Power transmission line tower component and intelligent forming and corrosion prevention integrated preparation method

Abstract

The application discloses a power transmission line tower component intelligent forming and corrosion prevention integrated preparation method, which comprises the following steps: incoming material information collection and initial model initialization; cutting process adaptive process parameter optimization and online size sensing based on a first optimization formula; adopting a cutting device to perform steel cutting and blanking; bending process parameter dynamic correction and springback compensation; inputting the cutting size deviation, steel yield strength and plate thickness into a second optimization formula to solve the bending process target process parameters; welding forming process heat input intelligent regulation and welding deformation online correction; inputting the component incoming steel original parameters, cutting size deviation and bending springback deviation into a third optimization formula to calculate the welding process target process parameters; forming component comprehensive state information summary output; hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving, based on a fourth optimization formula, outputting the hot-dip galvanizing target process parameters; passivation process adaptive matching and coating quality closed-loop feedback.

Description

Technical Field

[0001] This invention relates to the field of power transmission tower component processing and manufacturing technology, and in particular to an integrated preparation process for forming and processing key components such as angle steel and connecting plates of power transmission towers, as well as hot-dip galvanizing and anti-corrosion treatment. Specifically, it is a method for intelligent forming and anti-corrosion integrated preparation of power transmission tower components, which is applicable to the mass intelligent manufacturing of steel structure components for high-voltage and ultra-high-voltage power transmission towers and belongs to the category of power steel structure component manufacturing technology. Background Technology

[0002] As the core supporting carrier of overhead power transmission lines, the overall structural safety of transmission towers directly determines the operational reliability of the transmission lines. Angle steel and connecting plates are the most abundant critical load-bearing components in the towers. The geometric dimensional processing accuracy of these components, the mechanical properties of welded joints, and the quality of the hot-dip galvanized anti-corrosion layer collectively determine the overall structural strength, fatigue resistance, and outdoor service life of the tower. The existing manufacturing process for transmission tower components is divided into two independent production segments: component forming and processing, and hot-dip galvanizing and anti-corrosion treatment. Forming mainly includes plate cutting, bending, assembly welding, and component correction. Anti-corrosion treatment mainly includes component pretreatment, hot-dip galvanizing, and passivation post-treatment. The process parameters of these two production segments are set independently, and the production process relies on the experience of operators to adjust these parameters, lacking a closed-loop feedback control link between the various processes.

[0003] In the component forming and processing stage, existing cutting, bending, and welding equipment mostly adopts preset fixed process parameters for production. For steel used in iron towers with different thicknesses and materials, it is impossible to dynamically adjust the process parameters in real time according to the actual material fluctuations and rolling deformation deviations of the incoming material. During the flame cutting or plasma cutting process of angle steel components, the plate itself has residual rolling stress, and cutting thermal deformation will occur after cutting. During the bending process, the material springback varies with the yield strength of the steel and the thickness of the plate. Fixed bending stroke parameters will cause the bending angle and leg width of the component to exceed the tolerance. In the welding stage, unstable welding heat input will cause welding deformation, resulting in deviations in the flatness of the connecting plates and the relative dimensions of the welded angle steel components and the holes. Currently, the industry generally adopts an offline sampling inspection mode after production is completed. After the inspection finds components with out-of-tolerance dimensions, manual correction is then performed. The amount of correction by manual correction depends on the operator's experience, resulting in poor consistency of correction effect. Improper correction can also introduce additional residual internal stress, reducing the fatigue performance of tower components. In the case of mass production, the offline inspection mode cannot achieve real-time control of the dimensions of all components. Dimensional deviations of components will increase the difficulty of on-site assembly and installation of towers. When the bolt hole alignment deviation is serious, it will directly lead to the scrapping of components.

[0004] In the hot-dip galvanizing anti-corrosion process, hot-dip galvanizing is the mainstream anti-corrosion method for steel structural components of power transmission towers. The uniformity of the anti-corrosion layer thickness and the adhesion of the coating are the core indicators for evaluating the quality of galvanized components. In the existing hot-dip galvanizing production process, the immersion temperature, the immersion time of the component, and the passivation agent ratio parameters are mostly selected based on standard experience ranges. Operators roughly set process parameters based on the component's shape and wall thickness. Tower angle steel and connecting plate components have structural characteristics such as large differences in wall thickness, welded joints, and liquid retention in the component's internal cavity. Thick-walled components have high heat storage, while thin-walled components heat up quickly. When components of different specifications are simultaneously immersed in the zinc bath, fixed immersion process parameters will result in problems such as over-galvanizing of thin-walled components (leading to excessively thick coatings that easily peel off) and insufficient immersion and localized missed galvanizing of thick-walled components. Residual impurities from the component's pretreatment and surface deformation stress from the component forming process also affect the zinc immersion effect. Information on component deformation defects from the forming process cannot be transmitted to the galvanizing process, making it impossible to adapt galvanizing process parameters to the individual condition of the component. During the passivation treatment stage, the passivation solution temperature and passivation immersion time are set at a fixed value. When the coating thickness on the component surface is inconsistent, the quality of the passivation film formation is uneven, which further reduces the outdoor corrosion resistance and durability of the component.

[0005] In existing technologies, the forming and processing steps are isolated from the hot-dip galvanizing and anti-corrosion steps. Dimensional deviations, residual stress states, and surface quality information generated in the forming step cannot be transmitted to the hot-dip galvanizing step. Furthermore, the coating quality inspection results from the hot-dip galvanizing step cannot provide feedback to control the upstream cutting, bending, and welding processes. The production chain lacks cross-process closed-loop control logic. Some existing solutions introduce machine vision inspection methods, but these are limited to identifying surface defects in finished components. They lack a quantitative physical model between process parameters and the physical state of the components, and can only perform defect identification and alarm functions, not adaptive correction of upstream process parameters. Other intelligent transformation solutions directly apply general machine learning black-box models. Model training heavily relies on massive amounts of production sample data, resulting in poor model interpretability and a lack of physical mechanism constraints. Under the condition of small-batch, multi-specification component production for iron towers, the model's prediction accuracy decreases, and the process parameter output is prone to anomalies that do not conform to the physical boundaries of steel processing, making it unsuitable for direct use in industrial production.

[0006] Transmission tower components are critical to power safety. Their manufacturing processes not only need to ensure production efficiency but also require interpretable process control logic and physically constrained parameter outputs. This is to prevent substandard mechanical properties caused by unreasonable process parameters output by unconstrained black-box models. In summary, current production methods for transmission tower angle steel and connecting plates suffer from several problems: static setting of process parameters at each stage; lack of closed-loop feedback between stages; insufficient online correction capability for dimensional deviations; poor matching of hot-dip galvanizing process parameters; difficulty in consistently controlling the uniformity and adhesion of the anti-corrosion layer; and insufficient overall production intelligence. There is still room for improvement in component processing accuracy, structural strength, and service life. Therefore, there is an urgent need to develop an integrated manufacturing technology that balances high-precision forming of components with high-quality hot-dip galvanizing and anti-corrosion treatment. Summary of the Invention

[0007] The purpose of this invention is to overcome the technical defects in the existing technology, such as the independent forming and hot-dip galvanizing anti-corrosion processes of power transmission tower components, static configuration of process parameters, lack of cross-process closed-loop control, large dimensional deviations in component cutting, bending, welding and forming, and difficulty in stable control of the uniformity and adhesion of hot-dip galvanized coating. This invention provides a method for preparing power transmission tower components and an integrated intelligent forming and anti-corrosion method. This invention, based on the physical mechanisms of steel plastic deformation, welding heat conduction, and zinc immersion coating growth, constructs an artificial intelligence computational model with physical constraints. This model optimizes process parameters in real time throughout the entire process of cutting, bending, and welding angle steel and connecting plate components. Combined with online machine vision detection, it completes closed-loop correction of component dimensional deviations. Simultaneously, it incorporates component material, deformation, and surface condition information collected during the forming process into the hot-dip galvanizing process control, adaptively matching immersion temperature, immersion time, and passivation process parameters. This achieves integrated intelligent closed-loop production of the entire process, from component forming and hot-dip galvanizing to corrosion protection. This improves the geometric processing accuracy of tower components, reduces internal residual stress, improves the uniformity and adhesion of the hot-dip galvanized anti-corrosion layer, and enhances the overall structural strength and long-term outdoor service life of the tower components. This invention also provides a transmission line tower component and an integrated intelligent forming and corrosion protection manufacturing system for implementing the aforementioned manufacturing method.

[0008] To achieve the above technical objectives, the present invention adopts the following technical solution:

[0009] The first aspect of this invention is to provide a method for the integrated intelligent molding and corrosion protection of transmission line tower components, comprising the following steps:

[0010] S1, Incoming Material Information Acquisition and Initial Model Initialization: Obtain the original parameters of the incoming steel material for the components of the transmission tower to be processed. The original parameters of the incoming steel material include steel grade, plate thickness, initial yield strength of the plate, residual stress distribution of the plate during rolling, surface roughness of the plate, and geometric parameters of the component design drawings; input the original parameters of the incoming steel material into the artificial intelligence control model with physical constraints to complete the initial parameter assignment of the model.

[0011] S2, Adaptive process parameter optimization and online dimension perception for the cutting process: Based on the first optimization formula, the target process parameters for the cutting process are calculated in combination with the original parameters of the incoming steel. The target process parameters for the cutting process include cutting speed, cutting heat source power, and cutting heat source offset compensation. The steel is cut and blanked using a cutting device. After the steel is cut and blanked, the actual measured values ​​of the cross-sectional contour and edge dimensions of the cut component are collected by machine vision, and the cutting dimension deviation is calculated.

[0012] S3, Dynamic correction and springback compensation of bending process parameters: Input the cutting size deviation, steel yield strength and plate thickness into the second optimization formula to solve the target process parameters of the bending process, wherein the target process parameters of the bending process include the opening distance of the lower bending die, the downward stroke of the bending pressure head and the bending holding time; perform the bending process, and after the bending process is completed, the machine vision collects the actual measured values ​​of the bending component limb width and the bending inner angle to obtain the bending springback deviation;

[0013] S4, Intelligent control of heat input and online correction of welding deformation in welding forming process: The original parameters of the incoming steel of the component, the cutting size deviation and the bending springback deviation are input into the third optimization formula to calculate the target process parameters of the welding process. The target process parameters of the welding process include welding current, welding voltage, welding travel speed and welding intermittent cooling time; the welding operation of angle steel and connecting plate assembly is performed. During the welding process, the measured data of temperature field of welding area, flatness of component and relative position of bolt holes are collected in real time to obtain welding deformation deviation, and the welding heat input parameters are finely adjusted in real time based on the welding deformation deviation.

[0014] S5, Comprehensive status information of formed components is summarized and output: The cutting size deviation, bending springback deviation and welding deformation deviation are summarized, the estimated value of residual stress of the component and the surface morphology feature data of the component are extracted, and a comprehensive status data package of the formed component is generated. The comprehensive status data package of the formed component carries the individual component identifier and is transmitted to the downstream hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution and passivation process adaptive matching and coating quality closed-loop feedback process.

[0015] S6, Hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution: Receive the comprehensive status data packet of the formed component, complete the component degreasing, pickling and fluxing pretreatment processes; input the component plate thickness, comprehensive deformation deviation, residual stress prediction and surface roughness of the component into the fourth optimization formula, output the target hot-dip galvanizing process parameters, which include the zinc pot working temperature, component zinc immersion time and component in / out of the zinc pot lifting rate; execute the hot-dip galvanizing operation of the component according to the target hot-dip galvanizing process parameters.

[0016] S7, Adaptive matching of passivation process and closed-loop feedback of coating quality: After the hot-dip galvanizing of the component is completed, the coating thickness distribution and coating surface adhesion detection data are collected. Based on the measured coating data, the passivation solution temperature and passivation immersion time are adaptively adjusted. The coating quality detection results are fed back to S2-S4, and the model boundary constraint parameters of the first optimization formula, the second optimization formula and the third optimization formula are iteratively updated to achieve closed-loop control of the entire process of forming-corrosion protection.

[0017] S8. The finished components are inspected and put into storage. The finished components of the transmission line tower angle steel or connecting plate that have passed the inspection are output.

[0018] Preferably, the original parameters of the incoming steel in S1 also include the carbon equivalent of the steel, and the carbon equivalent of the steel has a value range of 0.20%-0.42%; all input variables of the artificial intelligence control model with physical constraints are set with physical boundary thresholds for steel processing. When the input variables exceed the physical boundary thresholds for steel processing, the artificial intelligence control model with physical constraints triggers constraint truncation and prohibits the output of process parameters that exceed the physical limits of steel processing.

[0019] Preferably, the first optimization formula in S2 is the calculation formula for the process optimization of cutting heat-induced deformation constraint, and the expression is shown in equation (1):

[0020] (1);

[0021] In the formula: The optimized cutting speed is expressed in mm / min. This is the baseline cutting speed, expressed in mm / min. The coefficient of thermal expansion of steel; The unit is °C, which represents the local temperature rise of the board material under the action of the cutting heat source. This is the elastic modulus of steel, in MPa. The thickness of the incoming sheet material is in mm; The measured yield strength of the steel is expressed in MPa. This is the amount of compensation for the offset of the cutting heat source, in mm.

[0022] Preferably, the second optimization formula in S3 is a calculation formula for bending elastic-plastic rebound compensation, and the expression is shown in equation (2):

[0023] (2);

[0024] In the formula: The actual downward stroke of the bending head is in mm; This is the reference value for the downward stroke of the bending base, in mm; This is the springback correction factor, which is dimensionless. The thickness of the incoming sheet material is in mm; The measured yield strength of the steel is expressed in MPa. This is the elastic modulus of steel, in MPa. The normalized value of the cutting dimension deviation obtained from S2 is dimensionless.

[0025] Preferably, the third optimization formula in S4 is a welding heat input adaptive control calculation formula, and its expression is shown in equation (3):

[0026] (3);

[0027] In the formula: The welding heat input is per unit length, expressed in J / mm. The welding thermal efficiency is dimensionless. Welding voltage, unit: V; Welding current, in amperes (A). The welding walking speed is expressed in mm / s. The reference unit is the welding heat input per unit length, expressed in J / mm. The cutting deviation weighting coefficient is dimensionless. This is a dimensionless, normalized cutting dimension deviation. This is the bending deviation weighting coefficient, which is dimensionless. The normalized bending springback deviation is dimensionless. These are the material property weighting coefficients, dimensionless. The measured yield strength of the steel is expressed in MPa. This is the elastic modulus of steel, expressed in MPa.

[0028] Preferably, the fourth optimization formula in S6 is a calculation formula for the growth kinetics of hot-dip galvanized coating, and its expression is shown in equation (4):

[0029] (4);

[0030] In the formula: The actual zinc immersion time of the component, in seconds; The reference zinc immersion time is expressed in seconds. The actual thickness of the sheet metal used in the component is in mm. For reference sheet thickness, unit: mm; The residual stress influence weighting coefficient is dimensionless. The normalized residual stress estimate for the formed component is dimensionless. is a surface roughness weighting coefficient, dimensionless; The measured surface roughness of the component is expressed in μm. Let be the function of the effect of zinc pot temperature. The temperature of the zinc pot is in °C.

[0031] Preferably, in the hot-dip galvanizing pretreatment process of S6, the pickling agent is a hydrochloric acid pickling system, the hydrochloric acid mass fraction is controlled at 8%-14%, and the pickling temperature is 25-45℃; the fluxing agent is a zinc chloride-ammonium chloride compound fluxing agent, the Baume degree of the fluxing agent solution is controlled at 18-24°Bé, and the fluxing temperature is 60-75℃; the aluminum element mass fraction of the zinc liquid inside the zinc pot is 0.005%-0.020%.

[0032] Preferably, in S7, the passivation treatment uses chromium-based or chromium-free passivation agents, and the pH of the passivation solution is controlled between 3.0 and 5.0. When the average thickness of the coating is greater than 85 μm, the passivation immersion time is set to 8-12 s; when the average thickness of the coating is 55-85 μm, the passivation immersion time is set to 4-8 s; and when the average thickness of the coating is less than 55 μm, the passivation immersion time is set to 2-4 s.

[0033] The second aspect of the present invention provides an integrated intelligent molding and corrosion protection manufacturing system for transmission line tower components, used to implement the method of the first aspect, comprising:

[0034] The incoming material information acquisition and initial model initialization unit (101) is used to acquire incoming material information and initialize the initial model: acquire the original parameters of the incoming steel material of the transmission tower components to be processed, the original parameters of the incoming steel material include steel grade, plate thickness, plate initial yield strength, plate rolling residual stress distribution, plate surface roughness and component design drawing geometric parameters; input the original parameters of the incoming steel material into the artificial intelligence control model with physical constraints to complete the initial parameter assignment of the model;

[0035] The cutting process adaptive process parameter optimization and online size perception unit (102) is used to perform cutting process adaptive process parameter optimization and online size perception: based on the first optimization formula, the target process parameters of the cutting process are calculated in combination with the original parameters of the incoming steel. The target process parameters of the cutting process include cutting speed, cutting heat source power and cutting heat source offset compensation amount; the steel is cut and blanked using a cutting device. After the steel is cut and blanked, the actual measured values ​​of the cross-sectional contour and edge size of the cut component are collected by machine vision, and the cutting size deviation is calculated.

[0036] The bending process parameter dynamic correction and springback compensation unit (103) is used to perform dynamic correction and springback compensation of bending process parameters: the cutting size deviation, steel yield strength and plate thickness are input into the second optimization formula to solve the target process parameters of the bending process, wherein the target process parameters of the bending process include the opening distance of the lower bending die, the downward stroke of the bending pressure head and the bending holding time; the bending process is performed, and after the bending process is completed, the machine vision collects the actual measured values ​​of the bending component limb width and the bending inner angle to obtain the bending springback deviation;

[0037] The intelligent control unit for heat input and online correction unit for welding deformation in the welding forming process (104) is used to perform intelligent control for heat input and online correction for welding deformation in the welding forming process: the original parameters of the incoming steel of the component, the cutting size deviation and the bending springback deviation are input into the third optimization formula to calculate the target process parameters of the welding process, wherein the target process parameters of the welding process include welding current, welding voltage, welding travel speed and welding intermittent cooling time; the welding operation of angle steel and connecting plate assembly is performed, and the measured data of temperature field of welding area, flatness of component and relative position of bolt holes are collected in real time during the welding process to obtain welding deformation deviation, and the welding heat input parameters are finely adjusted in real time based on the welding deformation deviation;

[0038] The forming component comprehensive status information summary and output unit (105) is used to summarize and output the forming component comprehensive status information: summarize the cutting size deviation, the bending springback deviation and the welding deformation deviation, extract the estimated value of the component residual stress and the component surface morphology feature data, and generate the forming component comprehensive status data package. The forming component comprehensive status data package carries the component individual identifier and is transmitted to the downstream hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution and passivation process adaptive matching and coating quality closed-loop feedback process.

[0039] The hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving unit (106) is used to perform hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving: receiving the comprehensive status data packet of the formed component, completing the component degreasing, pickling and fluxing pretreatment processes; inputting the component plate thickness, comprehensive deformation deviation, residual stress prediction and surface roughness of the component into the fourth optimization formula, and outputting the hot-dip galvanizing target process parameters, which include the zinc pot working temperature, component zinc immersion time and component in / out of zinc pot lifting rate; and executing the hot-dip galvanizing operation of the component according to the hot-dip galvanizing target process parameters.

[0040] The passivation process adaptive matching and coating quality closed-loop feedback unit (107) is used to perform passivation process adaptive matching and coating quality closed-loop feedback: after the hot-dip galvanizing operation of the component is completed, the component coating thickness distribution and coating surface adhesion detection data are collected, and the passivation liquid temperature and passivation immersion time are adaptively adjusted according to the actual coating data; the coating quality detection results are back transmitted to the cutting process adaptive process parameter optimization and online size sensing unit (102), the bending process parameter dynamic correction and springback compensation unit (103), and the welding forming process heat input intelligent control and welding deformation online correction unit (104), and the model boundary constraint parameters of the first optimization formula, the second optimization formula, and the third optimization formula are iteratively updated to realize the closed-loop control of the entire forming-corrosion prevention process.

[0041] The third aspect of the present invention is to provide an intelligent forming and corrosion protection integrated preparation device for transmission line tower components, used to implement the method of the first aspect, including: a material parameter acquisition module, a cutting processing module, a bending processing module, a welding processing module, a machine vision online detection module, a forming data processing module, a hot-dip galvanizing treatment module, a coating quality detection module, a closed-loop iterative control module, and a component identification reading and writing module;

[0042] The incoming material parameter acquisition module is used to acquire the original parameters of the incoming steel to be processed and transmit them to the forming data processing module.

[0043] The cutting and processing module receives the target process parameters for the cutting process output by the forming data processing module to complete the steel cutting and blanking;

[0044] The bending processing module receives the target process parameters for the bending process output by the forming data processing module and completes the bending process.

[0045] The welding processing module receives the target process parameters for the welding process output by the forming data processing module and completes the assembly welding process;

[0046] The machine vision online inspection module is deployed at the output ends of the cutting module, the bending module and the welding module respectively, and collects the actual measurement data of the component size and transmits it back to the forming data processing module;

[0047] The forming data processing module has a built-in artificial intelligence control model with physical constraints, and includes the first optimization formula, the second optimization formula, the third optimization formula and the fourth optimization formula. It receives the raw parameters of the incoming material and the machine vision measured size data, and outputs the target process parameters for cutting, bending, welding and hot-dip galvanizing respectively. It generates a comprehensive status data package of the formed component and sends it to the hot-dip galvanizing processing module.

[0048] The hot-dip galvanizing module includes a pretreatment submodule, a zinc pot immersion galvanizing submodule, and a passivation submodule, which are respectively used to perform pretreatment, hot-dip galvanizing, and passivation operations after receiving the comprehensive status data packet of the formed component.

[0049] The coating quality detection module is used to collect coating quality detection data including measured data of coating thickness and coating adhesion, and transmit the coating quality detection data to the closed-loop iterative control module.

[0050] The closed-loop iterative control module feeds back the coating quality detection results to the forming data processing module, iteratively updates the physical boundary constraint parameters of the artificial intelligence control model, and completes the closed-loop control of the entire process.

[0051] The component identification reading and writing module is set at each process flow station. The component identification reading and writing module reads the individual component identification and binds the corresponding process detection data to the individual component identification, which follows the component flow throughout the entire production chain. The system internally sets up a data physical boundary verification unit, which is connected to the forming data processing module. When the output process parameters of the first optimization formula, the second optimization formula, the third optimization formula and the fourth optimization formula exceed the physical boundary threshold of steel processing, the data physical boundary verification unit outputs an interception signal to prevent the transmission of abnormal process parameters to the lower processing unit.

[0052] The beneficial effects of the method and system of the present invention are as follows:

[0053] First, this invention constructs four sets of optimized calculation formulas based on the physical mechanisms of steel thermal deformation, elastoplastic bending, welding heat conduction, and zinc plating growth kinetics. All calculation formulas are based on physical models of solid mechanics, heat conduction, and metal wetting growth. Unlike black-box algorithms of large language models, the model has clear physical interpretability. By setting physical boundary constraints on steel, it can avoid outputting abnormal process parameters that exceed the processing limits of steel. It is suitable for the alternating production conditions of multiple specifications of transmission tower components and can achieve adaptive solution of process parameters without training with massive production samples.

[0054] Secondly, this invention connects the data links between the cutting, bending, and welding forming processes. The dimensional deviation of the previous process is transmitted to the downstream process and participates in the solution of downstream process parameters, realizing deviation compensation between processes. With the help of online machine vision detection, the dimensional deviation correction is completed during the production process, changing the traditional offline manual correction mode after production, reducing the additional residual stress introduced by manual correction, improving the geometric processing accuracy of the tower angle steel and connecting plate components, ensuring the bolt hole position, leg width, and flatness index of the components, and reducing the difficulty of on-site assembly of the tower.

[0055] Third, this invention breaks down the barriers between the forming process and the hot-dip galvanizing and anti-corrosion process. It packages the component plate thickness, deformation deviation, residual stress prediction, and component surface roughness information obtained in the forming process and transmits them to the downstream galvanizing process. It solves the hot-dip galvanizing immersion time, zinc pot temperature, and passivation process parameters for each individual component, instead of using uniform fixed process parameters. This effectively improves the problems of insufficient galvanizing of thick-walled components and over-galvanizing of thin-walled components, improves the uniformity of the hot-dip galvanized anti-corrosion layer thickness, enhances the coating adhesion, and reduces the incidence of component omissions and coating peeling defects.

[0056] Fourth, this invention enables the reverse transmission of hot-dip galvanizing coating quality inspection results back to the forming end model, iteratively updating the physical constraint boundaries of the model, forming an intelligent closed-loop control of the entire forming-corrosion protection process. As production continues, the process model is continuously iterated and optimized, and the production quality is continuously improved.

[0057] Fifth, this invention limits the parameter ranges of pickling agents, fluxing agents, zinc bath components, and passivating agents in the pretreatment of hot-dip galvanizing. It also elucidates the technical mechanism behind the selection of raw material parameters, which not only ensures the quality of component forming and processing but also ensures the basic conditions for hot-dip galvanizing coating film formation. The finished iron tower components have high dimensional accuracy, controlled internal residual stress, excellent anti-corrosion performance of the coating, and improved overall structural strength and outdoor service life of the components.

[0058] Sixth, this invention achieves full lifecycle data traceability for single components by setting up a component identification reading and writing module and a data physical boundary verification unit for dual security constraints, and double-intercepts abnormal process parameters, thereby ensuring the production safety of power transmission towers, a key load-bearing component of the power grid, and preventing unqualified components from entering the power grid project. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the specific embodiments or related technologies of the present invention, the drawings used in the description of the specific embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0060] Figure 1 This is a flowchart of an integrated intelligent molding and corrosion protection method for transmission line tower components provided by an embodiment of the present invention.

[0061] Figure 2 This is a schematic diagram of the intelligent molding and corrosion protection integrated manufacturing system for transmission line tower components provided in an embodiment of the present invention. Detailed Implementation

[0062] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0064] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0065] like Figure 1 As shown, the first aspect of the present invention is to provide a method for the integrated intelligent molding and corrosion protection of transmission line tower components, comprising the following steps:

[0066] S1, Incoming Material Information Acquisition and Initial Model Initialization: Obtain the original parameters of the incoming steel material for the components of the transmission tower to be processed. The original parameters of the incoming steel material include steel grade, plate thickness, initial yield strength of the plate, residual stress distribution of the plate during rolling, surface roughness of the plate, and geometric parameters of the component design drawings; input the original parameters of the incoming steel material into the artificial intelligence control model with physical constraints to complete the initial parameter assignment of the model.

[0067] In this embodiment, step S1 is the data source for the entire production process. The raw materials for transmission tower components are mostly hot-rolled carbon structural steel or low-alloy high-strength structural steel. After the steel is rolled out of the steel mill, residual rolling stress will objectively exist inside the plate. The yield strength of different batches of incoming steel will fluctuate slightly, and the surface roughness of the plate will be affected by the hot-rolled and pickled incoming material conditions. If fixed process parameters are used directly, the fluctuation of the incoming material will directly cause inconsistent deformation in subsequent cutting, bending, and welding. This step collects the steel grade, plate thickness, initial yield strength of the plate, distribution of residual rolling stress of the plate, surface roughness of the plate, and geometric parameters of the component design drawings. These physical parameters are used as model inputs, instead of just inputting the dimensions of the component drawings. This completes the initialization of the artificial intelligence control model with physical constraints, embedding the constitutive physical boundary of the steel material into the initial conditions of the model, and avoiding the model outputting process parameters that exceed the physical processing limits of the steel.

[0068] S2, Adaptive process parameter optimization and online dimension perception for the cutting process: Based on the first optimization formula, the target process parameters for the cutting process are calculated in combination with the original parameters of the incoming steel. The target process parameters for the cutting process include cutting speed, cutting heat source power, and cutting heat source offset compensation. The steel is cut and blanked using a cutting device. After the steel is cut and blanked, the actual measured values ​​of the cross-sectional contour and edge dimensions of the cut component are collected by machine vision, and the cutting dimension deviation is calculated.

[0069] In this embodiment, step S2 involves adaptive process parameter optimization and online dimension sensing for the cutting process. The cutting process targets the angle steel billets and connecting plates for the tower. Plasma cutting or flame cutting is used. During cutting, a large amount of heat is input to the plate, causing localized thermal expansion and subsequent thermal contraction upon cooling. The residual rolling stress inherent in the plate is superimposed on the cutting thermal stress, resulting in deviations in the edge dimensions and contour of the component after cutting. The first optimization formula uses the steel heat conduction equation and the physical model of cutting thermal deformation as basic constraints, taking the thickness of the incoming plate, yield strength, and residual stress distribution as input variables to solve for the cutting speed, cutting heat source power, and cutting heat source offset compensation. The heat source offset compensation is used to pre-compensate for the contour deviation caused by cutting thermal contraction. After cutting, machine vision collects the component cross-sectional contour and edge dimensions online, comparing them with the theoretical dimensions on the drawings to obtain the cutting dimension deviation. This deviation is not used for simple scrapping but is transmitted to the downstream bending process as input for bending process parameter correction, realizing the transmission of upstream process deviations to downstream processes, rather than independent production by each process.

[0070] S3, Dynamic correction and springback compensation of bending process parameters: Input the cutting size deviation, steel yield strength and plate thickness into the second optimization formula to solve the target process parameters of the bending process, wherein the target process parameters of the bending process include the opening distance of the lower bending die, the downward stroke of the bending pressure head and the bending holding time; perform the bending process, and after the bending process is completed, the machine vision collects the actual measured values ​​of the bending component limb width and the bending inner angle to obtain the bending springback deviation;

[0071] In this embodiment, step S3 involves dynamic correction and springback compensation of bending process parameters. During the bending and forming of angle steel, the steel undergoes elastoplastic deformation. After the pressure head is unloaded, the elastically deformed portion springs back. The amount of springback is directly related to the plate thickness and the steel yield strength. Simultaneously, the edge dimension deviation of the component caused by the previous cutting process alters the actual stress state of the bending cross-section, further changing the actual springback value. Existing processes rely solely on plate thickness to determine the pressing stroke, ignoring the fluctuations in the incoming material yield strength and the impact of cutting deviations from the previous process. The second optimization formula uses the elastoplastic constitutive model of steel as the physical basis, taking the cutting dimension deviation, steel yield strength, and plate thickness as inputs to solve for the lower die opening distance, the pressing stroke of the bending pressure head, and the bending holding time. The pressing stroke is superimposed with springback compensation, and the holding time is used to release local bending stress and reduce residual stress after bending. After bending, machine vision collects the measured values ​​of the bending leg width and inner angle to obtain the bending springback deviation. This springback deviation is then used as process input in the welding process.

[0072] S4, Intelligent control of heat input and online correction of welding deformation in welding forming process: The original parameters of the incoming steel of the component, the cutting size deviation and the bending springback deviation are input into the third optimization formula to calculate the target process parameters of the welding process. The target process parameters of the welding process include welding current, welding voltage, welding travel speed and welding intermittent cooling time; the welding operation of angle steel and connecting plate assembly is performed. During the welding process, the measured data of temperature field of welding area, flatness of component and relative position of bolt holes are collected in real time to obtain welding deformation deviation, and the welding heat input parameters are finely adjusted in real time based on the welding deformation deviation.

[0073] In this embodiment, the S4 step of the welding forming process involves intelligent control of heat input and online correction of welding deformation. Angle steel and connecting plates are welded together to form a tower assembly. Welding heat input causes localized high temperatures, and the weld area's thermal expansion and contraction during cooling lead to welding deformation. Pre-cutting deviations and bending springback deviations alter the assembly gap, which in turn changes the degree of welding heat deformation under the same welding parameters. The third optimization formula couples with a welding heat conduction physical model, using component material parameters, cutting size deviations, and bending springback deviations as inputs to calculate welding current, welding voltage, welding travel speed, and welding intermittent cooling time. The intermittent cooling time is used to control cumulative heat input, preventing overall component overheating and large torsional deformation. During the welding process, infrared temperature sensors collect the temperature field of the welding area, and machine vision collects component flatness and bolt hole relative positions to obtain welding deformation deviations. The welding heat input is then fine-tuned in real time, completing online correction during the welding process, rather than offline correction after all welding is complete, reducing additional residual stress caused by manual correction.

[0074] S5, Comprehensive status information of formed components is summarized and output: The cutting size deviation, bending springback deviation and welding deformation deviation are summarized, the estimated value of residual stress of the component and the surface morphology feature data of the component are extracted, and a comprehensive status data package of the formed component is generated. The comprehensive status data package of the formed component carries the individual component identifier and is transmitted to the downstream hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution and passivation process adaptive matching and coating quality closed-loop feedback process.

[0075] In this embodiment, step S5 summarizes and outputs the comprehensive status information of the formed component. The three forming processes of cutting, bending, and welding generate dimensional deviation data respectively. Based on the physical model of steel deformation, the estimated value of residual stress of the component is derived from the dimensional deviation. At the same time, the surface morphology features of the component obtained by machine vision recognition are extracted to construct a comprehensive status data package of the formed component. Each component is bound to a unique individual component identifier. The data package follows the component and completely transmits the physical state of the component in the forming process to the hot-dip galvanizing process. This breaks down the data barrier between the forming process and the anti-corrosion process. In the prior art, forming and galvanizing are independent of each other. The galvanizing process cannot know the deformation and stress state of the component in the forming stage. This step solves this problem.

[0076] S6, Hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution: Receive the comprehensive status data packet of the formed component, complete the component degreasing, pickling and fluxing pretreatment processes; input the component plate thickness, comprehensive deformation deviation, residual stress prediction and surface roughness of the component into the fourth optimization formula, output the target hot-dip galvanizing process parameters, which include the zinc pot working temperature, component zinc immersion time and component in / out of the zinc pot lifting rate; execute the hot-dip galvanizing operation of the component according to the target hot-dip galvanizing process parameters.

[0077] In this embodiment, step S6 involves pretreatment for hot-dip galvanizing and AI-driven solution of hot-dip galvanizing process parameters. The component first undergoes conventional pretreatment including degreasing, pickling, and fluxing. This pretreatment removes oil and oxide scale from the steel surface, while the fluxing treatment forms a fluxing salt film on the component surface, ensuring effective zinc immersion. The fourth optimization formula is based on zinc immersion kinetics and a physical model of coating growth. Inputs include the component plate thickness, overall deformation deviation, estimated residual stress, and surface roughness. Outputs include the zinc pot operating temperature, immersion time, and inlet / outlet lifting rate. The component plate thickness determines the overall heat capacity; thick-walled components store more heat and require a longer immersion time to reach the zinc pot temperature. Residual stress from component forming alters the zinc immersion performance on the component surface, with higher residual stress areas showing different immersion behavior. Surface roughness directly affects the growth rate of the zinc-iron alloy layer. This step incorporates all individual component states obtained during the forming process into the galvanizing parameter calculation, eliminating the use of uniform fixed process parameters. This allows each component to be matched with specific hot-dip galvanizing process parameters for hot-dip galvanizing operations.

[0078] S7, Adaptive matching of passivation process and closed-loop feedback of coating quality: After the hot-dip galvanizing of the component is completed, the coating thickness distribution and coating surface adhesion detection data are collected. Based on the measured coating data, the passivation solution temperature and passivation immersion time are adaptively adjusted. The coating quality detection results are fed back to S2-S4, and the model boundary constraint parameters of the first optimization formula, the second optimization formula and the third optimization formula are iteratively updated to achieve closed-loop control of the entire process of forming-corrosion protection.

[0079] In this embodiment, the S7 step passivation process features adaptive matching and closed-loop feedback of coating quality. After hot-dip galvanizing, eddy current thickness measurement equipment collects coating thickness distribution data at different locations on the component, and adhesion detection obtains coating adhesion data. The coating thickness directly determines the reaction time required for passivation film growth. Based on the measured coating data, the passivation solution temperature and passivation immersion time are adjusted to adapt to the current coating state of the component. The coating quality detection results are fed back to the cutting, bending, and welding forming processes, iteratively updating the model boundary constraint parameters of the first, second, and third optimization formulas. This achieves reverse optimization of the front-end forming process by the back-end anti-corrosion quality results, forming a complete closed-loop control of forming and anti-corrosion. If localized incomplete coating or low coating adhesion occurs, the comprehensive status data package of the formed component is traced to determine whether it is caused by excessive residual stress or surface morphology defects in the forming process. The physical constraint boundaries of the model are then updated to reduce the probability of similar defects occurring in subsequent similar components.

[0080] S8. The finished components are inspected and put into storage. The finished components of the transmission line tower angle steel or connecting plate that have passed the inspection are output.

[0081] In this embodiment, step S8 completes the comprehensive inspection of the finished product, and the component enters the finished product warehouse to obtain the finished component of the transmission line tower angle steel or connecting plate. The finished component has high processing dimensional accuracy, the residual stress inside the component is controlled, the hot-dip galvanized coating is uniform, and the coating adhesion is excellent, making it suitable for on-site assembly and use of high-voltage transmission towers.

[0082] In a preferred embodiment, the raw steel parameters in S1 further include the carbon equivalent of the steel, which ranges from 0.20% to 0.42%. All input variables of the physically constrained artificial intelligence control model are set with physical boundary thresholds for steel processing. When an input variable exceeds these thresholds, the physically constrained artificial intelligence control model triggers constraint truncation, prohibiting the output of process parameters exceeding the physical limits of steel processing.

[0083] In this embodiment, the steps for collecting incoming material information and initializing the model in S1 are further defined. The carbon equivalent of the hot-rolled low-alloy structural steel used in transmission towers directly affects the steel's welding hardening tendency and hot working deformation behavior. If the carbon equivalent is too high, cold cracks are easily generated during welding, the material's brittleness increases after high-temperature cutting, and the springback behavior during bending also changes. Therefore, the carbon equivalent of the steel is included in the initial parameters of the incoming material. A carbon equivalent value of 0.20%-0.42% is the mainstream carbon equivalent range for steel structures used in domestic transmission towers. The carbon equivalent values ​​of Q235 and Q355 series steels used in towers both fall within this range. The AI ​​control model with physical constraints differs from the unconstrained black-box machine learning model. Each input variable is set with a corresponding physical boundary threshold for steel processing, such as the lower limit of plate thickness, the upper and lower limits of yield strength, and the upper limit of carbon equivalent. When the parameters obtained from incoming material detection exceed the boundary threshold, the model directly triggers constraint truncation and will not continue to calculate and output process parameters. This avoids outputting unreasonable process parameters when the incoming material is abnormal, prevents the production of steel tower load-bearing components with unqualified mechanical properties, and improves the production safety of the process system.

[0084] In a preferred embodiment, the first optimization formula in S2 is a calculation formula for the optimization of the cutting thermal deformation constraint process, and the expression is shown in equation (1):

[0085] (1);

[0086] In the formula: The optimized cutting speed is expressed in mm / min. This is the baseline cutting speed, expressed in mm / min. The coefficient of thermal expansion of steel; The unit is °C, which represents the local temperature rise of the board material under the action of the cutting heat source. This is the elastic modulus of steel, in MPa. The thickness of the incoming sheet material is in mm; The measured yield strength of the steel is expressed in MPa. This is the amount of compensation for the offset of the cutting heat source, in mm.

[0087] In this embodiment, the calculation formula (1) is constructed based on the physical model of steel thermal conduction-thermal deformation. It does not belong to the large language model type algorithm, but to the machine learning optimization calculation formula under the constraint of the physical model. During the cutting process, the plate heats up rapidly in a localized area, and the thermal expansion leads to subsequent cooling and contraction deformation. The greater the plate thickness and the higher the yield strength, the greater the dimensional deviation caused by the thermal deformation during cutting. In the formula The item characterizes the material's ability to resist heat-induced deformation, and the material thickness. Increased yield strength enhances resistance to deformation, allowing for a moderate increase in cutting speed; As the height increases, the material's resistance to deformation improves, and the cutting speed is adjusted accordingly. Characterizing the thermal expansion effect caused by the cutting heat source, the greater the temperature rise, the more necessary it is to reduce the cutting speed and add a heat source offset compensation amount. This formula pre-compensates for shrinkage after cooling. It incorporates inherent physical constants of steel, such as the elastic modulus and coefficient of thermal expansion, into the calculation. It uses plate thickness, yield strength, and cutting temperature rise as real-time input variables to obtain a cutting speed suitable for the current incoming material. At the same time, it outputs heat source offset compensation for the cutting equipment trajectory correction, reducing dimensional deviations of components after cutting and reducing the workload of subsequent process corrections.

[0088] In a preferred embodiment, the second optimization formula in S3 is a calculation formula for bending elastic-plastic springback compensation, and the expression is shown in equation (2):

[0089] (2);

[0090] In the formula: The actual downward stroke of the bending head is in mm; This is the reference value for the downward stroke of the bending base, in mm; This is the springback correction factor, which is dimensionless. The thickness of the incoming sheet material is in mm; The measured yield strength of the steel is expressed in MPa. This is the elastic modulus of steel, in MPa. The normalized value of the cutting dimension deviation obtained from S2 is dimensionless.

[0091] In this embodiment, the second artificial intelligence calculation formula in step S3 is defined. This formula is based on the physical model of the elastic-plastic bending deformation of steel and belongs to the artificial intelligence optimization calculation formula based on a physical foundation. The bending head's downward stroke is calculated by superimposing the basic downward stroke reference value with the springback compensation. The springback compensation is positively correlated with the plate thickness and yield strength, and negatively correlated with the elastic modulus. A normalized cutting dimension deviation is added. The deviation in the edge dimensions of the component caused by the previous cutting process will change the actual stress-bearing section state at the bending position and change the actual springback level of the component. Therefore, the cutting deviation is normalized and then introduced into the calculation of the downward stroke. When the cutting dimension deviation is positively out of tolerance, the effective stress-bearing width of the component section increases, the springback under the same downward pressure conditions increases, and the downward stroke increases accordingly; when the cutting dimension deviation is negatively out of tolerance, the springback decreases, and the downward stroke decreases appropriately. This calculation formula (2) incorporates the deviation of the previous process into the bending process solution, realizes the deviation transfer compensation between processes, reduces the springback deviation after bending, improves the accuracy of the bending leg width and bending inner angle dimensions of angle steel, reduces the subsequent correction workload of the bent component, and reduces the residual internal stress introduced by bending.

[0092] As a preferred embodiment, the third optimization formula in S4 is a welding heat input adaptive control calculation formula, and its expression is shown in equation (3):

[0093] (3);

[0094] In the formula: The welding heat input is per unit length, expressed in J / mm. The welding thermal efficiency is dimensionless. Welding voltage, unit: V; Welding current, in amperes (A). The welding walking speed is expressed in mm / s. The reference unit is the welding heat input per unit length, expressed in J / mm. The cutting deviation weighting coefficient is dimensionless. This is a dimensionless, normalized cutting dimension deviation. This is the bending deviation weighting coefficient, which is dimensionless. The normalized bending springback deviation is dimensionless. These are the material property weighting coefficients, dimensionless. The measured yield strength of the steel is expressed in MPa. This is the elastic modulus of steel, expressed in MPa.

[0095] In this embodiment, the third optimization formula (3) in step S4 is constructed based on the physical model of welding heat conduction. The welding heat input directly determines the weld fusion effect and the welding deformation. The left side of the equation is the classic physical definition of welding heat input, and the right side of the equation introduces cutting deviation, bending springback deviation, and the ratio of material yield strength to elastic modulus as correction terms. When cutting and bending result in a large gap between component assembly, the welding heat input per unit length is appropriately increased to ensure the weld fusion quality; when the component yield strength is high, the welding deformation stress is greater under the same heat input conditions, so the heat input is appropriately reduced to avoid excessive welding deformation. The model adjusts the welding current. Welding voltage Welding walking speed To achieve the target heat input Simultaneously, by coordinating the intermittent cooling time during welding, the cumulative heat on the components is controlled, and welding torsional deformation is suppressed. This reduces welding deformation deviation while ensuring the mechanical properties of the welded joint, thereby reducing issues such as flatness and hole size deviations in the welded components, and lowering residual welding stress.

[0096] As a preferred embodiment, the fourth optimization formula in S6 is a calculation formula for the growth kinetics of hot-dip galvanized coating, and the expression is shown in equation (4):

[0097] (4);

[0098] In the formula: The actual zinc immersion time of the component, in seconds; The reference zinc immersion time is expressed in seconds. The actual thickness of the sheet metal used in the component is in mm. For reference sheet thickness, unit: mm; The residual stress influence weighting coefficient is dimensionless. The normalized residual stress estimate for the formed component is dimensionless. is a surface roughness weighting coefficient, dimensionless; The measured surface roughness of the component is expressed in μm. Let be the function of the effect of zinc pot temperature. The temperature of the zinc pot is in °C.

[0099] In this embodiment, the fourth AI calculation formula in step S6 is based on a physical model of zinc-iron alloy layer growth kinetics. The zinc immersion time is a key process parameter in hot-dip galvanizing. The thickness of the component sheet determines its heat capacity; thicker plates require a longer immersion time to allow the entire component to reach the zinc bath temperature. Residual stress from the component forming process alters the zinc bath wetting kinetics of the steel surface; increased residual stress changes the zinc-iron alloy layer growth rate. The surface roughness Ra of the component directly affects the zinc-iron alloy nucleation and growth process. The zinc pot temperature... Through temperature effect function A coupling correction is applied to the zinc immersion time. This formula incorporates the estimated residual stress and surface roughness from the forming process into the zinc immersion time calculation, instead of relying solely on the plate thickness to determine the immersion time. This allows for matching the immersion time to the individual condition of each formed component, reducing defects such as over-galvanization of thin-walled components and under-galvanization of thick-walled components, improving coating thickness uniformity, and enhancing coating adhesion.

[0100] In a preferred embodiment, in the hot-dip galvanizing pretreatment process of S6, the pickling agent is a hydrochloric acid pickling system, the mass fraction of hydrochloric acid is controlled at 8%-14%, and the pickling temperature is 25-45℃; the fluxing agent is a zinc chloride-ammonium chloride compound fluxing agent, the Baume degree of the fluxing agent solution is controlled at 18-24°Bé, and the fluxing temperature is 60-75℃; the mass fraction of aluminum element in the zinc liquid inside the zinc pot is 0.005%-0.020%.

[0101] In this embodiment, the parameters of the S6 hot-dip galvanizing pretreatment and zinc bath raw material composition are described. The technical reasons for the material selection include: the use of a hydrochloric acid pickling system, which dissolves iron oxide scale on the steel surface rapidly. Within the temperature range of 25-45℃, it can efficiently remove oxide scale and rust layers generated during hot rolling, while simultaneously controlling the over-corrosion rate of the steel substrate. If the hydrochloric acid mass fraction is below 8%, the oxide scale removal rate is too low, leading to decreased production efficiency. If the hydrochloric acid mass fraction is above 14%, the risk of over-corrosion of the substrate increases significantly, resulting in over-corrosion pitting on the component surface. This pitting will cause raised defects in the subsequent galvanized coating, reducing coating adhesion. Pickling temperatures below 25℃ result in a slow pickling reaction rate, while temperatures above 45℃ exacerbate hydrochloric acid volatilization, worsening the production environment and simultaneously accelerating steel substrate corrosion. The fluxing agent used is a zinc chloride-ammonium chloride compound fluxing agent. This compound system can form a uniform fluxing salt film on the surface of steel components. When the components are immersed in the zinc pot, the fluxing salt film quickly decomposes and breaks down the trace oxide layer on the surface of the components, ensuring that the zinc liquid fully wets the steel substrate. If the Baume degree of the fluxing agent is lower than 18°Bé, the fluxing salt film thickness is insufficient, the ability to break down the oxide film is insufficient, and the incomplete plating defect is easy to occur. If the Baume degree is higher than 24°Bé, the fluxing salt film is too thick, the salt film decomposes incompletely, and the coating surface is prone to residual particle defects. The fluxing temperature of 60-75℃ ensures the quality of the salt film drying and film formation. Adding aluminum to the zinc bath allows it to preferentially form a dense oxide film at the zinc-air interface, reducing zinc oxidation and burn-off. Simultaneously, trace amounts of aluminum participate in the zinc-iron alloy layer interface reaction, improving the coating's plasticity and adhesion. When the aluminum mass fraction is below 0.005%, the improvement in coating effect is minimal; when the aluminum mass fraction is above 0.020%, aluminum preferentially reacts with chlorine in the flux, inhibiting zinc wetting and increasing the risk of incomplete plating. Adopting the above raw material parameter range ensures pretreatment quality, providing a foundation for the formation of a high-quality hot-dip galvanized coating.

[0102] In a preferred embodiment, the passivation treatment in S7 uses a chromium-based passivation agent or a chromium-free passivation agent, and the pH of the passivation solution is controlled between 3.0 and 5.0. When the average thickness of the coating is greater than 85 μm, the passivation immersion time is set to 8-12 s; when the average thickness of the coating is 55-85 μm, the passivation immersion time is set to 4-8 s; and when the average thickness of the coating is less than 55 μm, the passivation immersion time is set to 2-4 s.

[0103] In this embodiment, the S7 passivation treatment agent and process parameters are specified. The purpose of passivation treatment is to form a dense passivation film on the surface of the hot-dip galvanized coating, isolating it from air and moisture, and improving the atmospheric corrosion resistance of the galvanized components. The pH of the passivation solution is controlled between 3.0 and 5.0. If the pH is too low, the passivation solution is too acidic, which will cause excessive dissolution of the zinc coating substrate, resulting in a thinner coating; if the pH is too high, the passivation film formation reaction is difficult to proceed, resulting in a thin and incomplete passivation film, and a decrease in the anti-corrosion effect. The passivation immersion time is matched with the average coating thickness. The thicker the coating, the longer the immersion time is required to complete the passivation film growth; when the coating is thin, the immersion time is shortened to avoid excessive dissolution of the coating. The immersion time is adaptively adjusted according to the measured coating thickness to achieve stable passivation film quality and improve the outdoor weather resistance and corrosion resistance of the tower components.

[0104] like Figure 2 As shown, a second aspect of the present invention provides an integrated intelligent molding and corrosion protection manufacturing system for transmission line tower components, used to implement the method of the first aspect, comprising:

[0105] The incoming material information acquisition and initial model initialization unit (101) is used to acquire incoming material information and initialize the initial model: acquire the original parameters of the incoming steel material of the transmission tower components to be processed, the original parameters of the incoming steel material include steel grade, plate thickness, plate initial yield strength, plate rolling residual stress distribution, plate surface roughness and component design drawing geometric parameters; input the original parameters of the incoming steel material into the artificial intelligence control model with physical constraints to complete the initial parameter assignment of the model;

[0106] The cutting process adaptive process parameter optimization and online size perception unit (102) is used to perform cutting process adaptive process parameter optimization and online size perception: based on the first optimization formula, the target process parameters of the cutting process are calculated in combination with the original parameters of the incoming steel. The target process parameters of the cutting process include cutting speed, cutting heat source power and cutting heat source offset compensation amount; the steel is cut and blanked using a cutting device. After the steel is cut and blanked, the actual measured values ​​of the cross-sectional contour and edge size of the cut component are collected by machine vision, and the cutting size deviation is calculated.

[0107] The bending process parameter dynamic correction and springback compensation unit (103) is used to perform dynamic correction and springback compensation of bending process parameters: the cutting size deviation, steel yield strength and plate thickness are input into the second optimization formula to solve the target process parameters of the bending process, wherein the target process parameters of the bending process include the opening distance of the lower bending die, the downward stroke of the bending pressure head and the bending holding time; the bending process is performed, and after the bending process is completed, the machine vision collects the actual measured values ​​of the bending component limb width and the bending inner angle to obtain the bending springback deviation;

[0108] The intelligent control unit for heat input and online correction unit for welding deformation in the welding forming process (104) is used to perform intelligent control for heat input and online correction for welding deformation in the welding forming process: the original parameters of the incoming steel of the component, the cutting size deviation and the bending springback deviation are input into the third optimization formula to calculate the target process parameters of the welding process, wherein the target process parameters of the welding process include welding current, welding voltage, welding travel speed and welding intermittent cooling time; the welding operation of angle steel and connecting plate assembly is performed, and the measured data of temperature field of welding area, flatness of component and relative position of bolt holes are collected in real time during the welding process to obtain welding deformation deviation, and the welding heat input parameters are finely adjusted in real time based on the welding deformation deviation;

[0109] The forming component comprehensive status information summary and output unit (105) is used to summarize and output the forming component comprehensive status information: summarize the cutting size deviation, the bending springback deviation and the welding deformation deviation, extract the estimated value of the component residual stress and the component surface morphology feature data, and generate the forming component comprehensive status data package. The forming component comprehensive status data package carries the component individual identifier and is transmitted to the downstream hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution and passivation process adaptive matching and coating quality closed-loop feedback process.

[0110] The hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving unit (106) is used to perform hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving: receiving the comprehensive status data packet of the formed component, completing the component degreasing, pickling and fluxing pretreatment processes; inputting the component plate thickness, comprehensive deformation deviation, residual stress prediction and surface roughness of the component into the fourth optimization formula, and outputting the hot-dip galvanizing target process parameters, which include the zinc pot working temperature, component zinc immersion time and component in / out of zinc pot lifting rate; and executing the hot-dip galvanizing operation of the component according to the hot-dip galvanizing target process parameters.

[0111] The passivation process adaptive matching and coating quality closed-loop feedback unit (107) is used to perform passivation process adaptive matching and coating quality closed-loop feedback: after the hot-dip galvanizing operation of the component is completed, the component coating thickness distribution and coating surface adhesion detection data are collected, and the passivation liquid temperature and passivation immersion time are adaptively adjusted according to the actual coating data; the coating quality detection results are back transmitted to the cutting process adaptive process parameter optimization and online size sensing unit (102), the bending process parameter dynamic correction and springback compensation unit (103), and the welding forming process heat input intelligent control and welding deformation online correction unit (104), and the model boundary constraint parameters of the first optimization formula, the second optimization formula, and the third optimization formula are iteratively updated to realize the closed-loop control of the entire forming-corrosion prevention process.

[0112] The third aspect of the present invention is to provide an intelligent forming and corrosion protection integrated preparation device for transmission line tower components, used to implement the method of the first aspect, including: a material parameter acquisition module, a cutting processing module, a bending processing module, a welding processing module, a machine vision online detection module, a forming data processing module, a hot-dip galvanizing treatment module, a coating quality detection module, a closed-loop iterative control module, and a component identification reading and writing module;

[0113] The incoming material parameter acquisition module is used to acquire the original parameters of the incoming steel to be processed and transmit them to the forming data processing module.

[0114] The cutting and processing module receives the target process parameters for the cutting process output by the forming data processing module to complete the steel cutting and blanking;

[0115] The bending processing module receives the target process parameters for the bending process output by the forming data processing module and completes the bending process.

[0116] The welding processing module receives the target process parameters for the welding process output by the forming data processing module and completes the assembly welding process;

[0117] The machine vision online inspection module is deployed at the output ends of the cutting module, the bending module and the welding module respectively, and collects the actual measurement data of the component size and transmits it back to the forming data processing module;

[0118] The forming data processing module has a built-in artificial intelligence control model with physical constraints, and includes the first optimization formula, the second optimization formula, the third optimization formula and the fourth optimization formula. It receives the raw parameters of the incoming material and the machine vision measured size data, and outputs the target process parameters for cutting, bending, welding and hot-dip galvanizing respectively. It generates a comprehensive status data package of the formed component and sends it to the hot-dip galvanizing processing module.

[0119] The hot-dip galvanizing module includes a pretreatment submodule, a zinc pot immersion galvanizing submodule, and a passivation submodule, which are respectively used to perform pretreatment, hot-dip galvanizing, and passivation operations after receiving the comprehensive status data packet of the formed component.

[0120] The coating quality detection module is used to collect coating quality detection data including measured data of coating thickness and coating adhesion, and transmit the coating quality detection data to the closed-loop iterative control module.

[0121] The closed-loop iterative control module feeds back the coating quality detection results to the forming data processing module, iteratively updates the physical boundary constraint parameters of the artificial intelligence control model, and completes the closed-loop control of the entire process.

[0122] The component identification reading and writing module is set at each process flow station. The component identification reading and writing module reads the individual component identification and binds the corresponding process detection data to the individual component identification, which follows the component flow throughout the entire production chain. The system internally sets up a data physical boundary verification unit, which is connected to the forming data processing module. When the output process parameters of the first optimization formula, the second optimization formula, the third optimization formula and the fourth optimization formula exceed the physical boundary threshold of steel processing, the data physical boundary verification unit outputs an interception signal to prevent the transmission of abnormal process parameters to the lower processing unit.

[0123] In this embodiment, the incoming material parameter acquisition module may include a spectral composition detection device, a plate thickness detection device, a residual stress detection device, and a surface roughness detector to acquire information such as steel grade, carbon equivalent, yield strength, residual stress, plate thickness, and surface roughness. The cutting processing unit includes plasma / flame cutting equipment; the bending processing unit includes CNC bending equipment; the welding processing unit includes automatic welding equipment; and the machine vision online inspection module includes multiple sets of industrial cameras and light sources, respectively arranged at the post-cutting, post-bending, and post-welding workstations, to achieve online inspection of components without disassembly, eliminating the need for offline sampling. The system comprises several modules. The forming data processing module, as the core computing unit, incorporates four sets of AI calculation formulas with physical model bases to solve for process parameters at each stage, generating a comprehensive status data package of the formed component carrying individual component identifiers. The hot-dip galvanizing unit is divided into a pretreatment subunit, a zinc pot immersion subunit, and a passivation subunit, receiving data packages to implement component-specific galvanizing process calls. The coating quality inspection module includes an eddy current coating thickness gauge and coating adhesion testing fixtures. The closed-loop iterative control module receives coating quality data and iterates backward through the forming end model constraint parameters to achieve a closed-loop process across forming and corrosion prevention. All the hardware units of the entire system work together to achieve integrated intelligent manufacturing of tower components from raw materials to finished products. The component identification reading and writing module can adopt an RFID radio frequency identification reading and writing device. Each tower component is bound with an RFID tag as an individual component identifier. When the component flows through each station of cutting, bending, welding, and hot-dip galvanizing, the component identification reading and writing module reads the tag and binds the detection data of this station to the corresponding component tag, realizing full-process data traceability for a single component. If coating defects or dimensional defects occur in the finished component, the original parameters and detection data of all processes can be traced back, which facilitates process iteration and optimization. The data physical boundary verification unit is set independently and does not rely on the constraints of the artificial intelligence calculation formula itself. It performs secondary hardware-level parameter interception. Once the process parameters output by the calculation formula exceed the physical processing boundary threshold of the steel, they are directly intercepted and will not be sent to the cutting, bending, welding, and galvanizing equipment. This provides double protection for the safety of the production process of key power components and avoids the production of unqualified tower load-bearing components due to abnormal process parameters.

[0124] Application Example 1

[0125] This embodiment manufactures Q355B angle steel components for high-voltage transmission towers. The angle steel specifications are ∠140×12, the designed leg width is 140mm, the plate thickness is 12mm, the steel grade is Q355B, the measured yield strength is 365MPa, the carbon equivalent is 0.34%, the elastic modulus is E=206000MPa, and the coefficient of thermal expansion is... The surface roughness of the incoming sheet material is Ra 18.2 μm.

[0126] S1, Incoming Material Information Acquisition and Model Initialization: The incoming material parameter acquisition module collects the steel grade Q355B, plate thickness 12mm, yield strength 365MPa, carbon equivalent 0.34%, rolling residual stress distribution, surface roughness Ra18.2μm, and component drawing geometric parameters; all original incoming material parameters are input into the forming data processing module's internal artificial intelligence control model with physical constraints to complete model initialization, and all input variables are verified for physical boundaries.

[0127] S2, Adaptive process parameter optimization and online dimension awareness for the cutting process: Calls the first optimization formula, the process optimization calculation formula for thermally induced deformation constraints, and the basic cutting speed reference value. Local temperature rise of the board under the action of the cutting heat source =720℃, substituting into the formula, the optimized cutting speed is obtained. Cutting heat source offset compensation amount The cutting and processing unit uses plasma cutting equipment to cut and blank angle steel billets. A machine vision online inspection module collects the edge contour dimensions of the cut components, compares them with the theoretical dimensions on the drawings, and calculates the normalized cutting dimension deviation. .

[0128] S3, Dynamic Correction and Springback Compensation of Bending Process Parameters: This involves calling the second optimized formula, the calculation formula for bending elastic-plastic springback compensation, and the reference value of the bending base compression stroke. springback correction coefficient The sheet material has a thickness of 12mm, a yield strength of 365MPa, an elastic modulus of 206000MPa, and a normalized cutting dimensional deviation. Substituting the values ​​into the calculation, we obtain the actual downward stroke of the bending indenter. The opening distance of the lower bending die is set to 120mm, and the bending pressure holding time is 14s. The CNC bending equipment performs angle steel bending. After processing, machine vision collects measured data of the bent leg width and inner angle to obtain the normalized bending springback deviation. .

[0129] S4, Intelligent control of welding heat input and online correction of welding deformation in the welding forming process: This utilizes the third optimized formula, the adaptive control calculation formula for welding heat input, and the baseline unit length welding heat input... Welding thermal efficiency Cutting deviation weighting coefficient Bending deviation weighting coefficient Material property weighting coefficient Substitute the normalized cutting dimension deviation. Normalized bending springback deviation With a yield strength of 365 MPa and an elastic modulus of 206,000 MPa, the calculated welding heat input per unit length is... The welding current was matched to 275A, the welding voltage to 31V, the welding travel speed to 5.7mm / s, and the inter-welding cooling time to 2.2s. The automatic welding equipment performed the welding operation of the angle steel auxiliary connecting plate assembly; during the welding process, infrared temperature field acquisition and machine vision real-time acquisition of component flatness and bolt hole relative positions were used; the heat input parameters were fine-tuned during the welding process to suppress welding deformation; and the welding deformation deviation was obtained after the welding was completed.

[0130] S5, Summary Output of Comprehensive Status Information for Molded Components: Summarizes cutting dimension deviation, bending springback deviation, and welding deformation deviation, and derives the normalized residual stress estimate of the molded component. The surface roughness of the component is 18.2μm; a comprehensive status data packet of the formed component is generated, and an RFID component individual identifier is bound. The data packet is transmitted to the downstream hot-dip galvanizing treatment unit.

[0131] S6, Hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter calculation: Components are transferred to the hot-dip galvanizing unit. The pretreatment subunit performs degreasing, pickling, and fluxing processes. Pickling uses a hydrochloric acid pickling system with a hydrochloric acid mass fraction of 11% and a pickling temperature of 33℃. The fluxing agent uses a zinc chloride-ammonium chloride compound flux with a Baume degree of 21°Bé and a fluxing temperature of 68℃. The zinc bath contains 0.012% aluminum. The fourth artificial intelligence calculation formula for hot-dip galvanizing coating growth kinetics process matching calculation formula is called; the baseline immersion time is used. Reference plate thickness The influence of residual stress on the weighting factor Surface roughness weighting coefficient Zinc pot working temperature Temperature effect function Substituting the thickness of the component plate as 12mm, the normalized residual stress estimate is obtained. The surface roughness of the component is Ra18.2μm; the actual zinc immersion time of the component was calculated. The lifting speed of components entering and exiting the zinc pot is set to 1.6 m / min; the zinc pot immersion subunit performs hot-dip galvanizing operations.

[0132] S7, Adaptive matching of passivation process and closed-loop feedback of coating quality: After hot-dip galvanizing is completed, the average coating thickness of the coating measured by the eddy current thickness measurement component of the coating quality detection module is 92μm, and the coating adhesion test is qualified; when the average coating thickness is greater than 85μm, a chromium-free passivation agent is selected, the pH of the passivation solution is controlled at 4.1, the passivation immersion time is set to 10s, and the passivation subunit performs passivation treatment; the coating quality detection data is transmitted to the closed-loop iterative control module, and then transmitted back to the forming data processing module to complete the model parameter iteration.

[0133] S8, finished product inspection and warehousing, yielding Q355B∠140×12 transmission tower angle steel finished components.

[0134] After the components in this embodiment were formed, the actual measurements showed that: the width deviation of the angle steel leg was ≤ ±0.45mm, the deviation of the inner angle of the bend was ≤ ±0.22°, and the flatness deviation of the component after welding was ≤ 0.7mm / m; the average thickness of the hot-dip galvanized coating was 92μm, with small coating thickness dispersion, and no peeling or flaking was observed in the coating adhesion cross-cut test; the residual stress level of the component was significantly lower than that of components produced by traditional fixed processes; the component underwent accelerated salt spray corrosion testing, and no red rust appeared after a salt spray test duration of more than 1350h; the overall processing accuracy and corrosion resistance of the component were excellent, meeting the technical standard requirements for high-voltage transmission tower components.

[0135] Reference Example 2

[0136] This embodiment describes the fabrication of a connecting plate component for an ultra-high voltage transmission tower. The connecting plate is 6mm thick, made of Q235B steel with a measured yield strength of 258MPa, a carbon equivalent of 0.27%, an elastic modulus E of 206000MPa, and a coefficient of thermal expansion of [missing information]. The surface roughness of the incoming sheet material is Ra13.7μm.

[0137] S1, Incoming Material Information Acquisition and Model Initialization: The incoming material parameter acquisition module acquires the steel grade Q235B, plate thickness 6mm, yield strength 258MPa, carbon equivalent 0.27%, rolling residual stress distribution, surface roughness Ra13.7μm, and geometric parameters of the connecting plate drawing; all original incoming material parameters are input into the forming data processing module to complete the initialization of the artificial intelligence control model with physical constraints, and the input variables are used to perform physical boundary verification.

[0138] S2, Adaptive process parameter optimization and online dimension sensing for the cutting process: Calls the first optimization formula, the baseline cutting speed value. Local temperature rise of the board under the action of the cutting heat source =660℃, substituting this into the calculation yields the optimized cutting speed. Cutting heat source offset compensation amount The plasma cutting equipment completes the cutting and blanking of the connecting plate material; the machine vision online inspection module collects the contour dimensions after cutting and calculates the normalized cutting dimension deviation. .

[0139] S3, Dynamic Correction and Springback Compensation of Bending Process Parameters: Calls the second optimization formula, and sets the baseline value for the bending base pressure stroke. springback correction coefficient Plate thickness 6mm, yield strength 258MPa, elastic modulus 206000MPa, normalized cutting dimensional deviation. Substituting into the calculation, the actual downward stroke of the bending head. The lower die opening distance is 72mm, and the bending holding time is 9s. After the bending process is completed, the bending dimensions are collected by machine vision to obtain the normalized bending springback deviation. .

[0140] S4, Intelligent control of heat input for welding forming process and online correction of welding deformation: Calling the third optimization formula, the baseline unit length welding heat input... Welding thermal efficiency Cutting deviation weighting coefficient Bending deviation weighting coefficient Material property weighting coefficient Substitute the normalized cutting dimension deviation. Normalized bending springback deviation Given a yield strength of 258 MPa and an elastic modulus of 206,000 MPa, calculate the target unit length welding heat input. The welding current is 220A, the welding voltage is 28V, the welding travel speed is 6.4mm / s, and the welding intermittent cooling time is 1.5s. The automatic welding equipment performs the welding operation of the connecting plate assembly. During the welding process, machine vision and infrared sensing collect the component deformation and temperature field data in real time, fine-tune the welding heat input parameters, control the welding deformation, and obtain the welding deformation deviation.

[0141] S5, Summary Output of Comprehensive Status Information for Molded Components: Summarizes cutting dimension deviation, bending springback deviation, and welding deformation deviation, and extrapolates the normalized residual stress estimate of the molded component. The surface roughness of the component is Ra13.7μm; a comprehensive status data packet of the formed component is generated, an RFID component individual identifier is bound, and the packet is sent to the downstream hot-dip galvanizing unit.

[0142] S6, Hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter calculation: The hot-dip galvanizing pretreatment subunit performs degreasing, pickling, and fluxing; hydrochloric acid pickling system, hydrochloric acid mass fraction 9%, pickling temperature 29℃; zinc chloride-ammonium chloride compound flux, flux Baume degree 20°Bé, fluxing temperature 66℃; zinc pot zinc liquid aluminum element mass fraction 0.011%. The fourth artificial intelligence calculation formula is called, and the baseline immersion time is determined. Reference plate thickness The influence of residual stress on the weighting factor Surface roughness weighting coefficient Zinc pot working temperature Temperature effect function Substituting the component plate thickness of 6mm, the normalized residual stress estimate is calculated. The surface roughness of the component is Ra13.7μm; the actual zinc immersion time of the component was calculated. The lifting speed of components entering and exiting the zinc pot is set to 1.8 m / min; the zinc pot immersion subunit performs hot-dip galvanizing operations.

[0143] S7, Adaptive matching of passivation process and closed-loop feedback of coating quality: The coating quality detection module detects that the average coating thickness of the connecting plate is 68μm, and the coating adhesion test is qualified; the average coating thickness is in the range of 55-85μm, chromium-based passivation agent is selected, the passivation solution pH is 3.8, and the passivation immersion time is set to 6s; the passivation subunit completes passivation; the coating quality detection data transmission closed-loop iterative control module sends back the forming data processing module iterative model constraint parameters.

[0144] S8, finished product inspection and warehousing, yielding the finished Q235B transmission tower connection plate component.

[0145] Actual measurements of the component forming in this embodiment show that: the cutting dimension deviation of the connecting plate is ≤ ±0.38mm, the bending dimension deviation is ≤ ±0.20°, and the flatness deviation after welding is ≤ 0.6mm / m; the average thickness of the hot-dip galvanized coating is 68μm, the coating thickness is uniformly distributed, and there is no peeling or flaking in the cross-cut adhesion test; no red rust appears in the salt spray accelerated corrosion test with a duration of more than 1220h; the overall dimensional accuracy of the component is high, and the coating quality is stable, meeting the production technical requirements of the connecting plate component for ultra-high voltage transmission towers.

[0146] Compared to traditional fixed-parameter production methods, the tower angle steel and connecting plate components produced in Application Examples 1 and 2 of this invention utilize a steel physical constitutive model for the cutting, bending, and welding processes. Material fluctuations and dimensional deviations from previous processes are incorporated into the process parameter calculations. Online sensing of component dimensional deviations and inter-process deviation compensation are implemented, significantly improving the geometric processing accuracy of the components. Residual stress after component forming is effectively suppressed, reducing the risk of stress corrosion during later service. The comprehensive state data of the components from the forming process is transferred to the hot-dip galvanizing process. Based on a coating growth kinetic model, the galvanizing and passivation process parameters are matched, significantly reducing coating thickness dispersion and improving coating adhesion. The back-end coating quality inspection results iterate against the front-end forming model constraint parameters, achieving integrated forming-corrosion protection closed-loop production. The entire production process does not require large-scale black-box sample training. The model possesses a clear physical mechanism, and the process parameter output is constrained by the physical boundaries of the steel. This is suitable for the alternating production of multiple specifications and small batches of power towers. The resulting tower components have high processing accuracy, stable structural strength, and extended service life in outdoor applications.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for the integrated intelligent molding and corrosion protection of transmission line tower components, comprising the following steps: S1, Incoming Material Information Acquisition and Initial Model Initialization: Obtain the original parameters of the incoming steel material for the components of the transmission tower to be processed. The original parameters of the incoming steel material include steel grade, plate thickness, initial yield strength of the plate, residual stress distribution of the plate during rolling, surface roughness of the plate, and geometric parameters of the component design drawings; input the original parameters of the incoming steel material into the artificial intelligence control model with physical constraints to complete the initial parameter assignment of the model. S2, Adaptive process parameter optimization and online size perception of the cutting process: Based on the first optimization formula, the target process parameters of the cutting process are calculated in combination with the original parameters of the incoming steel. The target process parameters of the cutting process include cutting speed, cutting heat source power and cutting heat source offset compensation amount. The steel is cut and blanked using a cutting device. After the steel is cut and blanked, the actual measured values ​​of the cross-sectional contour and edge dimensions of the cut component are collected by machine vision, and the cutting dimension deviation is calculated. S3, Dynamic correction and springback compensation of bending process parameters: Input the cutting size deviation, steel yield strength and plate thickness into the second optimization formula to solve the target process parameters of the bending process, wherein the target process parameters of the bending process include the opening distance of the lower bending die, the downward stroke of the bending pressure head and the bending holding time; perform the bending process, and after the bending process is completed, the machine vision collects the actual measured values ​​of the bending component limb width and the bending inner angle to obtain the bending springback deviation; S4, Intelligent control of heat input and online correction of welding deformation in welding forming process: The original parameters of the incoming steel of the component, the cutting size deviation and the bending springback deviation are input into the third optimization formula to calculate the target process parameters of the welding process. The target process parameters of the welding process include welding current, welding voltage, welding travel speed and welding intermittent cooling time; the welding operation of angle steel and connecting plate assembly is performed. During the welding process, the measured data of temperature field of welding area, flatness of component and relative position of bolt holes are collected in real time to obtain welding deformation deviation, and the welding heat input parameters are finely adjusted in real time based on the welding deformation deviation. S5, Comprehensive status information of formed components is summarized and output: The cutting size deviation, bending springback deviation and welding deformation deviation are summarized, the estimated value of residual stress of the component and the surface morphology feature data of the component are extracted, and a comprehensive status data package of the formed component is generated. The comprehensive status data package of the formed component carries the individual component identifier and is transmitted to the downstream hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution and passivation process adaptive matching and coating quality closed-loop feedback process. S6, Hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution: Receive the comprehensive status data packet of the formed component, complete the component degreasing, pickling and fluxing pretreatment processes; input the component plate thickness, comprehensive deformation deviation, residual stress prediction and surface roughness of the component into the fourth optimization formula, output the target hot-dip galvanizing process parameters, which include the zinc pot working temperature, component zinc immersion time and component in / out of the zinc pot lifting rate; execute the hot-dip galvanizing operation of the component according to the target hot-dip galvanizing process parameters. S7, Adaptive matching of passivation process and closed-loop feedback of coating quality: After the hot-dip galvanizing of the component is completed, the coating thickness distribution and coating surface adhesion detection data are collected. Based on the measured coating data, the passivation solution temperature and passivation immersion time are adaptively adjusted. The coating quality detection results are fed back to S2-S4, and the model boundary constraint parameters of the first optimization formula, the second optimization formula and the third optimization formula are iteratively updated to achieve closed-loop control of the entire process of forming-corrosion protection. S8. The finished components are inspected and put into storage. The finished components of the transmission line tower angle steel or connecting plate that have passed the inspection are output. Preferably, the original parameters of the incoming steel in S1 also include the carbon equivalent of the steel, and the carbon equivalent of the steel has a value range of 0.20%-0.42%; all input variables of the artificial intelligence control model with physical constraints are set with physical boundary thresholds for steel processing. When the input variables exceed the physical boundary thresholds for steel processing, the artificial intelligence control model with physical constraints triggers constraint truncation and prohibits the output of process parameters that exceed the physical limits of steel processing. Preferably, the first optimization formula in S2 is the calculation formula for the process optimization of cutting heat-induced deformation constraint, and the expression is shown in equation (1): （1）； In the formula: The optimized cutting speed is expressed in mm / min. This is the baseline cutting speed, expressed in mm / min. The coefficient of thermal expansion of steel; The unit is °C, which represents the local temperature rise of the board material under the action of the cutting heat source. This is the elastic modulus of steel, in MPa. The thickness of the incoming sheet material is in mm; The measured yield strength of the steel is expressed in MPa. This is the amount of compensation for the offset of the cutting heat source, in mm. Preferably, the second optimization formula in S3 is a calculation formula for bending elastic-plastic rebound compensation, and the expression is shown in equation (2): （2）； In the formula: The actual downward stroke of the bending head is in mm; This is the reference value for the downward stroke of the bending base, in mm; This is the springback correction factor, which is dimensionless. The thickness of the incoming sheet material is in mm; The measured yield strength of the steel is expressed in MPa. This refers to the elastic modulus of steel, expressed in MPa. The normalized value of the cutting dimension deviation obtained from S2 is dimensionless. Preferably, the third optimization formula in S4 is a welding heat input adaptive control calculation formula, and its expression is shown in equation (3): （3）； In the formula: The welding heat input is expressed per unit length, in J / mm. The welding thermal efficiency is dimensionless. Welding voltage, unit: V; Welding current, in amperes (A). The welding walking speed is expressed in mm / s. The reference unit is the welding heat input per unit length, expressed in J / mm. The cutting deviation weighting coefficient is dimensionless. This is a normalized cutting dimension deviation, dimensionless. This is the bending deviation weighting coefficient, which is dimensionless. The normalized bending springback deviation is dimensionless. These are the material property weighting coefficients, dimensionless. The measured yield strength of the steel is expressed in MPa. This is the elastic modulus of steel, expressed in MPa. Preferably, the fourth optimization formula in S6 is a calculation formula for the growth kinetics of hot-dip galvanized coating, and its expression is shown in equation (4): （4）； In the formula: The actual zinc immersion time of the component, in seconds; The reference zinc immersion time is expressed in seconds. The actual thickness of the sheet metal used in the component is in mm. For reference sheet thickness, unit: mm; The residual stress influence weighting coefficient is dimensionless. The normalized residual stress estimate for the formed component is dimensionless. is a surface roughness weighting coefficient, dimensionless; The measured surface roughness of the component is expressed in μm. Let be the function of the effect of zinc pot temperature. The temperature of the zinc pot is in °C. Preferably, in the hot-dip galvanizing pretreatment process of S6, the pickling agent is a hydrochloric acid pickling system, the hydrochloric acid mass fraction is controlled at 8%-14%, and the pickling temperature is 25-45℃; the fluxing agent is a zinc chloride-ammonium chloride compound fluxing agent, the Baume degree of the fluxing agent solution is controlled at 18-24°Bé, and the fluxing temperature is 60-75℃; the aluminum element mass fraction of the zinc liquid inside the zinc pot is 0.005%-0.020%. Preferably, in S7, the passivation treatment uses chromium-based or chromium-free passivation agents, and the pH of the passivation solution is controlled between 3.0 and 5.

0. When the average thickness of the coating is greater than 85 μm, the passivation immersion time is set to 8-12 s; when the average thickness of the coating is 55-85 μm, the passivation immersion time is set to 4-8 s; and when the average thickness of the coating is less than 55 μm, the passivation immersion time is set to 2-4 s. The second aspect of the present invention provides an integrated intelligent molding and corrosion protection manufacturing system for transmission line tower components, used to implement the method of the first aspect, comprising: The incoming material information acquisition and initial model initialization unit (101) is used to acquire incoming material information and initialize the initial model: acquire the original parameters of the incoming steel material of the transmission tower components to be processed, the original parameters of the incoming steel material include steel grade, plate thickness, plate initial yield strength, plate rolling residual stress distribution, plate surface roughness and component design drawing geometric parameters; input the original parameters of the incoming steel material into the artificial intelligence control model with physical constraints to complete the initial parameter assignment of the model; The cutting process adaptive process parameter optimization and online size sensing unit (102) is used to perform cutting process adaptive process parameter optimization and online size sensing: based on the first optimization formula, the target process parameters of the cutting process are calculated in combination with the original parameters of the incoming steel. The target process parameters of the cutting process include cutting speed, cutting heat source power and cutting heat source offset compensation amount. Steel cutting and blanking are performed using cutting equipment. After the steel cutting and blanking are completed, the actual measured values ​​of the cross-sectional contour and edge dimensions of the cut component are collected by machine vision, and the cutting dimension deviation is calculated. The bending process parameter dynamic correction and springback compensation unit (103) is used to perform dynamic correction and springback compensation of bending process parameters: the cutting size deviation, steel yield strength and plate thickness are input into the second optimization formula to solve the target process parameters of the bending process, wherein the target process parameters of the bending process include the opening distance of the lower bending die, the downward stroke of the bending pressure head and the bending holding time; the bending process is performed, and after the bending process is completed, the machine vision collects the actual measured values ​​of the bending component limb width and the bending inner angle to obtain the bending springback deviation; The intelligent control unit for heat input and online correction unit for welding deformation in the welding forming process (104) is used to perform intelligent control for heat input and online correction for welding deformation in the welding forming process: the original parameters of the incoming steel of the component, the cutting size deviation and the bending springback deviation are input into the third optimization formula to calculate the target process parameters of the welding process, wherein the target process parameters of the welding process include welding current, welding voltage, welding travel speed and welding intermittent cooling time; the welding operation of angle steel and connecting plate assembly is performed, and the measured data of temperature field of welding area, flatness of component and relative position of bolt holes are collected in real time during the welding process to obtain welding deformation deviation, and the welding heat input parameters are finely adjusted in real time based on the welding deformation deviation; The forming component comprehensive status information summary and output unit (105) is used to summarize and output the forming component comprehensive status information: summarize the cutting size deviation, the bending springback deviation and the welding deformation deviation, extract the estimated value of the component residual stress and the component surface morphology feature data, and generate the forming component comprehensive status data package. The forming component comprehensive status data package carries the component individual identifier and is transmitted to the downstream hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solution and passivation process adaptive matching and coating quality closed-loop feedback process. The hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving unit (106) is used to perform hot-dip galvanizing pretreatment and AI-driven hot-dip galvanizing process parameter solving: receiving the comprehensive status data packet of the formed component, completing the component degreasing, pickling and fluxing pretreatment processes; inputting the component plate thickness, comprehensive deformation deviation, residual stress prediction and surface roughness of the component into the fourth optimization formula, and outputting the hot-dip galvanizing target process parameters, which include the zinc pot working temperature, component zinc immersion time and component in / out of zinc pot lifting rate; and executing the hot-dip galvanizing operation of the component according to the hot-dip galvanizing target process parameters. The passivation process adaptive matching and coating quality closed-loop feedback unit (107) is used to perform passivation process adaptive matching and coating quality closed-loop feedback: after the hot-dip galvanizing operation of the component is completed, the component coating thickness distribution and coating surface adhesion detection data are collected, and the passivation liquid temperature and passivation immersion time are adaptively adjusted according to the actual coating data; the coating quality detection results are back transmitted to the cutting process adaptive process parameter optimization and online size sensing unit (102), the bending process parameter dynamic correction and springback compensation unit (103), and the welding forming process heat input intelligent control and welding deformation online correction unit (104), and the model boundary constraint parameters of the first optimization formula, the second optimization formula, and the third optimization formula are iteratively updated to realize the closed-loop control of the entire forming-corrosion prevention process. The third aspect of the present invention is to provide an intelligent forming and corrosion protection integrated preparation device for transmission line tower components, used to implement the method of the first aspect, including: a material parameter acquisition module, a cutting processing module, a bending processing module, a welding processing module, a machine vision online detection module, a forming data processing module, a hot-dip galvanizing treatment module, a coating quality detection module, a closed-loop iterative control module, and a component identification reading and writing module; The incoming material parameter acquisition module is used to acquire the original parameters of the incoming steel to be processed and transmit them to the forming data processing module. The cutting and processing module receives the target process parameters for the cutting process output by the forming data processing module to complete the steel cutting and blanking; The bending processing module receives the target process parameters for the bending process output by the forming data processing module and completes the bending process. The welding processing module receives the target process parameters for the welding process output by the forming data processing module and completes the assembly welding process; The machine vision online inspection module is deployed at the output ends of the cutting module, the bending module and the welding module respectively, and collects the actual measurement data of the component size and transmits it back to the forming data processing module; The forming data processing module has a built-in artificial intelligence control model with physical constraints, and includes the first optimization formula, the second optimization formula, the third optimization formula and the fourth optimization formula. It receives the raw parameters of the incoming material and the machine vision measured size data, and outputs the target process parameters for cutting, bending, welding and hot-dip galvanizing respectively. It generates a comprehensive status data package of the formed component and sends it to the hot-dip galvanizing processing module. The hot-dip galvanizing module includes a pretreatment submodule, a zinc pot immersion galvanizing submodule, and a passivation submodule, which are respectively used to perform pretreatment, hot-dip galvanizing, and passivation operations after receiving the comprehensive status data packet of the formed component. The coating quality detection module is used to collect coating quality detection data including measured data of coating thickness and coating adhesion, and transmit the coating quality detection data to the closed-loop iterative control module. The closed-loop iterative control module feeds back the coating quality detection results to the forming data processing module, iteratively updates the physical boundary constraint parameters of the artificial intelligence control model, and completes the closed-loop control of the entire process. The component identification reading and writing module is set at each process flow station. The component identification reading and writing module reads the individual component identification and binds the corresponding process detection data to the individual component identification, which follows the component flow throughout the entire production chain. The system internally sets up a data physical boundary verification unit, which is connected to the forming data processing module. When the output process parameters of the first optimization formula, the second optimization formula, the third optimization formula and the fourth optimization formula exceed the physical boundary threshold of steel processing, the data physical boundary verification unit outputs an interception signal to prevent the transmission of abnormal process parameters to the lower processing unit.

Images