Screw cold heading forming process

Through the combination of mechanical wire drawing, chemical phosphating and saponification treatment, multi-station cold heading machine and intelligent mold system, the problems of high friction resistance, many quality defects and difficult precision control in horizontal cold heading machine have been solved, and an efficient and precise screw forming process has been realized, which has improved production consistency and energy efficiency management.

CN120715159APending Publication Date: 2025-09-30YUANTUO PRECISION MACHINERY (SHANGHAI) CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202510949080.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-10
Publication Date
2025-09-30

AI Technical Summary

Technical Problem

When in use, existing horizontal cold heading machines have insufficient material surface treatment and lubrication performance, resulting in high friction resistance and rapid mold wear during the forming process; insufficient control of heat accumulation during the forming process can easily lead to quality defects such as cracks and collapsed angles; the mold system lacks intelligent adjustment capabilities and has difficulty coping with the impact of raw material fluctuations between batches or equipment micro-differences on forming accuracy; online detection and quality control methods are lagging and mostly rely on manual sampling, resulting in a high misjudgment rate and a long feedback cycle; there is a lack of a closed-loop data feedback mechanism, and process parameter updates mainly rely on manual experience, making it difficult to achieve dynamic optimization and energy efficiency management.

Method used

A composite process of mechanical wire drawing and chemical phosphating and saponification is used to treat metal wires to form a lubricating film layer. Precision control is achieved through fixed-length shearing and multi-station cold heading machines. Combined with local temperature control within the mold and an intelligent mold system, intelligent temperature control trimming and adaptive mold fine-tuning are performed. Online detection and automatic sorting are introduced, and an energy efficiency analysis and data feedback optimization mechanism is established.

Benefits of technology

Effectively reduce friction resistance, extend mold life, prevent quality defects, improve forming accuracy, ensure online detection accuracy, achieve dynamic optimization and energy efficiency management, and improve production consistency and product qualification rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120715159A_ABST
    Figure CN120715159A_ABST
Patent Text Reader

Abstract

The invention relates to the field of screw forming, and discloses a screw cold heading forming process which is characterized by comprising the following steps: step 1, wire preparation and surface pretreatment: based on the requirement of screw forming on the surface quality of raw materials, a metal wire is treated by adopting a mechanical wiredrawing and chemical phosphating and saponifying composite process, and the surface of the metal wire is treated; surface oxide skin is removed, a lubricating film layer is formed, and the forming flowability and the service life of the die in the subsequent cold heading process are effectively improved. A raw material is subjected to spheroidizing annealing treatment in a continuous spheroidizing annealing furnace, internal pearlite is converted into spherical cementite, the ductility and deformation uniformity of the material are improved, then a continuous wire drawing machine is adopted for multi-stage drawing diameter reduction, a three-stage chemical phosphating-saponification treatment process is matched, a low-friction and high-adhesion lubricating film layer is formed, and the service life of the lubricating film layer is prolonged. And a stable material basis and a lubricating environment are provided for the subsequent cold heading process, so that the problems that the friction resistance generated in the forming process of a screw material is large, and a mold is quickly worn are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of screw forming, in particular to a screw cold heading forming process. Background Art

[0002] Screws are tools that use the circular rotation of an object's inclined surface and friction to gradually fasten objects and parts. Screws are indispensable industrial necessities in daily life, such as the extremely small screws used in cameras, glasses, clocks, electronics and other products; the common screws used in televisions, electrical products, musical instruments, furniture, etc.; and the large screws used in engineering, construction, and bridges. Screws have important tasks in industry, and as long as there is industry on Earth, the function of screws will always be important. Screws have been a common invention in people's production and life for thousands of years. In terms of application areas, it is the first invention of mankind. Common forming methods for screws include forging, rolling, extrusion, stretching, cold heading, etc. Among them, the horizontal screw cold heading forming process refers to a manufacturing process that uses a horizontal cold heading machine to perform plastic deformation operations such as step-by-step upsetting, diameter reduction, punching, etc. on the screw raw material under normal temperature conditions to form it into a screw blank with a specific head shape, rod length and structural characteristics; traditional horizontal cold heading machines will cause the following problems when in use:

[0003] 1. Due to insufficient surface treatment and lubrication performance of the material, the friction resistance during the forming process is large and the mold wears quickly;

[0004] 2. Insufficient control of heat accumulation during the forming process can easily lead to quality defects such as cracks and collapsed angles, especially in complex heads or hexagonal structures;

[0005] 3. The mold system lacks intelligent adjustment capabilities, making it difficult to cope with the impact of raw material fluctuations between batches or equipment differences on forming accuracy;

[0006] 4. Online testing and quality control methods are lagging behind, relying heavily on manual spot checks, resulting in high misjudgment rates and long feedback cycles.

[0007] 5. There is a lack of a closed-loop data feedback mechanism, and process parameter updates mainly rely on manual experience, making it difficult to achieve dynamic optimization and energy efficiency management. Summary of the Invention

[0008] In response to the deficiencies in the prior art, the present invention provides a screw cold heading forming process, which solves the following problems when a horizontal cold heading machine is used: insufficient surface treatment and lubrication performance of the material lead to high friction resistance and rapid mold wear during the forming process; insufficient control of heat accumulation during the forming process easily causes quality defects such as cracks and collapsed angles, which are particularly obvious in complex heads or hexagonal structures; the mold system lacks intelligent adjustment capabilities, making it difficult to cope with the impact of fluctuations in raw materials between batches or slight equipment differences on forming accuracy; online detection and quality control methods lag behind and mostly rely on manual sampling, resulting in high misjudgment rate and long feedback cycle; there is a lack of a closed-loop data feedback mechanism, and the update of process parameters mainly relies on manual experience, making it difficult to achieve dynamic optimization and energy efficiency management.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: A screw cold heading forming process includes the following steps:

[0010] Step 1: Wire preparation and surface pretreatment

[0011] Based on the requirements of screw forming on the surface quality of raw materials, a composite process of mechanical wire drawing and chemical phosphating and saponification is used to treat the metal wire to remove the surface oxide scale and form a lubricating film layer, effectively improving the forming fluidity and die life in the subsequent cold heading process, and producing pre-treated wire;

[0012] Step 2: Cut to length

[0013] Based on the pre-processed wire, a high-speed fixed-length shearing device is used to cut the wire into small segments of the required length according to the set size through a servo control system, which is called a billet. This ensures the consistency and dimensional stability of subsequent forming and generates fixed-length billets.

[0014] Step 3: Primary upsetting (head pre-forming)

[0015] Based on the fixed-length blank, a one-station cold heading process is adopted to perform axial impact forming on one end of the blank in the die to form the basic outline of the screw head, laying the foundation for subsequent precision forming and generating a head preform blank;

[0016] Step 4: Multi-station precision cold heading

[0017] Based on the head preform, a multi-station cold heading machine is used to sequentially perform multiple deformation processes such as upsetting, diameter reduction, and punching. The shank diameter and head structure of the screw are precisely controlled. A closed mold system is used to ensure dimensional accuracy and structural integrity to generate the screw body blank.

[0018] Step 5: Intelligent temperature control trimming and forming

[0019] Based on the main blank of the screw body, in-mold local temperature control technology is used to induce micro-heating to briefly heat the key deformation area, and combined with the cold area to apply force to achieve higher-quality detail forming. It is especially suitable for preventing problems such as cracking and collapse of the head or hexagonal edge, and producing temperature-controlled formed semi-finished products;

[0020] Step 6: Adaptive mold fine-tuning correction

[0021] Based on the temperature-controlled forming of the semi-finished product, the intelligent mold system uses sensors to monitor the geometric deviation information of the semi-finished product, automatically fine-tune the mold pressing angle and closing gap, perform end finishing and transition boundary optimization, and generate a finished screw blank;

[0022] Step 7: Thread rolling

[0023] Based on the finishing screw blank, a non-cutting thread rolling process is adopted. The rod surface is plastically flowed into a thread shape by rolling the thread rolling die in the thread rolling machine, ensuring the thread engagement accuracy and surface hardening, and producing a semi-finished thread rolling screw;

[0024] Step 8: Online detection and automatic sorting

[0025] Based on the semi-finished thread-rolling screws, a high-speed image recognition system and multi-dimensional size / hardness testing equipment are used to fully inspect the screw head, thread, length, concentricity, surface defects and other indicators, and automatically eliminate defective or abnormal workpieces to produce qualified finished screws.

[0026] Step 9: Energy efficiency analysis and data feedback optimization

[0027] Based on the batch processing data of qualified screw products, the energy monitoring module and process analysis system are used to analyze and feedback the energy efficiency, mold life and process parameters of the entire production line, automatically update the forming parameter library to optimize subsequent batch production, and generate an optimized feedback parameter set.

[0028] Preferably, the step 1 of wire preparation and surface pretreatment includes the following specific steps:

[0029] Sub-step 1.1: Annealing of the original disk element

[0030] Based on the premise that cold heading has high requirements for material plasticity, medium carbon steel discs are selected as raw materials. They are first spheroidized and annealed in a continuous spheroidizing annealing furnace. The temperature is set at 720℃ and kept at this temperature for 4 hours to transform the internal pearlite into spheroidal cementite, thereby improving the material's ductility and deformation uniformity, and producing annealed discs.

[0031] Sub-step 1.2: Multi-pass wire drawing and reducing

[0032] Based on the annealed disc obtained in the previous step, a continuous wire drawing unit is used to perform three-stage wire drawing reduction in sequence, with the reduction rates controlled at 20%, 15%, and 10% respectively. Ultimately, the outer diameter of the wire is controlled within the cold heading requirement of ±0.05mm, and the grain extension direction of the material is aligned with the forming direction, which facilitates subsequent metal flow and produces drawn wire.

[0033] Sub-step 1.3: Surface degreasing and pickling

[0034] The drawn wire is fed into the surface pretreatment system, where it undergoes alkaline solution spray degreasing, water rinsing, and hydrochloric acid pickling to remove surface oil and scale, and ensure that the surface roughness is suitable for phosphating adhesion. The pickling solution concentration is maintained at 10% and the temperature is controlled at 40-50°C to produce clean and activated wire.

[0035] Sub-step 1.4: Chemical phosphate coating generation

[0036] After the cleaned and activated wire enters the phosphating tank reaction zone, a crystalline phosphate film with a thickness of 5 to 15 μm is formed by reaction in a zinc phosphate solution, which enhances the adhesion of the subsequent lubricating film. This process uses an automatic pH monitoring system to ensure process stability, maintaining it at 2.5 to 3.0 to produce phosphate-treated wire.

[0037] Sub-step 1.5: Saponification lubrication layer treatment

[0038] The phosphating wire is immersed in a sodium saponification solution to form a dry lubricating film with a thickness of about 1.0 to 1.5 μm, which can significantly reduce friction during the subsequent upsetting process and prevent direct adhesion between the material and the die. The final drying process is completed by a hot water spray drying process to avoid lubricant residue crystallization and produce saponified lubricated wire.

[0039] Sub-step 1.6: Coil Arrangement and Quality Review

[0040] The saponified lubricated wire is recycled into standard coils through an automatic coiling device, and the wire surface microcracks, lubricating film integrity and tensile yield strength are comprehensively tested using an online eddy current flaw detector + tensile testing unit to ensure that it meets the conditions for subsequent forming and use, thereby generating pre-treated wire.

[0041] Preferably, the step 2 of cutting to length comprises the following specific steps:

[0042] Sub-step 2.1: Wire introduction and tension control

[0043] Based on the pre-treated wire generated in step 1, it is introduced into the shearing module through the constant tension wire feeding device. The system adopts a servo tension controller + dual sensor closed-loop feedback loop to automatically adjust the tension range to 1.2~1.5N / mm 2, ensuring that the wire is in a constant tension state when entering the shearing zone, avoiding jumping or blocking, and generating constant tension for the wire;

[0044] Sub-step 2.2: Online laser length measurement and identification

[0045] The constant tension wire is introduced into the bidirectional laser length measurement system, and a three-axis laser displacement sensor is used to continuously scan and measure the wire's forward speed and instantaneous length. The built-in fuzzy PID control algorithm automatically determines the target shearing time point, outputs a precise displacement signal, and generates a shear positioning signal for the wire.

[0046] Sub-step 2.3: High-speed closed-loop cut-to-length

[0047] Based on the shear positioning signal wire, the shearing operation is performed by the servo punching and shearing integrated mechanism. The shearing blade is made of high-speed powder metallurgy alloy steel with a hardness of HRC 68 or above. Combined with the high-frequency response servo system, the shearing time is controlled within the range of 5-7ms. The shearing accuracy data is recorded synchronously to achieve online shearing dimensional error control within ±0.05mm, generating high-precision sheared blanks.

[0048] Sub-step 2.4: Billet conveying and alignment

[0049] The high-precision sheared blanks are transferred to a multi-channel vibration alignment mechanism, where linear vibration troughs and guide rails are used to align the direction and spacing of the blanks, facilitating their subsequent entry into the upsetting station. An image recognition system is also used to detect any anomalies such as length deviation and burrs, eliminating defective products and generating an aligned row of sheared blanks.

[0050] Sub-step 2.5: Verification and feedback of cutting result data

[0051] Based on the entire row of sheared blanks, the shear section length, end face verticality and residual stress data are collected at the shear end detection station. The data is analyzed through the industrial edge computing node, and parameter fine-tuning suggestions are sent to the shear control system to achieve continuous optimization of the shearing process and generate qualified shear control blanks.

[0052] Preferably, the step 3 of primary upsetting (head preforming) includes the following specific steps:

[0053] Sub-step 3.1: Blank positioning and clamping

[0054] Based on the fixed-length blank generated in step 2, it is accurately fed into a single-station closed mold system by a precision manipulator and an automatic feeding mechanism. Pneumatic clamping and servo displacement sensors are used to ensure the center alignment and axial stability of the blank in the mold cavity, providing a consistent initial state for subsequent impact loading, and generating a positioned and clamped blank.

[0055] Sub-step 3.2: Impact loading initial forming

[0056] The positioned and clamped blank is placed in a closed die system, and the high-speed hydraulic punch system is started to apply an axial impact load. The loading rate is controlled at 8-12 m / s to achieve preliminary upsetting of the blank head. In this stage, the main plastic deformation is concentrated at one end, and the metal is controlled to expand radially to form a primary drum-shaped structure, generating an impact-formed blank.

[0057] Sub-step 3.3: Induction of radial return flow paths

[0058] Based on the impact of the initial blank, the mold's lateral flow control groove structure is simultaneously activated during the impact holding phase, inducing the material to flow back in the radial and anti-axial directions under pressure, prompting the metal to evenly fill the concave area of ​​the mold cavity and improve the geometric symmetry of the head. This flow control structure is designed based on finite element simulation analysis results to ensure the optimal path and produce a flow-optimized blank.

[0059] Sub-step 3.4: Dynamic monitoring of the forming process

[0060] The displacement-load curve of the flow-controlled optimized billet is collected and analyzed in real time during its forming process. Multi-channel strain gauges and high-speed data acquisition cards are used to evaluate the material response characteristics at each loading moment to ensure that no overload or concentrated deformation areas occur during impact upsetting. The monitoring data is fed back to the subsequent mold adjustment and energy efficiency optimization system to generate qualified billets for forming monitoring.

[0061] Sub-step 3.5: Demolding and cleaning of the forming area

[0062] The qualified blanks after forming monitoring are removed from the mold, and a mechanical ejector + negative pressure adsorption mechanism is used in combination. During the mold empty window period, the in-mold gas purge and minimal lubrication system are used to remove residual burrs and metal dust to ensure the cleanliness of the mold and the stability of the next cycle, and to generate a head preform.

[0063] Preferably, the multi-station precision cold heading forming in step 4 includes the following specific steps:

[0064] Sub-step 4.1: Re-upsetting - Deepening of the head

[0065] Based on the head preform in step 3, it enters the first station die of the five-station cold heading machine and adopts a parallel bidirectional impact upsetting process to further compact the head geometry. The servo-controlled high-frequency impact system adjusts the stroke energy to improve the head density and shape symmetry, laying a precision foundation for complex geometric structures and generating a deep-forged head blank.

[0066] Sub-step 4.2: Reduction—Rod Size Adjustment

[0067] The deep-forging head blank is transported to the second station, where progressive diameter reduction forming technology is used. The tapered heading die and radial guide sleeve are used to precisely control the diameter reduction deformation of the rod. During this process, the in-die strain sensing unit monitors the rod stress state in real time to prevent metal cracking or center deviation caused by uneven diameter reduction, thus generating a reduced diameter rod blank.

[0068] Sub-step 4.3: Local extrusion—head detail reconstruction

[0069] Based on the reduced diameter shank blank, local reverse extrusion is performed in the third station, mainly used to reshape the detailed structure of the screw head. For cross-recessed and hexagonal cap flange screws, a closed reverse extrusion die and a floating core structure are used to precisely form the local geometric shape without increasing the total length of the blank. The die temperature is maintained at room temperature ±2°C to prevent thermal expansion errors and generate the head geometric structure blank.

[0070] Sub-step 4.4: Punching - Forming the hole at the rod end

[0071] For the head geometry blank, if it is designed as a hollow screw, the fourth station high-speed axial punching process is performed. A carbide hollow punch needle and a guide positioning mechanism are used to control the punch depth and maintain the coaxiality of the core shaft to ensure the hole diameter accuracy within ±0.02mm. For solid screws, this station is skipped and the solid rod end is retained to generate a hole or solid rod blank.

[0072] Sub-step 4.5: Rolling preform

[0073] For holed or solid rod blanks with knurling / toothed ring structures, local rolling pre-forming is performed at the fifth station according to structural requirements. It is mainly used to form special shapes such as anti-slip patterns, embossing or toothed rings on the rod. A dual-die counter-pressing synchronous rolling structure is adopted, and the die speed is synchronized through a constant speed controller to prevent pattern misalignment or knurling deformation, thereby generating complex geometric screw blanks.

[0074] Sub-step 4.6: Multi-point stress monitoring and closed-loop forming control

[0075] During the five-station cold heading process, multi-point force sensors and visual feedback modules are introduced throughout the process for complex geometric screw blanks. The impact force curves and metal flow images of each station are collected. Combined with the forming simulation feedback algorithm, the die gap, punch stroke and lubrication parameters are automatically adjusted to achieve closed-loop optimization control at each stage, improve forming consistency and die service life, and generate the main screw blank.

[0076] Preferably, the step 5 of intelligent temperature-controlled trimming and forming includes the following specific steps:

[0077] Sub-step 5.1: Target area identification and process parameter setting

[0078] Based on the screw body blank obtained in the previous step, a digital image recognition system combined with a 3D geometric comparison algorithm is used to automatically identify crack-prone locations such as the head edge and the transition zone of the shank. Based on the recognition results, the system sets the local heating target area, heating rate, and target temperature, providing a parameter basis for the next step of precise temperature control and generating a local heating area parameter set.

[0079] Sub-step 5.2: Initialize the in-mold induction heating device

[0080] Based on the local heating area parameter set, a high-frequency induction coil array is arranged in the mold, and heating preparation is initiated through a closed-loop temperature control system. The system uses a dual-channel feedback mechanism of thermocouple + infrared temperature measurement to ensure that the heating process is controlled within the target set value range of 60℃ to 120℃, ensuring heating uniformity and preventing overheating, thereby generating controllable heated screw blanks.

[0081] Sub-step 5.3: Temperature difference control die synchronization execution

[0082] The controllably heated screw blank is fed into a cold precision mold, where the temperature difference induces metal micro-flow to reconstruct the lattice. Synchronous stamping technology and self-lubricating mold cavity materials are used to complete edge trimming, shape fine-tuning, and head micro-crack relief. This process utilizes heat-cold coupling to enhance the microstructure density and produce a temperature-differential trimmed semi-finished product.

[0083] Sub-step 5.4: Edge integrity and density testing

[0084] Based on temperature differential trimming of semi-finished products, we introduce microscopic visual inspection and ultrasonic local density analysis technology to automatically determine the integrity of the formed edge, the recovery effect of material microcracks, and the degree of tissue densification. Products that do not meet critical standards are automatically withdrawn from the production line through an automatic return system, ensuring that only high-quality semi-finished products enter the next step, resulting in semi-finished products with qualified edge quality.

[0085] Sub-step 5.5: Intelligent Data Storage and Traceability Coding

[0086] Semi-finished products that meet edge quality standards are transferred to the online laser marking system, where each workpiece is coded with its temperature control parameters, die pressure curve, and density level to achieve product traceability. The coded data is then uploaded to the Manufacturing Execution System (MES) to provide a basis for subsequent mold wear and repair analysis, generating traceable semi-finished products.

[0087] Sub-step 5.6: Stable output and production line buffer management

[0088] The semi-finished products with traceability marks are collected in an automatic cache unit, and the feeding and discharging rhythm is controlled by batch and partition storage logic to provide stable input for the next step of mold fine-tuning. The system simultaneously records the temperature control stability and forming integrity of this batch of products as a basis for subsequent adjustments, and generates temperature-controlled formed semi-finished products.

[0089] Preferably, the step 6 of self-adapting the mold fine-tuning and correction includes the following specific steps:

[0090] Sub-step 6.1: Pre-processing of contour error detection

[0091] Based on the temperature-controlled forming of semi-finished products, a high-precision multi-point laser displacement sensor array is used in conjunction with a rotating platform system to perform 360-degree scanning and modeling of the workpiece. An edge recognition algorithm and a contour extraction program are used to extract data on the geometric contour ellipticity, end face deflection, and length deviation of the blank in real time to generate a blank geometric error data model.

[0092] Sub-step 6.2: Intelligent identification and analysis of error characteristics

[0093] Based on the blank geometric error data model, the neural network error analysis module automatically classifies and identifies the types and degrees of various forming deviations, outputs a set of optimal correction solution recommendations, including core parameters such as mold gap adjustment amount and press angle variation range, and generates a set of intelligent correction parameter recommendations;

[0094] Sub-step 6.3: In-mold micro-drive pre-adjustment settings

[0095] The intelligent correction parameter recommendation set is imported into the mold micro-drive control unit. Based on the piezoelectric drive + closed-loop displacement feedback system, the system automatically sets the initial state of each micro-adjustment mechanism, the initial state of the guide column micro-preload, the wedge adjustment plate angle, and the micro-inclination angle of the pressing surface, ensuring that all key parts are in the correction preparation state and generating the set state mold system.

[0096] Sub-step 6.4: Fine-tune the calibration operation execution

[0097] The set state mold system and the temperature-controlled formed semi-finished product are synchronously loaded into the working cavity, and the in-mold micro-drive profiling system is activated. According to the pre-correction parameters, micron-level profiling compensation operations are performed, including end face flattening, centering and rounding, and length adjustment. Multiple simultaneous operations ensure the geometric consistency of the forming and generate a primary correction screw blank.

[0098] Sub-step 6.5: Feedback refinement and adaptive adjustment

[0099] Based on the primary correction screw blank, the laser profile is scanned again, and the corrected geometric shape is matched and analyzed with the standard template. If there is a slight residual error, the adaptive closed-loop control logic is triggered to perform a secondary micro-compensation operation on the mold drive unit to complete the final correction process and generate a precision correction screw blank.

[0100] Sub-step 6.6: Workpiece unloading and result confirmation

[0101] The precision correction screw blank is unloaded from the mold system and enters the high-precision contact measuring tool and three-dimensional vision-assisted inspection system to conduct final confirmation of the key dimensions after correction, the coaxiality of the head, the rod length, and the symmetry dimensions to ensure that it meets the requirements of the next thread rolling process and generates a fine-tuned screw blank.

[0102] Preferably, the thread rolling step 7 includes the following specific steps:

[0103] Sub-step 7.1: Cleaning and positioning the blank before rolling

[0104] Based on the fine-finished screw blanks obtained in step 6, the workpieces are fed one by one into the front station of the thread rolling equipment through a vibrating feeding system and an automatic arrangement track. Alkaline ultrasonic cleaning and hot air drying devices are used to quickly clean their surfaces to remove residual oil film and micro-dust, ensuring the cleanliness of the interface for thread rolling and forming, and generating positioned clean screw blanks.

[0105] Sub-step 7.2: Thread rolling die preload and profile adjustment

[0106] The positioning clean screw blank is loaded into the die system of the constant speed rolling thread rolling machine. According to different thread specifications, high hardness tungsten-cobalt alloy thread rolling dies are selected. The tooth profile engagement angle and lead are preset through the die angle fine-tuning mechanism. The coaxiality error between the blank center and the die engagement line is achieved by using the visual calibration system to ≤0.02mm, generating a preloaded blank assembly.

[0107] Sub-step 7.3: Constant speed thread rolling

[0108] Based on the preloaded blank assembly, the main drive system of the thread rolling machine is activated, maintaining the die linear speed in a constant range of 20 to 30 m / min. Through a two-die intermeshing rolling method or a dynamic and static meshing rolling method, the material surface induces plastic flow mechanism to directly press out the thread profile without cutting. During this process, the thread rolling pressure closed-loop control system monitors and adjusts the die gap in real time to ≤±0.01mm, ensuring consistency between pitch and thread height, and producing a pre-rolled screw.

[0109] Sub-step 7.4: Thread profile accuracy detection and fine-tuning compensation

[0110] Online laser 3D profile scanning is performed on the initially rolled screws to extract data such as thread angle, pitch, and top circle runout. A self-learning thread deviation analysis model is then used to determine if there is any deviation from the set standard. If a deviation is detected, the automatic die angle compensation module is triggered to perform dynamic adjustments at the micron level to ensure the final thread profile is accurate and produce a precision threaded screw.

[0111] Sub-step 7.5: Surface Hardening and Residual Stress Adjustment

[0112] For precision threaded screws, choose whether to perform rolling-induced strengthening treatment based on product strength requirements. This treatment applies secondary micro-strain strengthening to the threaded area by continuing the low-speed rolling process, inducing dense dislocations and residual compressive stress in the surface layer to improve fatigue resistance. If not performed, this step is skipped to produce a surface-strengthened thread semi-finished product.

[0113] Sub-step 7.6: Finished product collection and initial inspection and classification

[0114] The surface-strengthened threaded semi-finished products are placed on the automatic sorting track and pre-classified and identified through a simple contact-type tooth profile inspection and an optical sensor rapid grading device. Workpieces with obvious tooth deviation or tooth bottom damage are removed in advance, and qualified parts are sorted into the next station for online inspection to produce thread-rolling screw semi-finished products.

[0115] Preferably, the online detection and automatic sorting in step 8 includes the following specific steps:

[0116] Sub-step 8.1: Screw Arrangement and Stable Delivery

[0117] Based on the semi-finished thread-rolling screws obtained in the previous step, a vibrating plate loading system and linear guide mechanism are used to arrange the screws one by one, ensuring that the screws enter the inspection area in a uniform direction. The synchronous photoelectric sensor and servo drive control system adjust the conveying speed and spacing in real time to avoid workpiece stacking and jitter, ensure image clarity, and generate a directional screw conveying flow.

[0118] Sub-step 8.2: Multi-angle industrial visual recognition inspection

[0119] Based on the directional conveying of screw flow, a multi-angle industrial camera system configured on the four sides and top uses structured light projection + high-speed shutter imaging technology to capture images of the screw's key dimensions, including head profile, thread geometry, shaft length, and coaxiality. An image stitching and reconstruction algorithm is then introduced to perform 3D appearance modeling and analysis, generating a multi-dimensional structural inspection dataset.

[0120] Sub-step 8.3: Laser scanning and surface defect identification

[0121] The multi-dimensional structural inspection data set is transmitted to the downstream high-resolution laser scanning system, which combines it with a screw rotation mechanism to perform 360° scanning without blind spots. The system uses a speckle interferometry algorithm and an edge reflection contrast model to identify tiny surface defects such as cracks, bumps, burrs, and missing teeth. The system has a high-precision feature with a false detection rate of less than 0.5%, generating a surface defect identification information set.

[0122] Sub-step 8.4: AI intelligent classification and judgment

[0123] Based on the surface defect recognition information set, the pre-trained deep convolutional neural network (CNN) model is called to input various detection features of the screws, including the screw size deviation, defect type, and distribution location. The system outputs whether the screws are qualified or not, and whether they are defective. The system uses an adaptive threshold learning mechanism to adjust the standard tolerance based on historical batches, effectively improving the judgment accuracy and generating a classification judgment label set.

[0124] Sub-step 8.5: Automatic sorting and rejection of defective products

[0125] The generated classification label set is synchronously sent to the sorting control module. This module links the pneumatic robot arm and the track diversion mechanism to ensure that qualified products are directed into the packaging section, and unqualified products are automatically removed and sent to the waste recycling channel. Under the guarantee of 100% online detection, this process can stably maintain an effective sorting accuracy rate of more than 99.5%, and generate a set of sorted screw products;

[0126] Sub-step 8.6: Archiving and Traceability Coding of Test Data

[0127] Based on the set of sorted screw products, the image, size, and defect judgment information of each product are used to generate a unique data fingerprint file, and the traceability code is directly engraved on the screw head or outer packaging through laser marking equipment or QR code inkjet printer to achieve full-process data tracking and batch traceability of the product, and generate qualified screw products.

[0128] Preferably, the energy efficiency analysis and data feedback optimization in step 9 includes the following specific steps:

[0129] Sub-step 9.1: Data collection of the entire cold heading process

[0130] Based on the complete manufacturing process of qualified screw products, a multi-channel industrial data acquisition system (SCADA) was deployed. This system integrated energy consumption sensors from the equipment PLC, die wear monitoring modules, and lubrication flow meter terminals to form a multi-dimensional production parameter database. This database provided the basis for subsequent modeling and generated a multi-source dataset for the cold heading manufacturing process.

[0131] Sub-step 9.2: Energy Efficiency Assessment and Modeling Analysis

[0132] Based on a multi-source dataset of the cold heading manufacturing process, a corresponding line energy consumption assessment model was established through a digital twin modeling platform. The model integrates the equipment operation cycle energy efficiency coefficient, the die life decay curve, and the lubricant usage efficiency model. Multivariate regression analysis and clustering algorithms are used to identify energy efficiency bottlenecks and redundant links, and an energy efficiency assessment report for the cold heading line is generated.

[0133] Sub-step 9.3: Optimization calculation of mold and lubrication cycle

[0134] Based on the cold heading line energy efficiency assessment report, the die attenuation curve and lubrication usage model were used. Parameter sensitivity analysis and Bayesian optimization algorithms were used to calculate the optimal die replacement cycle and lubricant injection strategy. This approach achieved a balance between minimizing unit energy consumption and maximizing die utilization, generating the optimal die lubrication strategy set.

[0135] Sub-step 9.4: Generate dynamic device parameter adjustment suggestions

[0136] Combined with the optimal mold lubrication strategy set, the process feedback rule engine generates specific equipment-level parameter adjustment recommendations, including key parameter settings for impact frequency, slide travel speed, and cooling system on-time. By establishing a mapping table with the actual equipment interface, instruction-level issuance is standardized and a table of optimized equipment setting parameters is generated.

[0137] Sub-step 9.5: Process database update and self-learning mechanism activation

[0138] Uploading the equipment optimization parameter table to the process parameter database of the Manufacturing Execution System (MES) triggers the self-learning process feedback module, which adjusts its internal parameter weighting coefficients based on historical batch performance, generates batch-by-batch optimization suggestions for quality and accuracy, and generates a dynamic optimization process parameter model.

[0139] Sub-step 9.6: Automatically update and push the next batch of setting values

[0140] Based on the dynamic optimization process parameter model, the MES system automatically pushes updated forming pressure settings, lubrication cycle control, mold calibration warning values ​​and other parameters to the next batch of screw cold heading tasks, realizing closed-loop optimization of the entire process; after the push is confirmed by the operator, it is formally introduced into production to generate an optimized feedback parameter set.

[0141] The present invention provides a screw cold heading forming process. It has the following beneficial effects:

[0142] 1. The present invention performs spheroidizing annealing treatment on the raw materials in a continuous spheroidizing annealing furnace to convert the internal pearlite into spheroidal cementite, thereby improving the material's ductility and deformation uniformity. A continuous wire drawing machine is then used for multi-stage drawing and diameter reduction, combined with a three-stage chemical phosphating-saponification treatment process to form a low-friction, high-adhesion lubricating film layer, providing a stable material foundation and lubrication environment for the subsequent cold heading process, thereby solving the problems of high friction resistance and rapid mold wear generated during the screw material forming process.

[0143] 2. The present invention places the obtained fixed-length blank in the first-station cold heading die system, utilizes an open-and-closed die structure in conjunction with a preloaded punch, to achieve axial impact plastic deformation and preliminarily form the outer contour of the screw head; uses finite element analysis to simulate and optimize the impact force path to reduce the risk of eccentricity or local collapse, and then introduces the head preform into a multi-station cold heading machine, generally with four or six stations, and uses a multi-station cold heading machine to sequentially perform multiple deformation processes such as upsetting, diameter reduction, and punching, and precisely control the rod diameter and head structure of the screw, thereby solving the problem of insufficient control of heat accumulation during the forming process, which easily causes quality defects such as cracks and collapsed angles.

[0144] 3. The present invention introduces a local induction heating module into the mold cavity through the obtained screw blank to perform short-term heating on the key deformation areas of the head, such as the cross slot and hexagonal edge, and combines the cold mold force application method to achieve plastic finishing of the hot soft zone, effectively preventing defects such as cracking and wrinkling; the heating time and temperature are automatically adjusted by the in-mold sensor + PID control algorithm to ensure process stability, and then it is sent to the intelligent mold system equipped with an electric micro-displacement actuator and a mold cavity stress sensing unit to monitor the geometric error and residual stress distribution in real time. Through the mold geometry fine-tuning feedback mechanism, the end face is smoothed, the edge contour is trimmed, and the shape and position accuracy is improved, thereby solving the problem that the mold system lacks intelligent adjustment capabilities and is difficult to cope with the impact of raw material fluctuations between batches or equipment micro-differences on forming accuracy.

[0145] 4. The present invention uses an industrial high-speed image recognition system and a laser measuring device to conduct 100% appearance and geometric dimension inspection of products, including thread integrity, head symmetry, length error, etc., and cooperates with an automatic sorting mechanism to eliminate defective products to ensure the qualified rate of incoming products, thereby solving the problems of lagging online detection and quality control methods, relying heavily on manual sampling, high misjudgment rate and long feedback cycle.

[0146] 5. The present invention uploads the qualified screw product data generated during the entire production process to the MES system and energy consumption monitoring platform, and analyzes indicators such as mold life consumption, energy consumption distribution, and finished product yield trend through data mining and variation source identification algorithms, automatically updates the parameter library of each process, forms a feedback control closed loop, and provides an optimization basis for subsequent batch production, thereby solving the problem of lack of a closed-loop data feedback mechanism, and the process parameter update mainly relies on manual experience, making it difficult to achieve dynamic optimization and energy efficiency management. BRIEF DESCRIPTION OF THE DRAWINGS

[0147] Figure 1 Schematic diagram of the process steps of the present invention;

[0148] Figure 2 This is a framework diagram of the wire preparation and surface pretreatment steps of the present invention;

[0149] Figure 3 This is a framework diagram of the fixed-length cutting steps of the present invention;

[0150] Figure 4 This is a framework diagram of the primary upsetting (head preforming) step of the present invention;

[0151] Figure 5 This is a framework diagram of the multi-station precision cold heading forming steps of the present invention;

[0152] Figure 6 This is a framework diagram of the intelligent temperature-controlled trimming and forming steps of the present invention;

[0153] Figure 7 This is a framework diagram of the adaptive mold fine-tuning and correction steps of the present invention;

[0154] Figure 8 It is a framework diagram of the thread rolling steps of the present invention;

[0155] Figure 9 This is a framework diagram of the online detection and automatic sorting steps of the present invention;

[0156] Figure 10 This is a framework diagram of the energy efficiency analysis and data feedback optimization steps of the present invention. DETAILED DESCRIPTION

[0157] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0158] Example:

[0159] Please see the attached Figure 1 -Attached Figure 10 The embodiment of the present invention provides a screw cold heading forming process, which is characterized by comprising the following steps:

[0160] Step 1: Wire preparation and surface pretreatment

[0161] Based on the requirements of screw forming on the surface quality of raw materials, a composite process of mechanical wire drawing and chemical phosphating and saponification is used to treat the metal wire to remove the surface oxide scale and form a lubricating film layer, effectively improving the forming fluidity and die life in the subsequent cold heading process, and producing pre-treated wire;

[0162] Step 2: Cut to length

[0163] Based on the pre-processed wire, a high-speed fixed-length shearing device is used to cut the wire into small segments of the required length according to the set size through a servo control system, which is called a billet. This ensures the consistency and dimensional stability of subsequent forming and generates fixed-length billets.

[0164] Step 3: Primary upsetting (head pre-forming)

[0165] Based on the fixed-length blank, a one-station cold heading process is adopted to perform axial impact forming on one end of the blank in the die to form the basic outline of the screw head, laying the foundation for subsequent precision forming and generating a head preform blank;

[0166] Step 4: Multi-station precision cold heading

[0167] Based on the head preform, a multi-station cold heading machine is used to sequentially perform multiple deformation processes such as upsetting, diameter reduction, and punching. The shank diameter and head structure of the screw are precisely controlled. A closed mold system is used to ensure dimensional accuracy and structural integrity to generate the screw body blank.

[0168] Step 5: Intelligent temperature control trimming and forming

[0169] Based on the main blank of the screw body, in-mold local temperature control technology is used to induce micro-heating to briefly heat the key deformation area, and combined with the cold area to apply force to achieve higher-quality detail forming. It is especially suitable for preventing problems such as cracking and collapse of the head or hexagonal edge, and producing temperature-controlled formed semi-finished products;

[0170] Step 6: Adaptive mold fine-tuning correction

[0171] Based on the temperature-controlled forming of the semi-finished product, the intelligent mold system uses sensors to monitor the geometric deviation information of the semi-finished product, automatically fine-tune the mold pressing angle and closing gap, perform end finishing and transition boundary optimization, and generate a finished screw blank;

[0172] Step 7: Thread rolling

[0173] Based on the finishing screw blank, a non-cutting thread rolling process is adopted. The rod surface is plastically flowed into a thread shape by rolling the thread rolling die in the thread rolling machine, ensuring the thread engagement accuracy and surface hardening, and producing a semi-finished thread rolling screw;

[0174] Step 8: Online detection and automatic sorting

[0175] Based on the semi-finished thread-rolling screws, a high-speed image recognition system and multi-dimensional size / hardness testing equipment are used to fully inspect the screw head, thread, length, concentricity, surface defects and other indicators, and automatically eliminate defective or abnormal workpieces to produce qualified finished screws.

[0176] Step 9: Energy efficiency analysis and data feedback optimization

[0177] Based on the batch processing data of qualified screw products, the energy monitoring module and process analysis system are used to analyze and feedback the energy efficiency, mold life and process parameters of the entire production line, automatically update the forming parameter library to optimize subsequent batch production, and generate an optimized feedback parameter set.

[0178] Among them, step 1 wire preparation and surface pretreatment includes the following specific steps:

[0179] Sub-step 1.1: Annealing of the original disk element

[0180] Based on the premise that cold heading has high requirements for material plasticity, medium carbon steel discs are selected as raw materials. They are first spheroidized and annealed in a continuous spheroidizing annealing furnace. The temperature is set at 720℃ and kept at this temperature for 4 hours to transform the internal pearlite into spheroidal cementite, thereby improving the material's ductility and deformation uniformity, and producing annealed discs.

[0181] Sub-step 1.2: Multi-pass wire drawing and reducing

[0182] Based on the annealed disc obtained in the previous step, a continuous wire drawing unit is used to perform three-stage wire drawing reduction in sequence, with the reduction rates controlled at 20%, 15%, and 10% respectively. Ultimately, the outer diameter of the wire is controlled within the cold heading requirement of ±0.05mm, and the grain extension direction of the material is aligned with the forming direction, which facilitates subsequent metal flow and produces drawn wire.

[0183] Sub-step 1.3: Surface degreasing and pickling

[0184] The drawn wire is fed into the surface pretreatment system, where it undergoes alkaline solution spray degreasing, water rinsing, and hydrochloric acid pickling to remove surface oil and scale, and ensure that the surface roughness is suitable for phosphating adhesion. The pickling solution concentration is maintained at 10% and the temperature is controlled at 40-50°C to produce clean and activated wire.

[0185] Sub-step 1.4: Chemical phosphate coating generation

[0186] After the cleaned and activated wire enters the phosphating tank reaction zone, a crystalline phosphate film with a thickness of 5 to 15 μm is formed by reaction in a zinc phosphate solution, which enhances the adhesion of the subsequent lubricating film. This process uses an automatic pH monitoring system to ensure process stability, maintaining it at 2.5 to 3.0 to produce phosphate-treated wire.

[0187] Sub-step 1.5: Saponification lubrication layer treatment

[0188] The phosphating wire is immersed in a sodium saponification solution to form a dry lubricating film with a thickness of about 1.0 to 1.5 μm, which can significantly reduce friction during the subsequent upsetting process and prevent direct adhesion between the material and the die. The final drying process is completed by a hot water spray drying process to avoid lubricant residue crystallization and produce saponified lubricated wire.

[0189] Sub-step 1.6: Coil Arrangement and Quality Review

[0190] The saponified lubricated wire is recycled into standard coils through an automatic coiling device, and the wire surface microcracks, lubricating film integrity and tensile yield strength are comprehensively tested using an online eddy current flaw detector + tensile testing unit to ensure that it meets the conditions for subsequent forming and use, thereby generating pre-treated wire.

[0191] Among them, step 2 of cutting to length includes the following specific steps:

[0192] Sub-step 2.1: Wire introduction and tension control

[0193] Based on the pre-treated wire generated in step 1, it is introduced into the shearing module through the constant tension wire feeding device. The system adopts a servo tension controller + dual sensor closed-loop feedback loop to automatically adjust the tension range to 1.2~1.5N / mm 2 , ensuring that the wire is in a constant tension state when entering the shearing zone, avoiding jumping or blocking, and generating constant tension for the wire;

[0194] Sub-step 2.2: Online laser length measurement and identification

[0195] The constant tension wire is introduced into the bidirectional laser length measurement system, and a three-axis laser displacement sensor is used to continuously scan and measure the wire's forward speed and instantaneous length. The built-in fuzzy PID control algorithm automatically determines the target shearing time point, outputs a precise displacement signal, and generates a shear positioning signal for the wire.

[0196] Sub-step 2.3: High-speed closed-loop cut-to-length

[0197] Based on the shear positioning signal wire, the shearing operation is performed by the servo punching and shearing integrated mechanism. The shearing blade is made of high-speed powder metallurgy alloy steel with a hardness of HRC 68 or above. Combined with the high-frequency response servo system, the shearing time is controlled within the range of 5-7ms. The shearing accuracy data is recorded synchronously to achieve online shearing dimensional error control within ±0.05mm, generating high-precision sheared blanks.

[0198] Sub-step 2.4: Billet conveying and alignment

[0199] The high-precision sheared blanks are transferred to a multi-channel vibration alignment mechanism, where linear vibration troughs and guide rails are used to align the direction and spacing of the blanks, facilitating their subsequent entry into the upsetting station. An image recognition system is also used to detect any anomalies such as length deviation and burrs, eliminating defective products and generating an aligned row of sheared blanks.

[0200] Sub-step 2.5: Verification and feedback of cutting result data

[0201] Based on the entire row of sheared blanks, the shear section length, end face verticality and residual stress data are collected at the shear end detection station. The data is analyzed through the industrial edge computing node, and parameter fine-tuning suggestions are sent to the shear control system to achieve continuous optimization of the shearing process and generate qualified shear control blanks.

[0202] Among them, step 3 primary upsetting (head preforming) includes the following specific steps:

[0203] Sub-step 3.1: Blank positioning and clamping

[0204] Based on the fixed-length blank generated in step 2, it is accurately fed into a single-station closed mold system by a precision manipulator and an automatic feeding mechanism. Pneumatic clamping and servo displacement sensors are used to ensure the center alignment and axial stability of the blank in the mold cavity, providing a consistent initial state for subsequent impact loading, and generating a positioned and clamped blank.

[0205] Sub-step 3.2: Impact loading initial forming

[0206] The positioned and clamped blank is placed in a closed die system, and the high-speed hydraulic punch system is started to apply an axial impact load. The loading rate is controlled at 8-12 m / s to achieve preliminary upsetting of the blank head. In this stage, the main plastic deformation is concentrated at one end, and the metal is controlled to expand radially to form a primary drum-shaped structure, generating an impact-formed blank.

[0207] Sub-step 3.3: Induction of radial return flow paths

[0208] Based on the impact of the initial blank, the mold's lateral flow control groove structure is simultaneously activated during the impact holding phase, inducing the material to flow back in the radial and anti-axial directions under pressure, prompting the metal to evenly fill the concave area of ​​the mold cavity and improve the geometric symmetry of the head. This flow control structure is designed based on finite element simulation analysis results to ensure the optimal path and produce a flow-optimized blank.

[0209] Sub-step 3.4: Dynamic monitoring of the forming process

[0210] The displacement-load curve of the flow-controlled optimized billet is collected and analyzed in real time during its forming process. Multi-channel strain gauges and high-speed data acquisition cards are used to evaluate the material response characteristics at each loading moment to ensure that no overload or concentrated deformation areas occur during impact upsetting. The monitoring data is fed back to the subsequent mold adjustment and energy efficiency optimization system to generate qualified billets for forming monitoring.

[0211] Sub-step 3.5: Demolding and cleaning of the forming area

[0212] The qualified blanks after forming monitoring are removed from the mold, and a mechanical ejector + negative pressure adsorption mechanism is used in combination. During the mold empty window period, the in-mold gas purge and minimal lubrication system are used to remove residual burrs and metal dust to ensure the cleanliness of the mold and the stability of the next cycle, and to generate a head preform.

[0213] Among them, step 4 multi-station precision cold heading forming includes the following specific steps:

[0214] Sub-step 4.1: Re-upsetting - Deepening of the head

[0215] Based on the head preform in step 3, it enters the first station die of the five-station cold heading machine and adopts a parallel bidirectional impact upsetting process to further compact the head geometry. The servo-controlled high-frequency impact system adjusts the stroke energy to improve the head density and shape symmetry, laying a precision foundation for complex geometric structures and generating a deep-forged head blank.

[0216] Sub-step 4.2: Reduction—Rod Size Adjustment

[0217] The deep-forging head blank is transported to the second station, where progressive diameter reduction forming technology is used. The tapered heading die and radial guide sleeve are used to precisely control the diameter reduction deformation of the rod. During this process, the in-die strain sensing unit monitors the rod stress state in real time to prevent metal cracking or center deviation caused by uneven diameter reduction, thus generating a reduced diameter rod blank.

[0218] Sub-step 4.3: Local extrusion—head detail reconstruction

[0219] Based on the reduced diameter shank blank, local reverse extrusion is performed in the third station, mainly used to reshape the detailed structure of the screw head. For cross-recessed and hexagonal cap flange screws, a closed reverse extrusion die and a floating core structure are used to precisely form the local geometric shape without increasing the total length of the blank. The die temperature is maintained at room temperature ±2°C to prevent thermal expansion errors and generate the head geometric structure blank.

[0220] Sub-step 4.4: Punching - Forming the hole at the rod end

[0221] For the head geometry blank, if it is designed as a hollow screw, the fourth station high-speed axial punching process is performed. A carbide hollow punch needle and a guide positioning mechanism are used to control the punch depth and maintain the coaxiality of the core shaft to ensure the hole diameter accuracy within ±0.02mm. For solid screws, this station is skipped and the solid rod end is retained to generate a hole or solid rod blank.

[0222] Sub-step 4.5: Rolling preform

[0223] For holed or solid rod blanks with knurling / toothed ring structures, local rolling pre-forming is performed at the fifth station according to structural requirements. It is mainly used to form special shapes such as anti-slip patterns, embossing or toothed rings on the rod. A dual-die counter-pressing synchronous rolling structure is adopted, and the die speed is synchronized through a constant speed controller to prevent pattern misalignment or knurling deformation, thereby generating complex geometric screw blanks.

[0224] Sub-step 4.6: Multi-point stress monitoring and closed-loop forming control

[0225] During the five-station cold heading process, multi-point force sensors and visual feedback modules are introduced throughout the process for complex geometric screw blanks. The impact force curves and metal flow images of each station are collected. Combined with the forming simulation feedback algorithm, the die gap, punch stroke and lubrication parameters are automatically adjusted to achieve closed-loop optimization control at each stage, improve forming consistency and die service life, and generate the main screw blank.

[0226] Among them, step 5 intelligent temperature control trimming and forming includes the following specific steps:

[0227] Sub-step 5.1: Target area identification and process parameter setting

[0228] Based on the screw body blank obtained in the previous step, a digital image recognition system combined with a 3D geometric comparison algorithm is used to automatically identify crack-prone locations such as the head edge and the transition zone of the shank. Based on the recognition results, the system sets the local heating target area, heating rate, and target temperature, providing a parameter basis for the next step of precise temperature control and generating a local heating area parameter set.

[0229] Sub-step 5.2: Initialize the in-mold induction heating device

[0230] Based on the local heating area parameter set, a high-frequency induction coil array is arranged in the mold, and heating preparation is initiated through a closed-loop temperature control system. The system uses a dual-channel feedback mechanism of thermocouple + infrared temperature measurement to ensure that the heating process is controlled within the target set value range of 60℃ to 120℃, ensuring heating uniformity and preventing overheating, thereby generating controllable heated screw blanks.

[0231] Sub-step 5.3: Temperature difference control die synchronization execution

[0232] The controllably heated screw blank is fed into a cold precision mold, where the temperature difference induces metal micro-flow to reconstruct the lattice. Synchronous stamping technology and self-lubricating mold cavity materials are used to complete edge trimming, shape fine-tuning, and head micro-crack relief. This process utilizes heat-cold coupling to enhance the microstructure density and produce a temperature-differential trimmed semi-finished product.

[0233] Sub-step 5.4: Edge integrity and density testing

[0234] Based on temperature differential trimming of semi-finished products, we introduce microscopic visual inspection and ultrasonic local density analysis technology to automatically determine the integrity of the formed edge, the recovery effect of material microcracks, and the degree of tissue densification. Products that do not meet critical standards are automatically withdrawn from the production line through an automatic return system, ensuring that only high-quality semi-finished products enter the next step, resulting in semi-finished products with qualified edge quality.

[0235] Sub-step 5.5: Intelligent Data Storage and Traceability Coding

[0236] Semi-finished products that meet edge quality standards are transferred to the online laser marking system, where each workpiece is coded with its temperature control parameters, die pressure curve, and density level to achieve product traceability. The coded data is then uploaded to the Manufacturing Execution System (MES) to provide a basis for subsequent mold wear and repair analysis, generating traceable semi-finished products.

[0237] Sub-step 5.6: Stable output and production line buffer management

[0238] The semi-finished products with traceability marks are collected in an automatic cache unit, and the feeding and discharging rhythm is controlled by batch and partition storage logic to provide stable input for the next step of mold fine-tuning. The system simultaneously records the temperature control stability and forming integrity of this batch of products as a basis for subsequent adjustments, and generates temperature-controlled formed semi-finished products.

[0239] Among them, step 6 of adaptive mold fine-tuning correction includes the following specific steps:

[0240] Sub-step 6.1: Pre-processing of contour error detection

[0241] Based on the temperature-controlled forming of semi-finished products, a high-precision multi-point laser displacement sensor array is used in conjunction with a rotating platform system to perform 360-degree scanning and modeling of the workpiece. An edge recognition algorithm and a contour extraction program are used to extract data on the geometric contour ellipticity, end face deflection, and length deviation of the blank in real time to generate a blank geometric error data model.

[0242] Sub-step 6.2: Intelligent identification and analysis of error characteristics

[0243] Based on the blank geometric error data model, the neural network error analysis module automatically classifies and identifies the types and degrees of various forming deviations, outputs a set of optimal correction solution recommendations, including core parameters such as mold gap adjustment amount and press angle variation range, and generates a set of intelligent correction parameter recommendations;

[0244] Sub-step 6.3: In-mold micro-drive pre-adjustment settings

[0245] The intelligent correction parameter recommendation set is imported into the mold micro-drive control unit. Based on the piezoelectric drive + closed-loop displacement feedback system, the system automatically sets the initial state of each micro-adjustment mechanism, the initial state of the guide column micro-preload, the wedge adjustment plate angle, and the micro-inclination angle of the pressing surface, ensuring that all key parts are in the correction preparation state and generating the set state mold system.

[0246] Sub-step 6.4: Fine-tune the calibration operation execution

[0247] The set state mold system and the temperature-controlled formed semi-finished product are synchronously loaded into the working cavity, and the in-mold micro-drive profiling system is activated. According to the pre-correction parameters, micron-level profiling compensation operations are performed, including end face flattening, centering and rounding, and length adjustment. Multiple simultaneous operations ensure the geometric consistency of the forming and generate a primary correction screw blank.

[0248] Sub-step 6.5: Feedback refinement and adaptive adjustment

[0249] Based on the primary correction screw blank, the laser profile is scanned again, and the corrected geometric shape is matched and analyzed with the standard template. If there is a slight residual error, the adaptive closed-loop control logic is triggered to perform a secondary micro-compensation operation on the mold drive unit to complete the final correction process and generate a precision correction screw blank.

[0250] Sub-step 6.6: Workpiece unloading and result confirmation

[0251] The precision correction screw blank is unloaded from the mold system and enters the high-precision contact measuring tool and three-dimensional vision-assisted inspection system to conduct final confirmation of the key dimensions after correction, the coaxiality of the head, the rod length, and the symmetry dimensions to ensure that it meets the requirements of the next thread rolling process and generates a fine-tuned screw blank.

[0252] Among them, step 7 thread rolling includes the following specific steps:

[0253] Sub-step 7.1: Cleaning and positioning the blank before rolling

[0254] Based on the fine-finished screw blanks obtained in step 6, the workpieces are fed one by one into the front station of the thread rolling equipment through a vibrating feeding system and an automatic arrangement track. Alkaline ultrasonic cleaning and hot air drying devices are used to quickly clean their surfaces to remove residual oil film and micro-dust, ensuring the cleanliness of the interface for thread rolling and forming, and generating positioned clean screw blanks.

[0255] Sub-step 7.2: Thread rolling die preload and profile adjustment

[0256] The positioning clean screw blank is loaded into the die system of the constant speed rolling thread rolling machine. According to different thread specifications, high hardness tungsten-cobalt alloy thread rolling dies are selected. The tooth profile engagement angle and lead are preset through the die angle fine-tuning mechanism. The coaxiality error between the blank center and the die engagement line is achieved by using the visual calibration system to ≤0.02mm, generating a preloaded blank assembly.

[0257] Sub-step 7.3: Constant speed thread rolling

[0258] Based on the preloaded blank assembly, the main drive system of the thread rolling machine is activated, maintaining the die linear speed in a constant range of 20 to 30 m / min. Through a two-die intermeshing rolling method or a dynamic and static meshing rolling method, the material surface induces plastic flow mechanism to directly press out the thread profile without cutting. During this process, the thread rolling pressure closed-loop control system monitors and adjusts the die gap in real time to ≤±0.01mm, ensuring consistency between pitch and thread height, and producing a pre-rolled screw.

[0259] Sub-step 7.4: Thread profile accuracy detection and fine-tuning compensation

[0260] Online laser 3D profile scanning is performed on the initially rolled screws to extract data such as thread angle, pitch, and top circle runout. A self-learning thread deviation analysis model is then used to determine if there is any deviation from the set standard. If a deviation is detected, the automatic die angle compensation module is triggered to perform dynamic adjustments at the micron level to ensure the final thread profile is accurate and produce a precision threaded screw.

[0261] Sub-step 7.5: Surface Hardening and Residual Stress Adjustment

[0262] For precision threaded screws, choose whether to perform rolling-induced strengthening treatment based on product strength requirements. This treatment applies secondary micro-strain strengthening to the threaded area by continuing the low-speed rolling process, inducing dense dislocations and residual compressive stress in the surface layer to improve fatigue resistance. If not performed, this step is skipped to produce a surface-strengthened thread semi-finished product.

[0263] Sub-step 7.6: Finished product collection and initial inspection and classification

[0264] The surface-strengthened threaded semi-finished products are placed on the automatic sorting track and pre-classified and identified through a simple contact-type tooth profile inspection and an optical sensor rapid grading device. Workpieces with obvious tooth deviation or tooth bottom damage are removed in advance, and qualified parts are sorted into the next station for online inspection to produce thread-rolling screw semi-finished products.

[0265] Among them, step 8 online detection and automatic sorting includes the following specific steps:

[0266] Sub-step 8.1: Screw Arrangement and Stable Delivery

[0267] Based on the semi-finished thread-rolling screws obtained in the previous step, a vibrating plate loading system and linear guide mechanism are used to arrange the screws one by one, ensuring that the screws enter the inspection area in a uniform direction. The synchronous photoelectric sensor and servo drive control system adjust the conveying speed and spacing in real time to avoid workpiece stacking and jitter, ensure image clarity, and generate a directional screw conveying flow.

[0268] Sub-step 8.2: Multi-angle industrial visual recognition inspection

[0269] Based on the directional conveying of screw flow, a multi-angle industrial camera system configured on the four sides and top uses structured light projection + high-speed shutter imaging technology to capture images of the screw's key dimensions, including head profile, thread geometry, shaft length, and coaxiality. An image stitching and reconstruction algorithm is then introduced to perform 3D appearance modeling and analysis, generating a multi-dimensional structural inspection dataset.

[0270] Sub-step 8.3: Laser scanning and surface defect identification

[0271] The multi-dimensional structural inspection data set is transmitted to the downstream high-resolution laser scanning system, which combines it with a screw rotation mechanism to perform 360° scanning without blind spots. The system uses a speckle interferometry algorithm and an edge reflection contrast model to identify tiny surface defects such as cracks, bumps, burrs, and missing teeth. The system has a high-precision feature with a false detection rate of less than 0.5%, generating a surface defect identification information set.

[0272] Sub-step 8.4: AI intelligent classification and judgment

[0273] Based on the surface defect recognition information set, the pre-trained deep convolutional neural network (CNN) model is called to input various detection features of the screws, including the screw size deviation, defect type, and distribution location. The system outputs whether the screws are qualified or not, and whether they are defective. The system uses an adaptive threshold learning mechanism to adjust the standard tolerance based on historical batches, effectively improving the judgment accuracy and generating a classification judgment label set.

[0274] Sub-step 8.5: Automatic sorting and rejection of defective products

[0275] The generated classification label set is synchronously sent to the sorting control module. This module links the pneumatic robot arm and the track diversion mechanism to ensure that qualified products are directed into the packaging section, and unqualified products are automatically removed and sent to the waste recycling channel. Under the guarantee of 100% online detection, this process can stably maintain an effective sorting accuracy rate of more than 99.5%, and generate a set of sorted screw products;

[0276] Sub-step 8.6: Archiving and Traceability Coding of Test Data

[0277] Based on the set of sorted screw products, the image, size, and defect judgment information of each product are used to generate a unique data fingerprint file, and the traceability code is directly engraved on the screw head or outer packaging through laser marking equipment or QR code inkjet printer to achieve full-process data tracking and batch traceability of the product, and generate qualified screw products.

[0278] Among them, step 9 energy efficiency analysis and data feedback optimization includes the following specific steps:

[0279] Sub-step 9.1: Data collection of the entire cold heading process

[0280] Based on the complete manufacturing process of qualified screw products, a multi-channel industrial data acquisition system (SCADA) was deployed. This system integrated energy consumption sensors from the equipment PLC, die wear monitoring modules, and lubrication flow meter terminals to form a multi-dimensional production parameter database. This database provided the basis for subsequent modeling and generated a multi-source dataset for the cold heading manufacturing process.

[0281] Sub-step 9.2: Energy Efficiency Assessment and Modeling Analysis

[0282] Based on a multi-source dataset of the cold heading manufacturing process, a corresponding line energy consumption assessment model was established through a digital twin modeling platform. The model integrates the equipment operation cycle energy efficiency coefficient, the die life decay curve, and the lubricant usage efficiency model. Multivariate regression analysis and clustering algorithms are used to identify energy efficiency bottlenecks and redundant links, and an energy efficiency assessment report for the cold heading line is generated.

[0283] Sub-step 9.3: Optimization calculation of mold and lubrication cycle

[0284] Based on the cold heading line energy efficiency assessment report, the die attenuation curve and lubrication usage model were used. Parameter sensitivity analysis and Bayesian optimization algorithms were used to calculate the optimal die replacement cycle and lubricant injection strategy. This approach achieved a balance between minimizing unit energy consumption and maximizing die utilization, generating the optimal die lubrication strategy set.

[0285] Sub-step 9.4: Generate dynamic device parameter adjustment suggestions

[0286] Combined with the optimal mold lubrication strategy set, the process feedback rule engine generates specific equipment-level parameter adjustment recommendations, including key parameter settings for impact frequency, slide travel speed, and cooling system on-time. By establishing a mapping table with the actual equipment interface, instruction-level issuance is standardized and a table of optimized equipment setting parameters is generated.

[0287] Sub-step 9.5: Process database update and self-learning mechanism activation

[0288] Uploading the equipment optimization parameter table to the process parameter database of the Manufacturing Execution System (MES) triggers the self-learning process feedback module, which adjusts its internal parameter weighting coefficients based on historical batch performance, generates batch-by-batch optimization suggestions for quality and accuracy, and generates a dynamic optimization process parameter model.

[0289] Sub-step 9.6: Automatically update and push the next batch of setting values

[0290] Based on the dynamic optimization process parameter model, the MES system automatically pushes updated forming pressure settings, lubrication cycle control, mold calibration warning values ​​and other parameters to the next batch of screw cold heading tasks, realizing closed-loop optimization of the entire process; after the push is confirmed by the operator, it is formally introduced into production to generate an optimized feedback parameter set.

[0291] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A screw cold heading process, characterized in that: The following steps are involved: Step 1: Wire preparation and surface pretreatment Based on the requirements of screw forming on the surface quality of raw materials, a composite process of mechanical wire drawing and chemical phosphating and saponification is used to treat the metal wire to remove the surface oxide scale and form a lubricating film layer, effectively improving the forming fluidity and die life in the subsequent cold heading process, and producing pre-treated wire; Step 2: Cut to length Based on the pre-processed wire, a high-speed fixed-length shearing device is used to cut the wire into small segments of the required length according to the set size through a servo control system, which is called a billet. This ensures the consistency and dimensional stability of subsequent forming and generates fixed-length billets. Step 3: Primary upsetting (head pre-forming) Based on the fixed-length blank, a one-station cold heading process is adopted to perform axial impact forming on one end of the blank in the die to form the basic outline of the screw head, laying the foundation for subsequent precision forming and generating a head preform blank; Step 4: Multi-station precision cold heading Based on the head preform, a multi-station cold heading machine is used to sequentially perform multiple deformation processes such as upsetting, diameter reduction, and punching. The shank diameter and head structure of the screw are precisely controlled. A closed mold system is used to ensure dimensional accuracy and structural integrity to generate the screw body blank. Step 5: Intelligent temperature control trimming and forming Based on the main blank of the screw body, in-mold local temperature control technology is used to induce micro-heating to briefly heat the key deformation area, and combined with the cold area to apply force to achieve higher-quality detail forming. It is especially suitable for preventing problems such as cracking and collapse of the head or hexagonal edge, and producing temperature-controlled formed semi-finished products; Step 6: Adaptive mold fine-tuning correction Based on the temperature-controlled forming of the semi-finished product, the intelligent mold system uses sensors to monitor the geometric deviation information of the semi-finished product, automatically fine-tune the mold pressing angle and closing gap, perform end finishing and transition boundary optimization, and generate a finished screw blank; Step 7: Thread rolling Based on the finishing screw blank, a non-cutting thread rolling process is adopted. The rod surface is plastically flowed into a thread shape by rolling the thread rolling die in the thread rolling machine, ensuring the thread engagement accuracy and surface hardening, and producing a semi-finished thread rolling screw; Step 8: Online detection and automatic sorting Based on the semi-finished thread-rolling screws, a high-speed image recognition system and multi-dimensional size / hardness testing equipment are used to fully inspect the screw head, thread, length, concentricity, surface defects and other indicators, and automatically eliminate defective or abnormal workpieces to produce qualified finished screws. Step 9: Energy efficiency analysis and data feedback optimization Based on the batch processing data of qualified screw products, the energy monitoring module and process analysis system are used to analyze and feedback the energy efficiency, mold life and process parameters of the entire production line, automatically update the forming parameter library to optimize subsequent batch production, and generate an optimized feedback parameter set.

2. A screw cold heading forming process according to claim 1, characterized in that: The step 1 of wire preparation and surface pretreatment includes the following specific steps: Sub-step 1.1: Annealing of the original disk element Based on the premise that cold heading has high requirements for material plasticity, medium carbon steel discs are selected as raw materials. They are first spheroidized and annealed in a continuous spheroidizing annealing furnace. The temperature is set at 720℃ and kept at this temperature for 4 hours to transform the internal pearlite into spheroidal cementite, thereby improving the material's ductility and deformation uniformity, and producing annealed discs. Sub-step 1.2: Multi-pass wire drawing and reducing Based on the annealed disc obtained in the previous step, a continuous wire drawing unit is used to perform three-stage wire drawing reduction in sequence, with the reduction rates controlled at 20%, 15%, and 10% respectively. Ultimately, the outer diameter of the wire is controlled within the cold heading requirement of ±0.05mm, and the grain extension direction of the material is aligned with the forming direction, which facilitates subsequent metal flow and produces drawn wire. Sub-step 1.3: Surface degreasing and pickling The drawn wire is fed into the surface pretreatment system, where it undergoes alkaline solution spray degreasing, water rinsing, and hydrochloric acid pickling to remove surface oil and scale, and ensure that the surface roughness is suitable for phosphating adhesion. The pickling solution concentration is maintained at 10% and the temperature is controlled at 40-50°C to produce clean and activated wire. Sub-step 1.4: Chemical phosphating film generation After the cleaned and activated wire enters the phosphating tank reaction zone, a crystalline phosphate film with a thickness of 5 to 15 μm is formed by reaction in a zinc phosphate solution, which enhances the adhesion of the subsequent lubricating film. This process uses an automatic pH monitoring system to ensure process stability, maintaining it at 2.5 to 3.0 to produce phosphate-treated wire. Sub-step 1.5: Saponification lubrication layer treatment The phosphating wire is immersed in a sodium saponification solution to form a dry lubricating film with a thickness of about 1.0 to 1.5 μm, which can significantly reduce friction during the subsequent upsetting process and prevent direct adhesion between the material and the die. The final drying process is completed by a hot water spray drying process to avoid lubricant residue crystallization and produce saponified lubricated wire. Sub-step 1.6: Coil Arrangement and Quality Review The saponified lubricated wire is recycled into standard coils through an automatic coiling device, and the wire surface microcracks, lubricating film integrity and tensile yield strength are comprehensively tested using an online eddy current flaw detector + tensile testing unit to ensure that it meets the conditions for subsequent forming and use, thereby generating pre-treated wire.

3. The screw cold heading forming process according to claim 1, characterized in that: The step 2 of cutting to length includes the following specific steps: Sub-step 2.1: Wire introduction and tension control Based on the pre-treated wire generated in step 1, it is introduced into the shearing module through the constant tension wire feeding device. The system adopts a servo tension controller + dual sensor closed-loop feedback loop to automatically adjust the tension range to 1.2~1.5N / mm 2 , ensuring that the wire is in a constant tension state when entering the shearing zone, avoiding jumping or blocking, and generating constant tension for the wire; Sub-step 2.2: Online laser length measurement and recognition The constant tension wire is introduced into the bidirectional laser length measurement system, and a three-axis laser displacement sensor is used to continuously scan and measure the wire's forward speed and instantaneous length. The built-in fuzzy PID control algorithm automatically determines the target shearing time point, outputs a precise displacement signal, and generates a shear positioning signal for the wire. Sub-step 2.3: High-speed closed-loop cut-to-length Based on the shear positioning signal wire, the shearing operation is performed by the servo punching and shearing integrated mechanism. The shearing blade is made of high-speed powder metallurgy alloy steel with a hardness of HRC 68 or above. Combined with the high-frequency response servo system, the shearing time is controlled within the range of 5-7ms. The shearing accuracy data is recorded synchronously to achieve online shearing dimensional error control within ±0.05mm, generating high-precision sheared blanks. Sub-step 2.4: Billet conveying and alignment The high-precision sheared blanks are transferred to a multi-channel vibration alignment mechanism, where linear vibration troughs and guide rails are used to align the direction and spacing of the blanks, facilitating their subsequent entry into the upsetting station. An image recognition system is also used to detect any anomalies such as length deviation and burrs, eliminating defective products and generating an aligned row of sheared blanks. Sub-step 2.5: Verification and feedback of cutting result data Based on the entire row of sheared blanks, the shear section length, end face verticality and residual stress data are collected at the shear end detection station. The data is analyzed through the industrial edge computing node, and parameter fine-tuning suggestions are sent to the shear control system to achieve continuous optimization of the shearing process and generate qualified shear control blanks.

4. The screw cold heading forming process according to claim 1, characterized in that: The step 3 of primary upsetting (head preforming) includes the following specific steps: Sub-step 3.1: Blank positioning and clamping Based on the fixed-length blank generated in step 2, it is accurately fed into a single-station closed mold system by a precision manipulator and an automatic feeding mechanism. Pneumatic clamping and servo displacement sensors are used to ensure the center alignment and axial stability of the blank in the mold cavity, providing a consistent initial state for subsequent impact loading, and generating a positioned and clamped blank. Sub-step 3.2: Impact loading initial forming The positioned and clamped blank is placed in a closed die system, and the high-speed hydraulic punch system is started to apply an axial impact load. The loading rate is controlled at 8-12 m / s to achieve preliminary upsetting of the blank head. In this stage, the main plastic deformation is concentrated at one end, and the metal is controlled to expand radially to form a primary drum-shaped structure, generating an impact-formed blank. Sub-step 3.3: Induction of radial return flow paths Based on the impact of the initial blank, the mold's lateral flow control groove structure is simultaneously activated during the impact holding phase, inducing the material to flow back in the radial and anti-axial directions under pressure, prompting the metal to evenly fill the concave area of ​​the mold cavity and improve the geometric symmetry of the head. This flow control structure is designed based on finite element simulation analysis results to ensure the optimal path and produce a flow-optimized blank. Sub-step 3.4: Dynamic monitoring of the forming process The displacement-load curve of the flow-controlled optimized billet is collected and analyzed in real time during its forming process. Multi-channel strain gauges and high-speed data acquisition cards are used to evaluate the material response characteristics at each loading moment to ensure that no overload or concentrated deformation areas occur during impact upsetting. The monitoring data is fed back to the subsequent mold adjustment and energy efficiency optimization system to generate qualified billets for forming monitoring. Sub-step 3.5: Demolding and cleaning of the forming area The qualified blanks after forming monitoring are removed from the mold, and a mechanical ejector + negative pressure adsorption mechanism is used in combination. During the mold empty window period, the in-mold gas purge and minimal lubrication system are used to remove residual burrs and metal dust to ensure the cleanliness of the mold and the stability of the next cycle, and to generate a head preform.

5. The screw cold heading forming process according to claim 1, characterized in that: The multi-station precision cold heading forming in step 4 includes the following specific steps: Sub-step 4.1: Re-upsetting - Deepening of the head Based on the head preform in step 3, it enters the first station die of the five-station cold heading machine and adopts a parallel bidirectional impact upsetting process to further compact the head geometry. The servo-controlled high-frequency impact system adjusts the stroke energy to improve the head density and shape symmetry, laying a precision foundation for complex geometric structures and generating a deep-forged head blank. Sub-step 4.2: Reduction—Rod Size Adjustment The deep-forging head blank is transported to the second station, where progressive diameter reduction forming technology is used. The tapered heading die and radial guide sleeve are used to precisely control the diameter reduction deformation of the rod. During this process, the in-die strain sensing unit monitors the rod stress state in real time to prevent metal cracking or center deviation caused by uneven diameter reduction, thus generating a reduced diameter rod blank. Sub-step 4.3: Local extrusion—head detail reconstruction Based on the reduced diameter shank blank, local reverse extrusion is performed in the third station, mainly used to reshape the detailed structure of the screw head. For cross-recessed and hexagonal cap flange screws, a closed reverse extrusion die and a floating core structure are used to precisely form the local geometric shape without increasing the total length of the blank. The die temperature is maintained at room temperature ±2°C to prevent thermal expansion errors and generate the head geometric structure blank. Sub-step 4.4: Punching - Forming the hole at the rod end For the head geometry blank, if it is designed as a hollow screw, the fourth station high-speed axial punching process is performed. A carbide hollow punch needle and a guide positioning mechanism are used to control the punch depth and maintain the coaxiality of the core shaft to ensure the hole diameter accuracy within ±0.02mm. For solid screws, this station is skipped and the solid rod end is retained to generate a hole or solid rod blank. Sub-step 4.5: Rolling preform For holed or solid rod blanks with knurling / toothed ring structures, local rolling pre-forming is performed at the fifth station according to structural requirements. It is mainly used to form special shapes such as anti-slip patterns, embossing or toothed rings on the rod. A dual-die counter-pressing synchronous rolling structure is adopted, and the die speed is synchronized through a constant speed controller to prevent pattern misalignment or knurling deformation, thereby generating complex geometric screw blanks. Sub-step 4.6: Multi-point stress monitoring and closed-loop forming control During the five-station cold heading process, multi-point force sensors and visual feedback modules are introduced throughout the process for complex geometric screw blanks. The impact force curves and metal flow images of each station are collected. Combined with the forming simulation feedback algorithm, the die gap, punch stroke and lubrication parameters are automatically adjusted to achieve closed-loop optimization control at each stage, improve forming consistency and die service life, and generate the main screw blank.

6. The screw cold heading forming process according to claim 1, characterized in that: The step 5 of intelligent temperature-controlled trimming and forming includes the following specific steps: Sub-step 5.1: Target area identification and process parameter setting Based on the screw body blank obtained in the previous step, a digital image recognition system combined with a 3D geometric comparison algorithm is used to automatically identify crack-prone locations such as the head edge and the transition zone of the shank. Based on the recognition results, the system sets the local heating target area, heating rate, and target temperature, providing a parameter basis for the next step of precise temperature control and generating a local heating area parameter set. Sub-step 5.2: Initialize the in-mold induction heating device Based on the local heating area parameter set, a high-frequency induction coil array is arranged in the mold, and heating preparation is initiated through a closed-loop temperature control system. The system uses a dual-channel feedback mechanism of thermocouple + infrared temperature measurement to ensure that the heating process is controlled within the target set value range of 60℃ to 120℃, ensuring heating uniformity and preventing overheating, thereby generating controllable heated screw blanks. Sub-step 5.3: Temperature difference control die synchronization execution The controllably heated screw blank is fed into a cold precision mold, where the temperature difference induces metal micro-flow to reconstruct the lattice. Synchronous stamping technology and self-lubricating mold cavity materials are used to complete edge trimming, shape fine-tuning, and head micro-crack relief. This process utilizes heat-cold coupling to enhance the microstructure density and produce a temperature-differential trimmed semi-finished product. Sub-step 5.4: Edge integrity and density testing Based on temperature differential trimming of semi-finished products, we introduce microscopic visual inspection and ultrasonic local density analysis technology to automatically determine the integrity of the formed edge, the recovery effect of material microcracks, and the degree of tissue densification. Products that do not meet critical standards are automatically withdrawn from the production line through an automatic return system, ensuring that only high-quality semi-finished products enter the next step, resulting in semi-finished products with qualified edge quality. Sub-step 5.5: Intelligent Data Storage and Traceability Coding Semi-finished products that meet edge quality standards are transferred to the online laser marking system, where each workpiece is coded with its temperature control parameters, die pressure curve, and density level to achieve product traceability. The coded data is then uploaded to the Manufacturing Execution System (MES) to provide a basis for subsequent mold wear and repair analysis, generating traceable semi-finished products. Sub-step 5.6: Stable output and production line buffer management The semi-finished products with traceability marks are collected in an automatic cache unit, and the feeding and discharging rhythm is controlled by batch and partition storage logic to provide stable input for the next step of mold fine-tuning. The system simultaneously records the temperature control stability and forming integrity of this batch of products as a basis for subsequent adjustments, and generates temperature-controlled formed semi-finished products.

7. The screw cold heading forming process according to claim 1, characterized in that: The step 6 of adaptive mold fine-tuning correction includes the following specific steps: Sub-step 6.1: Pre-processing of contour error detection Based on the temperature-controlled forming of semi-finished products, a high-precision multi-point laser displacement sensor array is used in conjunction with a rotating platform system to perform 360-degree scanning and modeling of the workpiece. An edge recognition algorithm and a contour extraction program are used to extract data on the geometric contour ellipticity, end face deflection, and length deviation of the blank in real time to generate a blank geometric error data model. Sub-step 6.2: Intelligent identification and analysis of error characteristics Based on the blank geometric error data model, the neural network error analysis module automatically classifies and identifies the types and degrees of various forming deviations, outputs a set of optimal correction solution recommendations, including core parameters such as mold gap adjustment amount and press angle variation range, and generates a set of intelligent correction parameter recommendations; Sub-step 6.3: In-mold micro-drive pre-adjustment settings The intelligent correction parameter recommendation set is imported into the mold micro-drive control unit. Based on the piezoelectric drive + closed-loop displacement feedback system, the system automatically sets the initial state of each micro-adjustment mechanism, the initial state of the guide column micro-preload, the wedge adjustment plate angle, and the micro-inclination angle of the pressing surface, ensuring that all key parts are in the correction preparation state and generating the set state mold system. Sub-step 6.4: Fine-tune the calibration operation execution The set state mold system and the temperature-controlled formed semi-finished product are synchronously loaded into the working cavity, and the in-mold micro-drive profiling system is activated. According to the pre-correction parameters, micron-level profiling compensation operations are performed, including end face flattening, centering and rounding, and length adjustment. Multiple simultaneous operations ensure the geometric consistency of the forming and generate a primary correction screw blank. Sub-step 6.5: Feedback refinement and adaptive adjustment Based on the primary correction screw blank, the laser profile is scanned again, and the corrected geometric shape is matched and analyzed with the standard template. If there is a slight residual error, the adaptive closed-loop control logic is triggered to perform a secondary micro-compensation operation on the mold drive unit to complete the final correction process and generate a precision correction screw blank. Sub-step 6.6: Workpiece unloading and result confirmation The precision correction screw blank is unloaded from the mold system and enters the high-precision contact measuring tool and three-dimensional vision-assisted inspection system to conduct final confirmation of the key dimensions after correction, the coaxiality of the head, the rod length, and the symmetry dimensions to ensure that it meets the requirements of the next thread rolling process and generates a fine-tuned screw blank.

8. The screw cold heading forming process according to claim 1, characterized in that: The thread rolling process in step 7 includes the following specific steps: Sub-step 7.1: Cleaning and positioning the blank before rolling Based on the fine-finished screw blanks obtained in step 6, the workpieces are fed one by one into the front station of the thread rolling equipment through a vibrating feeding system and an automatic arrangement track. Alkaline ultrasonic cleaning and hot air drying devices are used to quickly clean their surfaces to remove residual oil film and micro-dust, ensuring the cleanliness of the interface for thread rolling and forming, and generating positioned clean screw blanks. Sub-step 7.2: Thread rolling die preload and profile adjustment The positioning clean screw blank is loaded into the die system of the constant speed rolling thread rolling machine. According to different thread specifications, high hardness tungsten-cobalt alloy thread rolling dies are selected. The tooth profile engagement angle and lead are preset through the die angle fine-tuning mechanism. The coaxiality error between the blank center and the die engagement line is achieved by using the visual calibration system to ≤0.02mm, generating a preloaded blank assembly. Sub-step 7.3: Constant speed thread rolling Based on the preloaded blank assembly, the main drive system of the thread rolling machine is activated, maintaining the die linear speed in a constant range of 20 to 30 m / min. Through a two-die intermeshing rolling method or a dynamic and static meshing rolling method, the material surface induces plastic flow mechanism to directly press out the thread profile without cutting. During this process, the thread rolling pressure closed-loop control system monitors and adjusts the die gap in real time to ≤±0.01mm, ensuring consistency between pitch and thread height, and producing a pre-rolled screw. Sub-step 7.4: Thread profile accuracy detection and fine-tuning compensation Online laser 3D profile scanning is performed on the initially rolled screws to extract data such as thread angle, pitch, and top circle runout. A self-learning thread deviation analysis model is then used to determine if there is any deviation from the set standard. If a deviation is detected, the automatic die angle compensation module is triggered to perform dynamic adjustments at the micron level to ensure the final thread profile is accurate and produce a precision threaded screw. Sub-step 7.5: Surface Hardening and Residual Stress Adjustment For precision threaded screws, choose whether to perform rolling-induced strengthening treatment based on product strength requirements. This treatment applies secondary micro-strain strengthening to the threaded area by continuing the low-speed rolling process, inducing dense dislocations and residual compressive stress in the surface layer to improve fatigue resistance. If not performed, this step is skipped to produce a surface-strengthened thread semi-finished product. Sub-step 7.6: Finished product collection and initial inspection and classification The surface-strengthened threaded semi-finished products are placed on the automatic sorting track and pre-classified and identified through a simple contact-type tooth profile inspection and an optical sensor rapid grading device. Workpieces with obvious tooth deviation or tooth bottom damage are removed in advance, and qualified parts are sorted into the next station for online inspection to produce thread-rolling screw semi-finished products.

9. The screw cold heading forming process according to claim 1, characterized in that: The online detection and automatic sorting in step 8 includes the following specific steps: Sub-step 8.1: Screw Arrangement and Stable Delivery Based on the semi-finished thread-rolling screws obtained in the previous step, a vibrating plate loading system and linear guide mechanism are used to arrange the screws one by one, ensuring that the screws enter the inspection area in a uniform direction. The synchronous photoelectric sensor and servo drive control system adjust the conveying speed and spacing in real time to avoid workpiece stacking and jitter, ensure image clarity, and generate a directional screw conveying flow. Sub-step 8.2: Multi-angle industrial visual recognition inspection Based on the directional conveying of screw flow, a multi-angle industrial camera system configured on the four sides and top uses structured light projection + high-speed shutter imaging technology to capture images of the screw's key dimensions, including head profile, thread geometry, shaft length, and coaxiality. An image stitching and reconstruction algorithm is then introduced to perform 3D appearance modeling and analysis, generating a multi-dimensional structural inspection dataset. Sub-step 8.3: Laser scanning and surface defect identification The multi-dimensional structural inspection data set is transmitted to the downstream high-resolution laser scanning system, which combines it with a screw rotation mechanism to perform 360° scanning without blind spots. The system uses a speckle interferometry algorithm and an edge reflection contrast model to identify tiny surface defects such as cracks, bumps, burrs, and missing teeth. The system has a high-precision feature with a false detection rate of less than 0.5%, generating a surface defect identification information set. Sub-step 8.4: AI intelligent classification and judgment Based on the surface defect recognition information set, the pre-trained deep convolutional neural network (CNN) model is called to input various detection features of the screws, including the screw size deviation, defect type, and distribution location. The system outputs whether the screws are qualified or not, and whether they are defective. The system uses an adaptive threshold learning mechanism to adjust the standard tolerance based on historical batches, effectively improving the judgment accuracy and generating a classification judgment label set. Sub-step 8.5: Automatic sorting and rejection of defective products The generated classification label set is synchronously sent to the sorting control module. This module links the pneumatic robot arm and the track diversion mechanism to ensure that qualified products are directed into the packaging section, and unqualified products are automatically removed and sent to the waste recycling channel. Under the guarantee of 100% online detection, this process can stably maintain an effective sorting accuracy rate of more than 99.5%, and generate a set of sorted screw products; Sub-step 8.6: Archiving and Traceability Coding of Test Data Based on the set of sorted screw products, the image, size, and defect judgment information of each product are used to generate a unique data fingerprint file, and the traceability code is directly engraved on the screw head or outer packaging through laser marking equipment or QR code inkjet printer to achieve full-process data tracking and batch traceability of the product, and generate qualified screw products.

10. The screw cold heading forming process according to claim 1, characterized in that: The energy efficiency analysis and data feedback optimization in step 9 includes the following specific steps: Sub-step 9.1: Data collection of the entire cold heading process Based on the complete manufacturing process of qualified screw products, a multi-channel industrial data acquisition system (SCADA) was deployed. This system integrated energy consumption sensors from the equipment PLC, die wear monitoring modules, and lubrication flow meter terminals to form a multi-dimensional production parameter database. This database provided the basis for subsequent modeling and generated a multi-source dataset for the cold heading manufacturing process. Sub-step 9.2: Energy Efficiency Assessment and Modeling Analysis Based on a multi-source dataset of the cold heading manufacturing process, a corresponding line energy consumption assessment model was established through a digital twin modeling platform. The model integrates the equipment operation cycle energy efficiency coefficient, the die life decay curve, and the lubricant usage efficiency model. Multivariate regression analysis and clustering algorithms are used to identify energy efficiency bottlenecks and redundant links, and an energy efficiency assessment report for the cold heading line is generated. Sub-step 9.3: Optimization calculation of mold and lubrication cycle Based on the cold heading line energy efficiency assessment report, the die attenuation curve and lubrication usage model were used. Parameter sensitivity analysis and Bayesian optimization algorithms were used to calculate the optimal die replacement cycle and lubricant injection strategy. This approach achieved a balance between minimizing unit energy consumption and maximizing die utilization, generating the optimal die lubrication strategy set. Sub-step 9.4: Generate dynamic device parameter adjustment suggestions Combined with the optimal mold lubrication strategy set, the process feedback rule engine generates specific equipment-level parameter adjustment recommendations, including key parameter settings for impact frequency, slide travel speed, and cooling system on-time. By establishing a mapping table with the actual equipment interface, instruction-level issuance is standardized and a table of optimized equipment setting parameters is generated. Sub-step 9.5: Process database update and self-learning mechanism activation Uploading the equipment optimization parameter table to the process parameter database of the Manufacturing Execution System (MES) triggers the self-learning process feedback module, which adjusts its internal parameter weighting coefficients based on historical batch performance, generates batch-by-batch optimization suggestions for quality and accuracy, and generates a dynamic optimization process parameter model. Sub-step 9.6: Automatically update and push the next batch of setting values Based on the dynamic optimization process parameter model, the MES system automatically pushes updated forming pressure settings, lubrication cycle control, mold calibration warning values ​​and other parameters to the next batch of screw cold heading tasks, realizing closed-loop optimization of the entire process; after the push is confirmed by the operator, it is formally introduced into production to generate an optimized feedback parameter set.

Citation Information

Cited By

  • High-strength screw production optimization method based on self-adaptive control

    CN121008556A

  • Code traceability control method and system for thread gauge

    CN121563488A