions with no dimensional deviations. The dimensional deviation penalty coefficient characterizes the amplification of the efficiency penalty caused by dimensional deviations, typically ranging from 1 to 5 mm. -1 To avoid excessive exponential penalty, the calibration can be based on the product's precision level. For example, for high-precision fasteners, γ = 3-5mm. -1 For ordinary structural components, γ = 1-3 mm −1 Product dimensional deviations can be taken as the absolute difference between the actual measured values ​​and the ideal CAD values ​​of key feature dimensions of cold-forged parts (such as bolt dimensions across sides, head height, and rod diameter). This indicates the maximum permissible dimensional deviation, which can be set according to actual conditions.

[0096] The penalty constant can be understood as a constant that is artificially set much larger than the maximum value of the efficiency penalty term under qualified operating conditions. It is used to impose extreme penalties on out-of-tolerance solutions to ensure that the algorithm does not converge to the out-of-tolerance control parameters.

[0097] For acceptable working conditions (product size deviation is not greater than the maximum allowable size deviation), the size deviation is gently penalized by an exponential term. The larger the deviation, the higher the efficiency penalty value. The algorithm prioritizes the selection of control parameters with shorter processing cycles while ensuring that the size is acceptable. For unacceptable working conditions, an extreme penalty is forced by a large constant M, excluding out-of-tolerance solutions from the Pareto optimal solution set, ensuring that the optimized output control parameters only correspond to acceptable products.

[0098] Here, by minimizing the efficiency penalty term, i.e. minimizing the efficiency penalty target, a dynamic balance between production efficiency and dimensional accuracy can be achieved.

[0099] S24. Construct a multi-objective optimization function based on the energy consumption economic term, the quality loss term, and the efficiency penalty term, and obtain parameter adjustment instructions based on the multi-objective optimization function.

[0100] After obtaining the energy consumption economic term, mass loss term, and efficiency penalty term, the energy consumption economic term, mass loss term, and efficiency penalty term are dimensionlessly normalized to [0,1] and then non-dominated sorted. Finally, the Pareto front (non-dominated solution set) is calculated, and the compromise solution with the minimum Euclidean distance is selected.

[0101] For example, the multi-objective optimization function is expressed as: .in, These represent the energy consumption economic item, the quality loss item, and the efficiency penalty item, respectively. Represents the control vector. These are represented in order as feed rate, cold heading pressure, stroke frequency, and die closing clearance.

[0102] Here, before entering the multi-objective optimization process based on the multi-objective optimization function, the current material suitability condition is first determined based on the real-time deformation resistance. If the real-time deformation resistance is greater than the preset maximum deformation resistance threshold, it is determined to be a material overhard condition, and a pressure increase and speed reduction strategy is adopted. If the real-time deformation resistance is less than the preset minimum deformation resistance threshold and the root mean square signal of acoustic emission is greater than the acoustic emission saturation threshold, it is determined to be an internal damage condition. Otherwise, the multi-objective optimization process is entered.

[0103] Generally, upper and lower limits can be set for feeding speed, cold heading pressure, stroke frequency, and die closing clearance as constraints for multi-objective optimization functions.

[0104] The upper limit of stroke frequency can be determined according to the formula. Confirmed. Among them, These are represented, in order, the upper limit of stroke frequency, the punch speed, the material ductility-flow coupling coefficient, and the effective working stroke length.

[0105] Specifically, the maximum safe stroke frequency of cold heading equipment is the hard constraint upper limit of the stroke frequency f_stroke in multi-objective optimization. Exceeding this value will result in defects such as insufficient metal filling, product cracks, and die impact overload. It is used to provide the physical feasible domain boundary for the stroke frequency in the control vector, balancing production efficiency and forming quality. The punch speed is the axial real-time linear velocity of the punch in the cold heading machine, which directly determines the basic upper limit of the stroke frequency. The faster the punch speed, the higher the theoretical maximum stroke frequency. The die closing clearance is the process clearance reserved after the upper and lower dies are closed. It is used to compensate for the fluctuation of the billet size, the thermal expansion and elastic deformation of the die, and it represents the minimum flow / deformation space after the die is closed.

[0106] The ductility-flow coupling coefficient is used to quantify the correlation between a material's plastic flowability and its resistance to deformation, reflecting the material's filling fluidity under stress. A larger k value indicates smoother material flow under the same deformation resistance. Setting the ductility-flow coupling coefficient ensures... The dimension of the term is m, and it also characterizes the equivalent compensation of material properties to deformation space. The better the material fluidity, the larger the equivalent deformation space and the higher the upper limit of stroke frequency.

[0107] Real-time deformation resistance reflects a material's ability to resist plastic deformation in real time; the higher the value, the more difficult the material flow. Here, real-time deformation resistance is used as the denominator, reflecting the engineering logic of harder materials reducing speed while maintaining quality. The unit of mold closing clearance is generally μm, and the constant 1000 is used to convert the original unit μm of mold closing clearance to m.

[0108] This formula starts from the coupling relationship between punch speed, die geometry clearance, material plastic flow capacity, and effective stroke length, and calculates the maximum safe stroke frequency of cold heading equipment under the premise of ensuring forming quality (complete filling, no underfilling / cracking). It links the mechanical properties of materials with the motion parameters of equipment to achieve dynamic constraints on efficiency and quality.

[0109] To avoid damage to the equipment due to the actual stroke frequency exceeding the equipment's mechanical capacity, the mechanical limit frequency of the equipment is superimposed, so that the final upper limit of stroke frequency = min(upper limit of stroke frequency, mechanical limit frequency of equipment).

[0110] The healthy operating condition of the equipment can be obtained by weighted fusion of one or more parameters, such as the normalized peak / fluctuation value of cold heading pressure, the effective value of the main motor current, the amplitude of mold vibration, the mold temperature, and the displacement deviation of the feeding mechanism; the forming quality compliance condition is obtained by weighted fusion of one or more parameters, such as the normalized workpiece head size deviation, the straightness of the rod, the number of surface defects, and the hardness compliance rate, or by weighted fusion of one or more parameters, such as the geometric accuracy, surface quality, and dimensional deviation of the product.

[0111] S3, based on parameter adjustment commands, adjusts in real time the feeding speed of the feeding drive mechanism and the cold heading pressure, stroke frequency, and mold closing clearance of the cold heading forming drive mechanism to achieve closed-loop adaptive control of the forming process.

[0112] Generally, closed-loop adaptive control during the molding process can be achieved based on the PID algorithm.

[0113] By collecting and analyzing material characteristic parameters (such as hardness and diameter deviation), the compatibility between materials and process parameters can be accurately determined, avoiding molding defects caused by batch fluctuations in materials. By monitoring equipment operating parameters (such as pressure fluctuations and vibration amplitude) in real time, potential faults such as mold wear and motor overload can be identified in advance, preventing quality problems caused by the expansion of faults. By detecting molding quality parameters (such as dimensional deviations and surface defects) in real time, the quality of finished workpieces can be directly controlled, preventing the outflow of unqualified products.

[0114] In summary, the automatic control method for cold heading forming machine described in this invention can achieve accurate collaborative identification and judgment of multi-dimensional working conditions, greatly improving product quality stability and pass rate. It solves the problem that existing technologies only monitor and control single parameters (such as pressure, displacement, etc.), and cannot simultaneously identify three core working conditions: material compatibility, equipment operating health status, and forming quality compliance, and lack multi-dimensional collaborative judgment capabilities.

[0115] In one embodiment, such as Figure 9 As shown, the automatic control method for the cold heading machine further includes the following steps: S4, through the IoT communication module, uploads the operating condition information, parameter adjustment information and fault warning information processed by the edge computing module to the remote operation and maintenance platform in real time, and at the same time receives parameter configuration instructions or model update instructions issued by the remote operation and maintenance platform.

[0116] S5, the remote operation and maintenance platform receives, stores, and processes the operating condition information, parameter adjustment information, and fault warning information from the edge computing module, and issues parameter configuration instructions and model update instructions to realize remote monitoring and predictive maintenance of equipment.

[0117] Specifically, based on the healthy operating conditions of the equipment, fault warning commands are output. When the confidence level of the fusion result is too low and a level three fault warning is triggered, such as mold jamming, motor overload, or serious material abnormalities leading to molding risks, an emergency stop command is output. The command can be transmitted to the main controller via the CANopen industrial bus, along with fault type codes such as "E001-Cold heading pressure exceeds threshold" and "E002-Mold vibration exceeds limit", which facilitates the main controller to accurately execute interlocking actions.

[0118] The main controller can adopt a redundant design with dual PLCs operating in parallel to avoid interlock failure due to a single PLC malfunction. After receiving a trigger signal, the main controller immediately activates the safety interlock mechanism. After the signal is verified to be correct, it starts the preset safety interlock program, simultaneously cutting off the normal operation control signals from the main controller to the feeding drive mechanism and the cold heading drive mechanism, switching to the emergency stop control channel, and then triggering the equipment's built-in emergency power supply module to ensure stable power supply to the main controller, safety interlock components, and alarm unit during the emergency stop process, preventing the shutdown operation from being interrupted due to power failure.

[0119] The main controller outputs graded stop commands to the feeding drive mechanism (generally composed of a servo motor, ball screw, electromagnetic brake, and feeding clutch) through the emergency stop control channel to ensure rapid and smooth shutdown and avoid material jamming or damage to the mechanism. The cold heading drive mechanism (generally composed of a hydraulic system, main punch, and mold opening and closing mechanism) undertakes the high-pressure forming task. Emergency stop must prioritize ensuring pressure unloading and mold locking to avoid high-pressure impact causing mold damage or workpiece splashing.

[0120] Thus, based on the industrial operation scenario of cold heading forming machines, by realizing remote equipment monitoring and predictive maintenance, safety accidents can be effectively prevented and equipment and production losses can be reduced.

[0121] The technical features of the embodiments described can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A cold heading forming machine, comprising a feeding drive mechanism and a cold heading forming drive mechanism, characterized in that, The cold heading forming machine also includes: The multi-dimensional parameter acquisition module is used to simultaneously acquire material characteristic parameters, equipment operating parameters, and forming quality parameters during the cold heading process. The acquired raw parameter data is preprocessed and transmitted to the edge computing module. The edge computing module is used to perform real-time analysis of the preprocessed parameter data, identify the current material compatibility conditions, equipment operating health conditions, and molding quality compliance conditions, and output parameter adjustment instructions based on the current material compatibility conditions or fault warning instructions based on the equipment operating health conditions. The main controller, based on parameter adjustment instructions, adjusts the feeding speed of the feeding drive mechanism and the cold heading pressure, stroke frequency, and mold closing clearance of the cold heading forming drive mechanism in real time to achieve closed-loop adaptive control of the forming process. Among them, the feeding drive mechanism and the cold heading drive mechanism are electrically connected to the main controller, the multi-dimensional parameter acquisition module is electrically connected to the edge computing module, and the edge computing module is bidirectionally electrically connected to the main controller.

2. The cold heading forming machine as described in claim 1, characterized in that, The cold heading forming machine also includes: The remote operation and maintenance platform is used to receive, store, and process operating condition information, parameter adjustment information, and fault warning information from the edge computing module, enabling remote monitoring and predictive maintenance of equipment.

3. A cold heading forming machine as described in claim 2, characterized in that, The cold heading forming machine also includes: The IoT communication module, using 5G industrial modules or WiFi-6 communication protocol, is used to enable bidirectional data interaction between the edge computing module and the remote operation and maintenance platform.

4. A cold heading forming machine as described in claim 3, characterized in that, The edge computing module includes: The strain hardening term acquisition unit is used to acquire the real-time strain rate, plastic strain, and strain rate reference value of the material, and to acquire the strain hardening term representing the material hardening effect based on the real-time strain rate, plastic strain, and strain rate reference value. The thermal softening term acquisition unit is used to acquire the real-time temperature of the mold and the real-time temperature of the material, acquire the mold temperature difference value based on the real-time temperature of the mold and the real-time temperature of the material, and acquire the thermal softening term to represent the thermal softening effect based on the mold temperature difference. The micro-damage term acquisition unit is used to acquire the root mean square signal of acoustic emission and to acquire micro-damage terms for quantifying the degree of material damage based on the root mean square signal of acoustic emission. The adaptability condition acquisition unit is used to acquire real-time deformation resistance based on strain hardening, thermal softening and micro-damage terms, and to identify the current material adaptability condition based on the real-time deformation resistance.

5. A cold heading forming machine as described in claim 4, characterized in that, The micro-damage term acquisition unit includes: The threshold judgment subunit is used to obtain the acoustic emission saturation threshold and the acoustic emission damage threshold, and to determine whether the acoustic emission root mean square signal is greater than the acoustic emission damage threshold and less than the acoustic emission saturation threshold. The softening factor acquisition subunit is used to acquire the micro-damage softening factor, which is used to characterize the resistance softening ratio caused by micro-damage, based on the acoustic emission damage difference between the acoustic emission root mean square signal and the acoustic emission damage threshold. The micro-damage acquisition subunit is used to acquire micro-damage terms based on the micro-damage softening factor.

6. An automatic control method for a cold heading forming machine, applied to the cold heading forming machine as described in any one of claims 1-5, characterized in that, The automatic control method for the cold heading machine includes the following steps: Simultaneously collect material characteristic parameters, equipment operating parameters, and forming quality parameters during the cold heading process, and preprocess the collected raw parameter data; The pre-processed parameter data is analyzed in real time to identify the current material compatibility conditions, equipment operating health conditions, and molding quality compliance conditions. Based on the current material compatibility conditions, parameter adjustment instructions are output, or based on the equipment operating health conditions, fault warning instructions are output. Based on parameter adjustment commands, the feeding speed of the feeding drive mechanism and the cold heading pressure, stroke frequency, and mold closing clearance of the cold heading forming drive mechanism are adjusted in real time to achieve closed-loop adaptive control of the forming process.

7. The automatic control method for a cold heading forming machine as described in claim 6, characterized in that, The automatic control method for the cold heading machine includes the following steps: The IoT communication module uploads the operating condition information, parameter adjustment information and fault warning information processed by the edge computing module to the remote operation and maintenance platform in real time, and at the same time receives parameter configuration instructions or model update instructions issued by the remote operation and maintenance platform. The remote operation and maintenance platform receives, stores, and processes operating condition information, parameter adjustment information, and fault warning information from the edge computing module, and issues parameter configuration instructions and model update instructions to achieve remote monitoring and predictive maintenance of equipment.

8. The automatic control method for a cold heading forming machine as described in claim 7, characterized in that, The specific methods for real-time analysis of preprocessed parameter data include the following steps: Obtain the real-time strain rate, plastic strain, and strain rate reference value of the material, and obtain the strain hardening term to represent the material hardening effect based on the real-time strain rate, plastic strain, and strain rate reference value. Obtain the real-time temperature of the mold and the real-time temperature of the material. Obtain the mold temperature difference value based on the real-time temperature of the mold and the real-time temperature of the material. Obtain the heat softening term to represent the heat softening effect based on the mold temperature difference value. Acquire the root mean square (RMS) acoustic emission signal, and obtain the microscopic damage term for quantifying the degree of material damage based on the RMS acoustic emission signal. The real-time deformation resistance is obtained based on the strain hardening term, thermal softening term, and micro-damage term, and the current material compatibility condition is identified based on the real-time deformation resistance.

9. The automatic control method for a cold heading forming machine as described in claim 8, characterized in that, The specific method for obtaining microscopic damage parameters includes the following steps: Obtain the acoustic emission saturation threshold and acoustic emission damage threshold, and determine whether the root mean square signal of acoustic emission is greater than the acoustic emission damage threshold and less than the acoustic emission saturation threshold. If so, the micro-damage softening factor, used to characterize the resistance softening ratio caused by micro-damage, is obtained based on the acoustic emission damage difference between the acoustic emission root mean square signal and the acoustic emission damage threshold. The micro-damage term is obtained based on the micro-damage softening factor.

10. The automatic control method for a cold heading forming machine as described in claim 9, characterized in that, The specific method for adjusting output parameters includes the following steps: Obtain the total drive power of the system and the output of qualified products per unit time. Based on the total drive power of the system and the output of qualified products per unit time, obtain the energy consumption economic item used to quantify the comprehensive energy consumption of a single product. Obtain 3D point cloud data of the ideal product and 3D scanning data of the actual product. Calculate the structural similarity index based on the 3D point cloud data of the ideal product and the 3D scanning data of the actual product. Obtain the quality loss item for evaluating product quality based on the structural similarity index. Obtain the theoretical processing time per unit and the product size deviation value, and obtain the efficiency penalty item for evaluating production efficiency based on the theoretical processing time per unit and the product size deviation value. A multi-objective optimization function is constructed based on the energy consumption economic term, the quality loss term, and the efficiency penalty term, and parameter adjustment instructions are obtained based on the multi-objective optimization function.

Images