A process parameter intelligent early warning method and system for a galvanizing workshop

Abstract

The application relates to the technical field of steel pipe hot galvanizing process, in particular to a process parameter intelligent early warning method and system for a galvanizing workshop. The method comprises the following steps: acquiring a multi-source process parameter set of a steel pipe to be galvanized; performing steel pipe geometric feature recognition and physical section division based on the multi-source process parameter set, to generate a steel pipe segmented feature information set containing a straight pipe section, an end section and a special-shaped geometric section; performing differentiated risk assessment on the straight pipe section, the end section and the special-shaped geometric section based on the steel pipe segmented feature information set, to generate a regional risk feature information set; performing multi-section risk comprehensive discrimination and early warning level decision based on the regional risk feature information set, to generate and output a process parameter intelligent early warning report. The application can realize fine real-time monitoring of different structural regions of the steel pipe, identify defects such as inner wall blockage, zinc layer thickness deviation and special-shaped section slag inclusion in advance, improve the galvanizing quality and reduce the rework cost.

Description

Technical Field

[0001] This application relates to the field of hot-dip galvanizing process technology for steel pipes, and in particular to an intelligent early warning method and system for process parameters in galvanizing workshops. Background Technology

[0002] In the hot-dip galvanizing of steel pipes, product quality control is crucial for ensuring the corrosion resistance and extending the service life of the pipes. This process typically includes pretreatment, hot-dip galvanizing, internal and external purging, and cooling, and is widely used in industries such as petroleum, chemical, construction, and municipal water supply. Existing production control systems are usually equipped with programmable logic controllers (PLCs), industrial gateways, and data acquisition modules to monitor basic process parameters such as zinc bath temperature, immersion rate, and purging pressure in real time, and to trigger alarms for exceeding preset fixed thresholds. Operators view real-time data through on-site touchscreens or host computer interfaces, and, combined with periodic post-production sampling inspections or offline flaw detection results, manually assess and intervene in the production line status to maintain basic production order.

[0003] However, current technologies lack differentiated real-time risk warning mechanisms for different pipe structure regions when dealing with steel pipes with complex geometries. Because straight pipe sections, ends, and irregular structures such as elbows and diameter changes exhibit significant differences in thermodynamic behavior and hydrodynamic characteristics, a uniform monitoring model struggles to accurately identify potential defects in specific areas, such as zinc nodules, pipe blockage, uneven zinc layer thickness distribution, or slag accumulation. This results in some quality risks not being effectively detected and addressed before defects form. Summary of the Invention

[0004] This application provides an intelligent early warning method and system for process parameters in galvanizing workshops to solve the above-mentioned problems.

[0005] Firstly, this application provides an intelligent early warning method for process parameters in a galvanizing workshop, the method comprising: A multi-source process parameter set of the steel pipe to be galvanized is obtained. Based on the multi-source process parameter set, the geometric features of the steel pipe are identified and the physical segments are divided to generate a steel pipe segment feature information set including straight pipe segments, end segments and irregular geometric segments. Based on the segmented feature information set of the steel pipe, a dynamic blockage risk assessment of the inner wall is performed on the straight pipe segment, a zinc layer thickness distribution offset prediction is performed on the end segment, and a risk assessment of zinc liquid residue and slag accumulation is performed on the irregular geometric segment, generating a risk feature information set for different regions. Based on the aforementioned regional risk characteristic information set, multi-segment risk comprehensive judgment and early warning level decision are performed, and intelligent early warning reports of process parameters are generated and output.

[0006] Through the aforementioned technical solution, this application constructs a segmented feature information set for steel pipes, including straight pipe sections, end sections, and irregularly shaped geometric sections, by acquiring multi-source process parameter sets and identifying the geometric features and physical segments of the steel pipes. Based on this, targeted risk assessments are performed on the three types of areas, generating regional risk feature information sets. Finally, multi-segment risk comprehensive judgment and early warning level decision are performed, generating and outputting intelligent early warning reports for process parameters. Through the coordinated efforts of dynamic blockage risk assessment of the inner wall of straight pipe sections, zinc layer thickness distribution shift prediction of end sections, and zinc liquid residue and slag accumulation risk assessment of irregularly shaped geometric sections, the system achieves full-coverage monitoring of defects in different geometric feature areas along the entire length of the steel pipe. The high-priority setting of the comprehensive defect risk index for irregularly shaped sections ensures that serious structural defects can be identified and trigger high-risk early warnings immediately, while the refined analysis of straight pipe sections and end sections supplements the monitoring blind spots of conventional quality problems. This divide-and-conquer strategy improves the accuracy and timeliness of early warnings, enabling effective intervention before defects occur, reducing rework and material waste, and solving the technical challenges of traditional methods failing to effectively monitor defects in complex structures such as elbows and diameter changes, as well as uneven zinc coating at the ends.

[0007] Optionally, the process of generating the segmented feature information set of the steel pipe includes: Read the digital control database of the galvanizing process to obtain the set of physical parameters of the galvanized steel pipe; By using a sensor array arranged in the zinc pot, internal and external blowing stations and cooling section, real-time process parameter streams including zinc liquid temperature, steel pipe temperature, internal blowing pressure, external blowing pressure, zinc immersion angle, zinc immersion speed, residence time, pull-out speed and multi-point temperature of the steel pipe surface are collected and combined to form the multi-source process parameter set. Based on the physical parameter set of the specifications, the geometric features of the steel pipe are identified and the physical segments are divided. Each steel pipe is divided along its length into straight pipe segments, end segments, and irregular geometric segments including bends, right angles, and diameter change areas, thereby generating the steel pipe segment feature information set.

[0008] Through the aforementioned technical solution, this application achieves automation and precision in steel pipe geometric feature identification and physical segment division through the synergistic effect of the above technical features. By reading the galvanizing process digital control database to obtain the set of specification physical parameters, the theoretical geometric framework of the steel pipe is established. Simultaneously, a sensor array deployed in the zinc pot, internal and external blowing stations, and cooling section collects real-time process parameter streams, constructing a dynamic data dimension reflecting the actual production situation. Based on this, the system uses the static set of specification physical parameters as a spatial positioning reference and the dynamic multi-source process parameter set as a status verification basis. The two work together to support the execution of steel pipe geometric feature identification and physical segment division. Specifically, data such as bend angles and diameter change positions in the specification physical parameter set directly guide the locking of irregular geometric segments, while the real-time data provided by the sensor array ensures that the division of straight pipe segments and end segments remains accurate even when the steel pipe experiences minor deformation or operational fluctuations. This static and dynamic combined data processing mechanism not only solves the problem of generalized risk assessment caused by ignoring geometric differences in traditional methods, but also makes full use of the existing digital control infrastructure in the workshop. It can achieve high-precision segmented management without additional hardware costs, thereby improving the pertinence and accuracy of subsequent risk assessment of internal wall dynamic blockage, prediction of zinc layer thickness distribution deviation, and risk assessment of zinc liquid residue and slag accumulation.

[0009] Optionally, the process of identifying the geometric features of the steel pipe and dividing it into physical segments includes: Based on the physical parameter set of the specifications, the bend angle, bending radius, starting position of diameter change and diameter change ratio of the galvanized steel pipe are extracted and marked as the irregular geometric segment; Using the two ends of the irregular geometric segment as the dividing points, the remaining pipe body is divided into straight pipe segments along the length direction, and each predefined length range at both ends of the straight pipe segment is separately marked as the end segment.

[0010] Through the aforementioned technical solutions, this application achieves refined analysis of complex steel pipe geometries by leveraging the synergistic effect of the aforementioned technical features. First, by extracting key parameters such as elbow angles and bending radii to pinpoint irregular geometric segments, the boundaries of the highest-risk areas are established. Then, using the irregular segments as a benchmark, straight pipe segments are derived in reverse, ensuring the continuity of the regular area division. Finally, a predefined length range is introduced to independently mark end segments, capturing end areas prone to defects due to special heat dissipation and fluid boundary conditions. These three steps are interconnected. The precise positioning of irregular segments provides a clear cutoff point for the division of straight pipe segments, while the independent extraction of end segments compensates for the shortcomings of the straight pipe model in edge effects. This hierarchical and progressive division mechanism not only eliminates the risk of overlapping or omissions in areas but also allows subsequent assessments of straight pipe inner wall blockage, predictions of end thickness offsets, and analysis of slag inclusion risks in irregular segments to run independently within their respective spatial domains using dedicated algorithms, improving the overall structure and diagnostic accuracy of the early warning system.

[0011] Optionally, the process of generating the regional risk feature information set includes: Based on the segmented feature information set of the steel pipe, perform independent area risk assessments in parallel: For the straight pipe section, assess the risk of zinc nodules and pipe blockage on the inner wall to generate a risk level for the inner wall of the straight pipe; For the end segment, the offset between the end zinc layer thickness and the middle thickness is predicted, and an end thickness deviation index is generated; For the irregular geometric segment, assess the risk of zinc residue and slag accumulation, and generate a comprehensive defect risk index for the irregular segment; The risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular section are combined to generate the risk feature information set of the sub-regions.

[0012] Through the aforementioned technical solution, this application achieves a refined monitoring mode of "one pipe, three parts" by executing independent risk assessment mechanisms for straight pipe sections, end sections, and irregularly shaped sections in parallel. The risk level of the straight pipe inner wall, the thickness deviation index of the end section, and the comprehensive defect risk index of the irregularly shaped section complement each other, covering the three main types of defects: pipe blockage, uneven thickness, and residual slag inclusions. By generating a regional risk characteristic information set by comprehensively evaluating the results of these three dimensions, not only is the identification accuracy of different types of defects improved, but the system can also provide differentiated handling suggestions based on the specific source of risk. This modular and parallel architecture enhances the functional completeness of the early warning system, enabling precise intervention before defects occur, thereby effectively improving the quality of galvanized products and reducing zinc consumption.

[0013] Optionally, the process of generating the risk level of the inner wall of the straight pipe includes: Analyze the real-time temperature difference between the zinc liquid temperature and the steel pipe temperature to determine the probability level of premature solidification of the zinc liquid at the front end of the straight pipe inner wall. During the zinc immersion stage of steel pipes, non-periodic changes in the zinc pot liquid level fluctuations are monitored to infer the degree of local blockage on the inner wall; During the internal blowing stage, the rise time and peak stabilization time of the internal blowing pressure curve are analyzed to identify the obstruction strength of the zinc nodule on the airflow and obtain the characteristics of the internal blowing pressure curve. Based on the combined characteristics of the temperature difference probability level, the degree of local blockage of the inner wall, and the internal blowing pressure curve, the risk of the straight pipe inner wall is divided into three levels: low risk, medium risk, and high risk, thus generating the risk level of the straight pipe inner wall.

[0014] Through the aforementioned technical solutions, this application constructs a robust closed-loop monitoring system for the risk of straight pipe inner walls by leveraging the synergistic effect of the aforementioned technical features. Firstly, by utilizing real-time temperature difference analysis between the zinc bath and the steel pipe, the solidification tendency is predicted from a thermodynamic perspective, enabling early risk identification. Based on this, combined with monitoring the non-periodic changes in the zinc bath surface fluctuations during the galvanizing stage, minor obstructions already formed on the inner wall are indirectly perceived from a fluid dynamics perspective. Furthermore, through refined analysis of the rise time and peak stabilization time of the internal blowing pressure curve during the internal blowing stage, the severity of blockage is directly quantified from an aerodynamic perspective. These three aspects are not isolated but rather progressive and mutually reinforcing: abnormal temperature differences are often the trigger for blockage, surface fluctuations are the dynamic manifestation of the blockage process, and the pressure curve characteristics directly reflect the blockage outcome. By integrating the feature information from these three dimensions, the system can accurately classify the risk of the inner wall of straight pipe into three levels: low, medium, and high. This not only improves the accuracy and timeliness of risk identification, but also provides clear early warning signals before pipe blockage accidents occur, guiding operators to adjust process parameters or take intervention measures in a timely manner, thereby effectively preventing production losses and quality defects caused by premature solidification of zinc liquid.

[0015] Optionally, the process of generating the end thickness deviation index includes: Based on the zinc immersion angle, the zinc immersion speed, the dwell time, and the pull-out speed, and in conjunction with the zinc bath temperature, a theoretical zinc immersion time distribution along the length of the steel pipe is established. In the initial cooling stage after the steel pipe exits the zinc pot, the change in the amount of zinc liquid adhering to the end due to the difference in heat dissipation rate is identified based on the temperature of multiple points on the surface of the steel pipe. During the external blowing stage, the difference in external blowing pressure between the end and the middle is analyzed to determine the local air resistance difference at the end caused by the change in the thickness of the zinc liquid. Based on the theoretical zinc immersion time distribution, the changes in zinc liquid adhesion, and the local air resistance differences, the offset direction and offset level of the end zinc layer thickness relative to the middle are determined by setting a thinner threshold and a thicker threshold, and the end thickness deviation index is generated.

[0016] Through the aforementioned technical solutions, this application achieves refined prediction of the zinc layer thickness distribution at the end of the steel pipe by leveraging the synergistic effect of the above-mentioned technical features. First, based on the theoretical and actual immersion time distribution established using motion parameters such as immersion angle and speed, potential immersion time limitations at the end are predicted from the process source. Second, by utilizing multi-point surface temperature monitoring during the initial cooling phase, the abnormal heat dissipation caused by the end's geometry and the resulting dynamic changes in zinc liquid adhesion are captured in real time, compensating for the shortcomings of relying solely on motion parameters. Finally, by analyzing the pressure difference between the end and middle during the external blowing stage, the actual zinc layer deposition thickness is verified from the perspective of airflow resistance. These three aspects construct a closed-loop evaluation system from three dimensions: input process conditions, intermediate thermal evolution, and output fluid feedback. By setting thresholds for thinness and thickness, multi-dimensional data are fused into a single end thickness deviation index, not only eliminating the random errors of single sensor data but also clarifying the direction and severity of defect deviation, thus providing a scientific and quantitative basis for solving end thickness or thinness problems and guiding the real-time adjustment of process parameters.

[0017] Optionally, the process of generating the comprehensive defect risk index for the irregular segment includes: During the zinc immersion stage of steel pipe, the change in flow resistance of zinc liquid as it flows through the irregular section is calculated in real time. When the resistance suddenly increases, it is determined that the zinc liquid solidifies prematurely or is blocked by slag. In the initial cooling stage after the steel pipe exits the zinc pot, cold spots or hot spots in the temperature field caused by zinc liquid residue or slag inclusions are identified to obtain information on temperature field anomalies. During the cooling stage of the steel pipe, the micro-vibration signal generated by the release of residual stress in the irregular section when the steel pipe passes through is captured, and the abnormal amplitude of micro-vibration is analyzed. Based on the combined information of the flow resistance change, the abnormal temperature field, and the abnormal amplitude of the micro-vibration, a comprehensive defect risk index for the irregular segment is obtained through weighted fusion, and the defect type is determined to be zinc liquid residue or slag accumulation.

[0018] Through the aforementioned technical solution and steps, this application achieves multi-dimensional collaborative early warning for irregularly shaped geometric sections. It monitors fluid blockage during the galvanizing process in real time by observing changes in flow resistance, identifies surface heat distribution non-uniformity in the early cooling stage by combining temperature field anomaly information, and captures the mechanical response of internal mass distribution in the later cooling stage using micro-vibration anomaly amplitude. These three methods provide a comprehensive assessment of the irregularly shaped section from three independent dimensions: fluid dynamics, thermal, and mechanical. Based on this, the three indicators are weighted and fused, overcoming the limitations of single detection methods being susceptible to environmental interference and dynamically adjusting the risk assessment results according to the contribution of each indicator. This multi-source information fusion mechanism enables the system to issue early warnings at the early stages of defect formation (such as resistance changes during galvanizing), and to verify and refine these warnings in subsequent stages (such as temperature and vibration during cooling), ultimately accurately determining whether the defect type is zinc residue or slag accumulation. This provides clear handling guidelines for workshop operators and improves the quality control level of irregularly shaped galvanized steel pipe sections.

[0019] Optionally, the process of generating and outputting the intelligent early warning report for process parameters includes: Based on the aforementioned regional risk characteristic information set, multi-segment risk comprehensive judgment is performed on different sections of each steel pipe to determine the comprehensive health level of each steel pipe; Based on the overall health level, a corresponding recommended treatment strategy is matched from the preset treatment strategy library. The recommended treatment strategies include maintaining the current parameter operation, suggesting adjustment of specific process parameters, triggering secondary auxiliary processing, or offline testing. The comprehensive health level, the risk details of each region included in the regional risk characteristic information set, and the recommended treatment strategy are integrated to generate the intelligent early warning report of process parameters, and output to the workshop control visualization interface.

[0020] Through the aforementioned technical solution, this application constructs a three-tiered intelligent early warning architecture of assessment, decision-making, and output by leveraging the synergistic effect of the steps described above. Comprehensive judgment based on regional risk characteristic information sets solves the problem that a single indicator cannot fully reflect the overall quality of the steel pipe. In particular, by assigning the highest priority to risks in irregularly shaped sections, it ensures zero missed detection of serious defects. Dynamic matching between the preset handling strategy library and the comprehensive health level transforms abstract risk data into specific executable instructions, achieving closed-loop automation of process control. Furthermore, the output of early warning reports, integrating detailed risk information and handling suggestions, to a visual interface enhances the user-friendliness of human-computer interaction, enabling on-site operators and management to simultaneously grasp the production situation and respond quickly. This complete process, from multi-dimensional data fusion to hierarchical decision-making and intuitive display, avoids delays or errors caused by insufficient human experience, reduces scrap rates and zinc consumption, and improves the intelligence level and operational efficiency of the galvanizing production line.

[0021] Optionally, the multi-segment risk comprehensive assessment includes: The risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular section are prioritized, with the comprehensive defect risk index of the irregular section having the highest priority. When the comprehensive defect risk index of the irregular segment is in a high-risk state, the comprehensive health level is directly determined to be high-risk. When the comprehensive defect risk index of the irregular segment is in a non-high-risk state, the comprehensive health level is determined to be either a general concern level or a warning level based on the combination of the risk level of the straight pipe inner wall and the end thickness deviation index.

[0022] Through the aforementioned technical solution, this application achieves the organic synergy of multiple risk characteristics through the priority ranking and hierarchical judgment mechanism. By setting the comprehensive defect risk index of the irregularly shaped segment as the highest priority and directly determining the comprehensive health level as high-risk when it is in a high-risk state, the system constructs the first line of defense against fatal defects, ensuring that the bottom line of production safety is not breached. On this basis, when the risk of the irregularly shaped segment is controllable, a secondary judgment is further made by combining the combined state of the risk level of the straight pipe inner wall and the end thickness deviation index to generate a general concern level or a warning level, thereby realizing differentiated responses to defects of different severity. This hierarchical decision-making logic not only solves the judgment conflict problem when multiple sources of risk coexist, but also optimizes the allocation of early warning resources, making high-risk alarms absolutely authoritative, while general deviations receive reasonable attention and handling suggestions.

[0023] Secondly, this application provides an intelligent early warning system for process parameters in galvanizing workshops, the system comprising: The steel pipe segmentation module is used to acquire a multi-source process parameter set of the steel pipe to be galvanized, and based on the multi-source process parameter set, to perform steel pipe geometric feature identification and physical segment division, and generate a steel pipe segmentation feature information set including straight pipe segments, end segments and irregular geometric segments; The risk feature module is used to perform dynamic blockage risk assessment on the inner wall of the straight pipe section based on the segment feature information set of the steel pipe, perform zinc layer thickness distribution offset prediction on the end section, and perform zinc liquid residue and slag accumulation risk assessment on the irregular geometric section, and generate regional risk feature information set. The intelligent early warning module is used to make comprehensive judgments on multi-segment risks and make early warning level decisions based on the risk feature information set of the sub-regions, and generate and output intelligent early warning reports of process parameters. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application; Figure 2 A flowchart illustrating an intelligent early warning method for process parameters in a galvanizing workshop, provided as an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an intelligent early warning system for process parameters in a galvanizing workshop, provided as an embodiment of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the related objects before and after it have an "or" relationship.

[0028] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0029] Figure 1This diagram illustrates an application scenario provided by this application. It comprehensively showcases the end-to-end architecture of galvanizing workshop process data, from acquisition to application. The architecture includes on-site process sensors, a Siemens S7-200 SMART PLC and analog input modules, an industrial switch, an industrial IoT gateway, and a cloud-based IoT management platform. It also features a Weintek industrial touchscreen and an external large display screen, as well as a WeChat mini-program mobile application. On-site sensors collect process parameters such as temperature, pressure, and liquid level. After PLC signal processing and local logic control, the data is converted and transmitted via the switch and IoT gateway, then uploaded to the cloud platform for centralized management, real-time monitoring, and historical data storage and export. Simultaneously, the touchscreen allows for on-site configuration of production screens and visualization of parameters. This architecture leverages existing industrial hardware to form a complete data acquisition and transmission channel, combining local interlocking control with remote cloud monitoring to ensure that galvanizing process production data is collectable, monitorable, and traceable throughout the entire process.

[0030] The specific implementation method can be referred to in the following embodiments, wherein the data mentioned in the embodiments are only for reference and examples, so that relevant personnel can better understand them.

[0031] Figure 2 This document presents a flowchart illustrating an intelligent early warning method for process parameters in a galvanizing workshop, as provided in one embodiment of this application. The method described in this embodiment can be applied to the server in the aforementioned scenario. (See also...) Figure 2 The process, specifically the implementation steps, are as follows: Example 1: Common quality problems in the hot-dip galvanizing process of steel pipes include zinc nodules or even complete blockage of the pipe due to premature solidification of the molten zinc on the inner wall; uneven zinc layer thickness distribution at the pipe ends due to differences in the immersion trajectory and cooling rate; and slag inclusions at geometric abrupt changes in shape, such as bends, right angles, and diameter changes. Current technologies largely rely on post-production sampling or offline flaw detection, lacking real-time, segmented intelligent early warning systems for the entire production process. This makes it difficult to fully utilize existing data acquisition and control equipment in the workshop for geometric feature identification and risk assessment of steel pipes, resulting in defects often being discovered only after they occur, leading to zinc waste and increased rework costs.

[0032] To address the aforementioned problems, this application provides an intelligent early warning method for process parameters in galvanizing workshops, the method comprising: Step 1: Obtain the multi-source process parameter set of the steel pipe to be galvanized. Based on the multi-source process parameter set, perform geometric feature identification and physical segment division of the steel pipe to generate a segment feature information set of the steel pipe including straight pipe segments, end segments and irregular geometric segments. The multi-source process parameter set refers to a multi-dimensional data set used to characterize the state of the steel pipe and environmental conditions during the galvanizing process. It originates from real-time data streams collected by sensor arrays deployed in the zinc bath, internal and external blowing stations, and cooling section. Specifically, it includes physical quantities such as zinc bath temperature, steel pipe temperature, internal blowing pressure, external blowing pressure, immersion angle, immersion speed, residence time, pull-out speed, and multi-point temperature on the steel pipe surface. This multi-source process parameter set serves as the foundational input data for subsequent geometric feature identification and risk assessment. Steel pipe geometric feature identification and physical segment division refer to logically dividing each steel pipe to be galvanized into three independent physical regions along its length direction based on the set of specification physical parameters (such as steel pipe length, wall thickness, elbow angle, bending radius, starting position of diameter change, and diameter change ratio) read from the galvanizing process digital control database. Specifically, the system marks structurally abrupt changes, including bends, right angles, and diameter variations, as irregular geometric segments. Using the ends of these irregular geometric segments as boundaries, the remaining regular pipe body is divided into straight segments. Furthermore, predefined length ranges (e.g., 50mm at each end) at both ends of each straight segment are separately marked as end segments. This segmentation method generates a set of segmented feature information for the steel pipe that accurately describes the geometric properties of different sections, providing a spatial benchmark for subsequent differentiated risk assessment. For example, for a steel pipe with a 90-degree bend, the system automatically extracts the bend angle and radius, marking them as irregular geometric segments. The long straight sections before and after the bend are marked as straight segments, and specific sections at the very front and rear of the pipe are marked as end segments. This segmentation process, based on geometric fingerprints, decouples the complex steel pipe structure into independent units with clear physical meaning, thereby enabling targeted modeling of defect mechanisms in different regions.

[0033] Step 2: Based on the segmented feature information set of the steel pipe, perform dynamic blockage risk assessment on the inner wall of the straight pipe segment, perform zinc layer thickness distribution offset prediction on the end segment, and perform zinc liquid residue and slag accumulation risk assessment on the irregular geometric segment to generate regional risk feature information set; The dynamic blockage risk assessment of the inner wall is a specialized analysis process for straight pipe sections, designed to identify the risk of zinc nodules and blockage on the inner wall. This assessment determines the likelihood of premature solidification of zinc at the front end of the straight pipe by analyzing the real-time temperature difference between the molten zinc and the steel pipe. It also monitors the non-periodic changes in the zinc bath level during the immersion process to infer the degree of local blockage on the inner wall, and combines the rise time and peak stabilization time of the internal blowing pressure curve during the internal blowing stage to identify the degree of obstruction to airflow caused by zinc nodules. Finally, by combining the above-mentioned temperature difference likelihood level, the degree of local blockage on the inner wall, and the characteristics of the internal blowing pressure curve, the risk of the straight pipe inner wall is classified into three levels: low risk, medium risk, and high risk, generating a risk level for the straight pipe inner wall. The zinc layer thickness distribution offset prediction is a specialized analysis process for the end section, aiming to address the inconsistency between the zinc layer thickness at the end and in the middle. This prediction is based on the immersion angle, immersion speed, dwell time, and pull-out speed, combined with the zinc bath temperature to establish a theoretical immersion time distribution along the length of the steel pipe. During the initial cooling stage after the steel pipe exits the zinc bath, the temperature at multiple points on the pipe surface is used to identify changes in zinc adhesion at the ends due to differences in heat dissipation rates. During the external blowing stage, the difference in external blowing pressure between the ends and the middle is analyzed to determine local air resistance differences. By combining the theoretical immersion time distribution, changes in zinc adhesion, and local air resistance differences, and by setting thinning and thickening thresholds, the direction and level of the zinc layer thickness deviation at the ends relative to the middle are determined, generating an end thickness deviation index. Zinc bath residue and slag inclusion risk assessment is a specialized analysis process for irregularly shaped geometric sections, used to detect defects at geometric abrupt changes. This assessment calculates the flow resistance changes of molten zinc as it flows through the shaped section during the zinc immersion stage. A sudden increase in resistance is identified as premature solidification of the molten zinc or slag inclusion blockage. In the initial cooling stage, it identifies cold or hot spots in the temperature field caused by molten zinc residue or slag inclusion, obtaining anomaly information. During the cooling stage, it captures micro-vibration signals generated by the release of residual stress in the shaped section as the steel pipe passes through, analyzing the abnormal amplitude of the micro-vibration. By combining the changes in flow resistance, anomaly information in the temperature field, and abnormal amplitude of the micro-vibration, a weighted fusion is performed to obtain a comprehensive defect risk index for the shaped section, and the defect type is determined to be molten zinc residue or slag accumulation. For example, when a sudden increase in flow resistance exceeding 15% of the baseline is detected in a shaped section of a steel pipe during zinc immersion, and an infrared thermal imager detects a cold spot at the corresponding location, while an accelerometer captures an abnormal micro-vibration peak, the system determines that there is a high-risk molten zinc residue in that area. By executing the three independent risk assessment models in parallel, the system can simultaneously output the risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular section, which together constitute a risk feature information set for different regions, thus achieving refined capture of different types of defects.

[0034] Step 3: Based on the risk characteristic information set of different regions, perform multi-segment risk comprehensive judgment and early warning level decision, and generate and output intelligent early warning reports of process parameters.

[0035] The multi-segment risk comprehensive judgment refers to the process of prioritizing and logically integrating the risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregularly shaped section. In this process, the comprehensive defect risk index of the irregularly shaped section has the highest priority. When the comprehensive defect risk index of the irregularly shaped section is in a high-risk state, the overall health level is directly determined to be high-risk; when the comprehensive defect risk index of the irregularly shaped section is not in a high-risk state, the overall health level is determined to be either a general concern level or a warning level based on the combination of the risk level of the straight pipe inner wall and the end thickness deviation index. The early warning level decision is based on the determined overall health level, matching the corresponding recommended handling strategy from the preset handling strategy library. Recommended handling strategies include maintaining the current parameter operation, suggesting adjustments to specific process parameters, triggering secondary auxiliary treatment, or offline testing. The intelligent early warning report for process parameters is a structured document generated by integrating the overall health level, the risk details of each region contained in the regional risk characteristic information set, and the recommended handling strategies. It is output to the workshop control visualization interface and cloud platform for operators or management systems to view and respond to in real time. For example, if the system determines that a steel pipe's overall health level is high-risk and the defect type is zinc residue on the inside of the elbow, the generated warning report will be immediately pushed to the management personnel's terminal. The elbow area of ​​the steel pipe will be highlighted in red on the workshop touchscreen, and a suggestion will pop up to divert the steel pipe to the inspection platform. Through this tiered judgment and strategy matching mechanism, the system can not only accurately reflect the overall quality status of the steel pipe but also provide specific intervention guidance, thereby realizing a shift from passive detection to proactive early warning.

[0036] Example 2: In another embodiment, this application provides a process for generating a segmented feature information set for steel pipes, which includes the following steps: Step 1: Read the digital control database of the galvanizing process to obtain the set of physical parameters of the galvanized steel pipe; The specification physical parameter set refers to the basic design data of steel pipes stored in the workshop's digital control system. This data originates from the digital control database of the galvanizing process, which is typically integrated into a Siemens PLC or host computer management system. Specifically, the specification physical parameter set includes the total length, wall thickness, outer diameter, and geometric abrupt change characteristic parameters of the steel pipe. These parameters cover key dimensional information such as bend angle, bending radius, coordinates of the starting position of the diameter change, and the diameter ratio. These parameters provide a static benchmark for subsequent geometric feature identification, ensuring the system can know the theoretical shape of the steel pipe to be processed in advance. For example, when producing a batch of irregularly shaped pipes containing 90-degree bends, the system reads from the database that the bend angle of this batch of steel pipes is 90±1 degrees, the bending radius is 1.5 times the pipe diameter, and the starting position of the diameter change is 2000mm from the pipe end. This database-based pre-reading method avoids repetitive geometric measurements for each steel pipe, improves data processing efficiency, and provides accurate spatial coordinates for subsequently dividing the steel pipes into different physical segments.

[0037] Step 2: By deploying sensor arrays in the zinc pot, internal and external blowing stations, and cooling section, real-time process parameter streams are collected, including zinc liquid temperature, steel pipe temperature, internal blowing pressure, external blowing pressure, zinc immersion angle, zinc immersion speed, residence time, pull-out speed, and multi-point temperature on the steel pipe surface. These data are then combined to form a multi-source process parameter set. The multi-source process parameter set is a dynamic data stream obtained through real-time acquisition and fusion of sensor arrays distributed at key stations of the galvanizing production line. Specifically, the sensor array includes temperature and level sensors installed in the zinc pot area, pressure sensors and angle encoders at the internal and external blowing stations, and infrared thermal imagers and accelerometers in the cooling section. Zinc liquid temperature and steel pipe temperature characterize the heat exchange state; internal and external blowing pressure reflect the airflow's control effect on the zinc layer; immersion angle, immersion speed, residence time, and pull-out speed collectively describe the trajectory and duration of the steel pipe in the zinc liquid; and multi-point temperatures on the steel pipe surface are used to monitor the uniformity of heat dissipation during the cooling process. This multi-source process parameter set, composed of these parameters, dynamically reflects the actual physicochemical changes occurring during production and corrects for deviations that may arise from relying solely on static specification data. For example, in a certain production run, although the database recorded a zinc immersion speed set to 1.2 m / s, the actual pull-out speed, collected in real time by the sensor array, fluctuated between 1.15 m / s and 1.25 m / s due to mechanical variations. The system incorporated this real-time fluctuation data into a multi-source process parameter set, thereby more accurately assessing the risk of zinc layer adhesion during that time period. By combining static specification data with dynamic process flow, a data closed loop from theoretical pre-setting to actual perception was achieved, improving the robustness of risk identification.

[0038] Step 3: Based on the physical parameter set of specifications, perform geometric feature identification and physical segment division of steel pipes, and divide each steel pipe into straight pipe segments, end segments and irregular geometric segments including elbows, right angles and diameter change areas along the length direction, and generate a steel pipe segment feature information set; The segmented feature information set of the steel pipe is structured data generated by spatially mapping the entire length of the steel pipe using a logical judgment algorithm, based on the aforementioned set of physical parameters. The specific segmentation method is as follows: First, based on the bend angle, bending radius, and diameter change location in the physical parameter set, areas where the geometric structure undergoes abrupt changes are identified and marked as irregular geometric segments. These segments encompass irregular areas such as bends, right-angle turns, and diameter changes. Second, using the two ends of the irregular geometric segments as dividing points, the remaining continuous regular pipe body is marked as straight pipe segments. Finally, at both ends of the straight pipe segments, a predefined length range (e.g., 50mm to 100mm each) is extended towards the pipe opening and separately marked as end segments. This segmentation method discretizes the continuous steel pipe entity into independent assessment units with different risk characteristics, enabling subsequent risk assessment models to differentiate between the blockage risk of straight pipe segments, the thickness deviation risk of end segments, and the residual slag inclusion risk of irregular segments. For example, for a steel pipe with a total length of 6 meters and a 90-degree bend in the middle, the system identifies the bend and the transition area before and after it (a total of 300mm) as an irregular geometric segment, the 80mm range at each end as end segments, and the remaining 5540mm as straight pipe segments. It then generates a segmented feature information set containing the start and end coordinates and type labels of these three segments. This achieves the digital reconstruction of the steel pipe's geometric features, laying a solid spatial foundation for subsequent precise regional early warning.

[0039] Example 3: In another embodiment, this application provides a method for identifying the geometric features of a steel pipe and dividing it into physical segments. The method includes the following steps: Step 1: Based on the physical parameter set of the specifications, extract the bend angle, bending radius, starting position of the diameter change and the diameter change ratio of the galvanized steel pipe, and mark them as irregular geometric segments; The physical parameter set refers to the basic geometric description data of the steel pipe to be processed, read from the digital control database of the galvanizing process or the production order system. The elbow angle refers to the angle at which the axial direction of the steel pipe changes, characterizing the degree of bending; the bending radius refers to the radius of curvature of the elbow centerline, defining the spatial span of the bending area; the starting position of the diameter change refers to the axial coordinate of the point where the steel pipe diameter changes abruptly; and the diameter ratio refers to the ratio of the diameters of the two sections of the steel pipe before and after the diameter change. The extraction process involves parsing the structured fields in the physical parameter set to locate the above four key geometric fingerprints, and marking areas that meet preset abrupt change conditions (e.g., elbow angles greater than 0 degrees and less than 180 degrees, or diameter ratios not equal to 1) in the steel pipe length coordinate system, defining them as irregular geometric segments. This step aims to identify the areas in the steel pipe structure with the most complex fluid dynamics and the highest defect risk from a macroscopic geometric perspective. By prioritizing the identification of these structural abrupt change points, it is possible to avoid incorrectly including irregular areas in the straight pipe model for calculation, thus providing accurate spatial anchor points for subsequent flow resistance analysis and temperature field anomaly detection for irregular segments.

[0040] Step 2: Using the two ends of the irregular geometric segment as the dividing points, divide the remaining pipe body into straight pipe segments along the length direction, and mark each predefined length range at both ends of the straight pipe segment as an end segment.

[0041] The dividing point can refer to the starting and ending coordinates of the aforementioned irregular geometric segments along the axial direction of the steel pipe. A straight pipe segment can refer to the continuous portion of the steel pipe with a regular geometric shape and constant cross-section after deducting all irregular geometric segments. The predefined length range is an empirical value derived from historical process data, used to define the pipe end region significantly affected by boundary effects; for example, it can be set to a range of 50mm to 100mm from the physical endpoint of the steel pipe or the connection point of the irregular segments. The division process first involves cutting the entire length of the steel pipe using the boundary coordinates of the irregular geometric segments. After removing the irregular portions, the remaining continuous intermediate interval is identified as the straight pipe segment. Subsequently, at the two physical ends of the straight pipe segment, a predefined length range is cut inwards, and this portion is independently marked as the end segment. The core area of ​​the straight pipe segment, excluding the end segments, remains as the main body of the straight pipe segment. This division strategy of first abrupt change and then regularity, first the whole and then the part, effectively solves the problem of blurred boundaries between ends and straight pipes in traditional methods. For example, for a steel pipe with a total length of 6 meters and a 90-degree bend in the middle, the system first identifies the bend area as the irregular section, and automatically classifies the remaining two parts as straight pipe section candidate areas. Then, 50mm is cut from the end of each straight pipe section as the end segment, and the rest is confirmed as the straight pipe section. Through this standardized partitioning logic, the automated analysis of the steel pipe structure is achieved, avoiding errors caused by manual intervention, ensuring that all steel pipe ends are uniformly modeled, which is conducive to establishing a universal end thickness prediction model. At the same time, it also provides a reliable spatial partitioning basis for subsequent application of differentiated risk assessment models for different areas.

[0042] Example 4: In one optional embodiment, this application provides a process for generating a regional risk feature information set. The method further includes specific steps for performing independent regional risk assessments in parallel based on the steel pipe segment feature information set.

[0043] Step 1: For straight pipe sections, assess the risk of zinc nodules and pipe blockage on the inner wall, and generate the risk level of the inner wall of the straight pipe; In this context, a straight pipe section refers to the main body of a steel pipe with a regular geometric shape along its length, without bends or diameter changes. Assessing the risk of zinc nodules and pipe blockage involves monitoring key physical quantities affecting the flow and solidification of the molten zinc during hot-dip galvanizing to determine whether protrusions (zinc nodules) formed by premature solidification of the molten zinc appear on the inner wall of the straight pipe, or whether excessive accumulation leads to a reduction in the pipe's cross-sectional area or even complete blockage. This risk level for the inner wall of the straight pipe is obtained through comprehensive analysis of the real-time temperature difference between the molten zinc and the steel pipe, the fluctuation characteristics of the zinc pot surface, and the characteristics of the internal blowing pressure curve. Specifically, the system first analyzes the real-time temperature difference between the zinc bath and the initial temperature of the steel pipe. If the temperature difference is too large, it indicates a high probability that the zinc bath will solidify prematurely at the front end of the straight pipe's inner wall. Second, during the zinc immersion stage, the system monitors the non-periodic changes in the zinc bath's surface level. Abnormal fluctuations in the surface level often indicate localized blockages on the inner wall, leading to changes in the volume of displaced zinc bath. Finally, during the internal blowing stage, the system analyzes the rise time and peak stabilization time of the internal blowing pressure curve. Zinc nodules alter the cross-sectional area of ​​the airflow channel, thus hindering airflow and changing the dynamic characteristics of pressure buildup. Based on the combined analysis of the temperature difference probability level, the degree of localized blockage on the inner wall, and the characteristics of the internal blowing pressure curve, the system classifies the risk of the straight pipe's inner wall into three levels: low, medium, and high risk. For example, when the rise time of the internal blowing pressure curve is more than 20% longer than the standard baseline value, and the peak stabilization time exhibits violent fluctuations, the system determines that the airflow is severely obstructed, generating a high-risk level for the straight pipe's inner wall. Through this multi-dimensional dynamic assessment, the system can accurately identify the unique blockage risks in the straight pipe section, providing a quantitative basis for subsequent process adjustments.

[0044] Step 2: For the end segment, predict the offset between the end zinc layer thickness and the middle thickness, and generate the end thickness deviation index; The end section refers to the pipe area within a predefined length range at both ends of a straight pipe section. Due to edge effects and differences in heat dissipation conditions, this area is prone to uneven zinc layer thickness. The predicted deviation between the end zinc layer thickness and the middle thickness refers to the calculated deviation of the actual zinc layer adhesion in the end area relative to the standard thickness in the middle of the steel pipe. This end thickness deviation index is generated based on zinc immersion motion parameters, zinc liquid temperature distribution, and external blowing pressure difference. Specifically, the system first establishes a theoretical zinc immersion time distribution model along the length of the steel pipe based on the zinc immersion angle, zinc immersion speed, residence time, and pull-out speed, combined with the zinc liquid temperature. Then, in the initial cooling stage after the steel pipe exits the zinc pot, multi-point temperature data of the steel pipe surface collected by an infrared thermal imager is used to identify the phenomenon of reduced or increased zinc liquid adhesion at the end due to the faster heat dissipation rate than in the middle. Simultaneously, during the external blowing stage, the difference in external blowing pressure between the end and the middle is analyzed. Changes in the end zinc layer thickness will cause differences in local air resistance, which will be reflected in the pressure reading. By considering the theoretical distribution of zinc immersion time, variations in zinc liquid adhesion, and local differences in air resistance, the system determines the direction (thinner or thicker) and level of zinc layer thickness deviation at the end relative to the middle section by setting thinner and thicker thresholds. For example, if the temperature drop rate at the end surface is significantly higher than that at the middle section, and the corresponding external blowing pressure difference indicates reduced air resistance, the system determines that the zinc layer at that location is too thin and calculates a specific end thickness deviation index (e.g., -15μm). The generation of this index allows the system to quantify end quality defects, thereby enabling targeted optimization of the zinc immersion trajectory or external blowing parameters.

[0045] Step 3: For irregularly shaped geometric segments, assess the risk of zinc residue and slag accumulation, and generate a comprehensive defect risk index for the irregularly shaped segments; The irregular geometric section refers to the pipe area containing abrupt changes in geometric structure such as bends, right angles, and diameter changes. These areas are prone to zinc stagnation or impurity deposition due to complex flow fields. Assessing the risk of zinc slag residue and slag accumulation involves determining whether there are undischarged excess zinc slag or zinc oxide slag defects inside the irregular section. The comprehensive defect risk index for this irregular section is obtained by weighted fusion of changes in flow resistance, abnormal temperature field information, and abnormal amplitude of micro-vibration. Specifically, during the zinc immersion stage of the steel pipe, the system calculates the changes in flow resistance of the zinc slag as it flows through the irregular section in real time. When the resistance suddenly increases, it is determined that the zinc slag has solidified prematurely or that there is slag blockage. In the initial cooling stage after the steel pipe exits the zinc pot, abnormal temperature field information caused by zinc slag residue (forming hot spots) or slag inclusion (forming cold spots) is identified. During the cooling stage of the steel pipe, the micro-vibration signal generated by the release of residual stress in the irregular section as the steel pipe passes through is captured, and the abnormal amplitude of micro-vibration is analyzed, because slag inclusion or residue can cause changes in structural stiffness, thus triggering abnormal vibration. By combining the characteristics of the above three dimensions, a comprehensive defect risk index for irregularly shaped pipe sections is obtained through weighted fusion, and the defect type is simultaneously determined to be either zinc residue or slag accumulation. For example, when the differential pressure sensor detects a sudden increase of 15% in the resistance flowing through the elbow, and the infrared thermal imager finds obvious cold spots on the inside of the elbow, while the accelerometer captures a micro-vibration peak at a specific frequency, the system calculates a high comprehensive defect risk index for the irregularly shaped pipe section (e.g., 0.85) and determines the defect type to be slag accumulation. This multi-physical quantity fusion evaluation mechanism fills the gap in online inspection of irregularly shaped pipe fittings.

[0046] Step 4: Combine the risk level of the inner wall of the straight pipe, the end thickness deviation index, and the comprehensive defect risk index of the irregular section to generate a risk characteristic information set for different regions.

[0047] The regional risk characteristic information set is a complete data structure that summarizes the risk assessment results of three different physical sections of the steel pipe, used to support subsequent comprehensive judgment and early warning decisions. Generating this information set involves logically integrating and encapsulating the risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregularly shaped section, all generated independently in the aforementioned steps. This process is not a simple numerical superposition, but rather the construction of a structured set containing risk details, risk type identifiers, and confidence levels for each region. Specifically, the system maps the three-level risk status of the straight pipe section, the quantitative deviation index of the end section, and the comprehensive defect index and defect type label of the irregularly shaped section to a unique identifier for the current steel pipe. For example, for a steel pipe to be galvanized, the generated regional risk characteristic information set might include: straight pipe section - low risk, end section - deviation index +12μm (too thick), and irregularly shaped section - risk index 0.3 (low risk, no defects). The generation of this information set enables a panoramic digital description of the quality status of the entire steel pipe, providing a data foundation for subsequent steps to conduct multi-segment risk comprehensive assessment and early warning level decision-making, and ensuring the accuracy and traceability of the early warning report.

[0048] Example 5: In one optional embodiment, this application provides a process for generating the risk level of the inner wall of a straight pipe, the method comprising the following steps: Step 1: Analyze the real-time temperature difference between the zinc liquid temperature and the steel pipe temperature to determine the probability level of premature solidification of the zinc liquid at the front end of the straight pipe inner wall. The zinc bath temperature refers to the real-time temperature of the molten zinc in the galvanizing pot, typically obtained using an immersion thermocouple or infrared thermometer. The steel pipe temperature refers to the surface or core temperature of the steel pipe before it enters the zinc pot. The real-time temperature difference is the numerical difference between the two at the same time point. This temperature difference probability level is used to quantify the thermodynamic tendency of the zinc bath to solidify due to excessively rapid heat exchange upon contact with the inner wall of the steel pipe. Specifically, the system presets a critical temperature difference threshold range. When the real-time temperature difference exceeds the upper limit of this range, it is judged as a high probability level, meaning that the zinc bath is very likely to solidify rapidly at the front end of the straight pipe inner wall to form an initial zinc coating. When the real-time temperature difference is within the range, it is judged as a medium probability level; and when it is below the lower limit, it is a low probability level. For example, if the zinc bath temperature is 450℃, while the steel pipe preheating temperature is only 200℃, the temperature difference is 250℃, far exceeding the set high-risk threshold of 180℃, and the system marks it as a high probability level. This thermodynamic-based prediction can identify potential solidification risks before physical blockage occurs, providing a basis for subsequent comprehensive rating.

[0049] Step 2: During the zinc immersion stage of the steel pipe, monitor the non-periodic changes in the zinc pot liquid level to infer the degree of local blockage on the inner wall; The zinc immersion stage refers to the period during which the steel pipe is fully or partially immersed in molten zinc for heat exchange and alloying reactions. The non-periodic fluctuations in the zinc pot surface refer to irregular undulations caused by abnormal airflow within the steel pipe or obstructed zinc flow, after excluding regular disturbances such as mechanical stirring and water impact. This variation indirectly reflects the unobstructed flow status of the straight pipe's inner wall. Specifically, when localized zinc deposits on the inner wall of the straight pipe reduce the flow cross-sectional area, the gas escape or zinc replacement process inside the steel pipe becomes unstable, causing irregular, violent fluctuations or vibrations at specific frequencies on the zinc pot surface. The system collects time-series data on the liquid level using a high-precision level sensor, extracts the non-periodic component after filtering, and matches its amplitude with a preset blockage correlation model to infer the degree of local blockage on the inner wall. For example, if random pulse-like fluctuations with an amplitude greater than 5 mm and no fixed period are detected, it is inferred that there is moderate to severe local blockage on the inner wall. This characteristic, combined with the aforementioned probability level of temperature difference, indicates that a large temperature difference can easily lead to solidification, and the accumulation of solidified material directly causes non-periodic fluctuations in the liquid level. Together, these factors verify the authenticity of the risk to the inner wall.

[0050] Step 3: During the internal blowing stage, analyze the rise time and peak stabilization time of the internal blowing pressure curve, identify the resistance strength of the zinc nodule to the airflow, and obtain the characteristics of the internal blowing pressure curve. The internal blowing stage refers to the process of using high-pressure gas to purge the inner wall of the steel pipe after it has been removed from the zinc molten metal, removing excess zinc and preventing blockage. The internal blowing pressure curve records the trajectory of pressure change within the pipe over time during the internal blowing process. The rise time refers to the time required for the pressure to rise from its initial value to 90% of the set working pressure, and the peak stabilization time refers to the duration for the pressure to maintain fluctuations within the allowable error range after reaching its peak. These two parameters together constitute the characteristics of the internal blowing pressure curve, used to directly quantify the obstruction strength of the zinc nodules on airflow. Specifically, if there are severe zinc nodules on the inner wall of the straight pipe, the airflow channel narrows, leading to slow pressure build-up, manifested as a prolonged rise time. Simultaneously, due to turbulence and pulsation generated when the airflow passes through the narrow gap, the pressure peak is difficult to maintain stability, manifested as a shortened peak stabilization time or increased fluctuation amplitude. For example, under normal circumstances, the pressure rise time is 0.5 seconds, and the peak stabilization time is 3.0 seconds; if the rise time is detected to be extended to 1.2 seconds and the peak fluctuation exceeds ±15%, it is identified as strong obstruction.

[0051] Step 4: Based on the combined characteristics of temperature difference probability level, degree of local blockage of inner wall and internal blowing pressure curve, the risk of straight pipe inner wall is divided into three levels: low risk, medium risk and high risk, thus generating the risk level of straight pipe inner wall.

[0052] The "integration" refers to using a weighted scoring method or a logical rule engine to fuse the qualitative or quantitative indicators obtained from the first three steps into a unified risk criterion. Low risk, medium risk, and high risk levels correspond to different levels of urgency in handling the situation. Specifically, the system sets the following judgment logic: when the temperature difference probability level is low, and no obvious non-periodic fluctuations in the liquid level are detected, while the internal blowing pressure curve characteristics are normal, it is judged as low risk, indicating that the straight pipe inner wall is unobstructed and there are no hidden dangers; when the temperature difference probability level is medium, or there are slight fluctuations in the liquid level, or the pressure curve is slightly abnormal but not exceeding the standard, it is judged as medium risk, suggesting that attention should be paid to fine-tuning the process parameters; when the temperature difference probability level is high, accompanied by significant non-periodic fluctuations in the liquid level, or the internal blowing pressure rise time is severely delayed and the peak value is extremely unstable, it is directly judged as high risk, indicating that a pipe blockage accident is imminent. Through this multi-physical quantity fusion judgment mechanism, the problem of false alarms from a single indicator is effectively avoided. For example, a large temperature difference with a normal pressure curve may simply indicate insufficient preheating of the steel pipe without any substantial blockage. In this case, it would be classified as medium risk rather than high risk, thus avoiding unnecessary shutdown intervention. The final risk level of the straight pipe inner wall will serve as an important component of the regional risk characteristic information set and will be transmitted to the subsequent early warning decision-making module.

[0053] Example 6: In one optional implementation, this application provides a process for generating an end thickness deviation index, the method comprising the following steps: Step 1: Based on the zinc immersion angle, zinc immersion speed, dwell time, and pull-out speed, and in conjunction with the zinc bath temperature, establish the theoretical zinc immersion time distribution along the length of the steel pipe; The theoretical zinc immersion time distribution refers to the theoretical duration sequence of contact between the steel pipe surface and the molten zinc, calculated based on the trajectory parameters of the steel pipe within the zinc pot and the thermodynamic environment parameters. This distribution is obtained by substituting the immersion angle, immersion speed, residence time, and pull-out speed into a preset kinematic model, and introducing a correction coefficient for the zinc liquid temperature on the viscosity and fluidity of the zinc liquid. Its function is to quantify the degree of zinc immersion at different lengths of the steel pipe under ideal conditions, providing a benchmark for subsequent judgment of abnormal zinc layer thickness at the ends. Specifically, the immersion angle determines the length of the inclined path when the steel pipe enters and leaves the zinc pot, the immersion speed and pull-out speed directly affect the time per unit length of pipe to pass through the pot, while the residence time compensates for the static reaction time during the complete immersion stage of the steel pipe. For example, when the zinc immersion angle is 45 degrees, the immersion speed is 1.2 m / s, the pull-out speed is 1.0 m / s, and the zinc bath temperature is maintained at 450℃, the system can calculate that the theoretical immersion time in the 0-100 mm region from the pipe end is about 0.5 seconds shorter than that in the middle region, thus forming a trough in the end time on the distribution curve. Through this coupled modeling of multidimensional motion parameters and temperature parameters, the dynamic immersion process of the steel pipe in the zinc pot can be accurately reproduced, avoiding the errors caused by estimating based on a single speed parameter.

[0054] Step 2: In the initial cooling stage after the steel pipe exits the zinc pot, based on the temperature at multiple points on the surface of the steel pipe, identify the changes in the amount of zinc liquid adhering to the end due to the difference in heat dissipation rate. The change in zinc molten metal adhesion refers to the dynamic change in the amount of residual liquid zinc on the steel pipe surface during the cooling process, which is directly related to the final solidified zinc layer thickness. This change is obtained by real-time acquisition of multi-point temperature data along the length of the steel pipe surface using an infrared thermal imager or array temperature sensor placed at the inlet of the cooling section, and by analyzing the difference in temperature drop slope between the end and middle regions. Its function is to capture the accelerated heat dissipation phenomenon caused by the tip effect at the end, and thus infer the loss or shrinkage of the zinc molten metal before solidification. Specifically, because the end of the steel pipe has a large exposed area and no adjacent pipe body for heat conduction support, its heat dissipation rate is significantly higher than that in the middle, causing the zinc molten metal at the end to cool rapidly and increase in viscosity. Some unsolidified zinc molten metal may flow back or thin due to gravity or surface tension. For example, if the temperature within a 50mm range at the end of the pipe drops by 80°C within 3 seconds after being removed from the heat, while the temperature within the same length in the middle drops by only 40°C, it is determined that there is a significant rapid cooling characteristic at that end, corresponding to an unexpected reduction in the amount of zinc molten metal adhesion. By tracking the evolution of this temperature field in real time, the source of coating defects caused by uneven heat exchange can be accurately located.

[0055] Step 3: During the external blowing stage, analyze the difference in external blowing pressure between the end and the middle to determine the difference in local air resistance at the end caused by the change in the thickness of the zinc liquid. The local air resistance difference refers to the resistance fluctuation caused by the uneven zinc layer thickness leading to changes in the cross-sectional area of ​​the airflow channel when a high-speed airflow blows away excess zinc liquid from the surface of the steel pipe. This difference is calculated by comparing the air pressure readings of the external blowing nozzles acting on the end region of the steel pipe with those acting on the middle region, and calculating the real-time difference between the two. Its purpose is to verify the actual accumulation state of the zinc layer at the end, because a thicker zinc layer will narrow the airflow channel, thus causing an increase in back pressure, while a thinner zinc layer may lead to airflow leakage or pressure reduction. Specifically, the external blowing system usually maintains a constant air supply flow rate. When the airflow passes through the end with a thicker zinc layer, the flow cross-sectional area decreases, resulting in an increase in local static pressure; conversely, if the zinc layer at the end is thinner, the air resistance decreases, and the pressure reading decreases accordingly. For example, if the reference external blowing pressure in the middle is set to 0.6 MPa, if the pressure reading in the end region is consistently stable at 0.75 MPa, it indicates that the zinc layer thickness at that location is significantly greater than in the middle, forming a large local air resistance; if the reading is only 0.5 MPa, it suggests that the zinc layer at the end may be too thin. By analyzing this pressure difference characteristic, the spatial unevenness of zinc layer thickness can be indirectly but sensitively reflected from a fluid dynamics perspective.

[0056] Step 4: Based on the theoretical zinc immersion time distribution, the change in zinc liquid adhesion amount, and the local gas resistance difference, by setting thin and thick thresholds, determine the offset direction and offset level of the zinc layer thickness at the end relative to the middle, and generate the end thickness deviation index.

[0057] The end thickness deviation index is a quantitative risk indicator used to characterize the degree and direction of deviation of the zinc layer thickness at the end of the steel pipe from the standard value. This index is obtained by normalizing the theoretical zinc immersion time deviation, temperature drop rate difference, and external blowing pressure difference obtained in the preceding steps, assigning corresponding weight coefficients, and then weighting and fusing them. Its function is to unify multi-source heterogeneous physical quantities into an intuitive risk level for subsequent early warning decisions. Specifically, the system presets a lower threshold for thinner thickness (e.g., -15%) and an upper threshold for thicker thickness (e.g., +20%). When the calculated index falls into the negative over-threshold range, it is determined to be a risk of end thinning; when it falls into the positive over-threshold range, it is determined to be a risk of end thickening. The deviation is further classified into slight, moderate, and severe levels based on the magnitude of the exceedance. For example, if the theoretical model shows insufficient zinc immersion, temperature monitoring confirms rapid cooling, and the external blowing pressure shows low air resistance, all three pointing to the same conclusion, the generated end thickness deviation index will show a high-confidence negative value, directly triggering a thinning warning. This cross-validation and comprehensive judgment using multiple evidence chains improves the accuracy and robustness of end-coating quality assessment.

[0058] Example 7: In one possible implementation, this application embodiment provides a process for generating a comprehensive defect risk index for irregularly shaped segments, the method comprising the following steps: Step 1: During the zinc immersion stage of the steel pipe, calculate the change in flow resistance of the zinc liquid as it flows through the irregular section in real time. When the resistance suddenly increases, it is determined that the zinc liquid has solidified prematurely or that there is slag blockage. The change in flow resistance refers to the dynamic pressure loss caused by abrupt changes in the flow channel cross-section, flow direction, or obstruction by internal foreign objects when molten zinc flows through irregularly shaped sections such as bends, right angles, or diameter changes. This data is obtained by real-time acquisition of differential pressure signals by high-precision differential pressure sensors installed at the inlet and outlet of the zinc pot, combined with the zinc immersion rate. Its function is to serve as the primary basis for determining whether any fluid dynamic anomalies have occurred within the irregularly shaped sections. Specifically, the system presets a baseline flow resistance curve. When the real-time calculated flow resistance value exceeds the baseline value by a certain percentage within a short period (e.g., a sudden increase exceeding 15%), it is considered a sudden increase in resistance. This sudden increase usually indicates that the molten zinc has solidified prematurely at the dead corners of the irregularly shaped section due to excessive heat dissipation, or that zinc oxide slag has accumulated and blocked the flow channel. By monitoring changes in flow resistance, the physical blockage state inside the zinc immersion process can be accurately detected in the early stages, providing crucial fluid dynamic characteristic input for subsequent risk assessment.

[0059] Step 2: In the initial cooling stage after the steel pipe exits the zinc pot, identify cold spots or hot spots in the temperature field caused by zinc residue or slag inclusions to obtain abnormal temperature field information. Temperature field anomaly information refers to abnormal temperature distribution characteristics exhibited by specific areas on the steel pipe surface due to uneven zinc layer thickness or the presence of inclusions. This information is obtained by analyzing high-resolution thermal imaging data acquired through a full-width scan of the steel pipe surface immediately after it exits the zinc bath using an infrared thermal imager. Its function is to locate potential defect areas by utilizing differences in thermal conductivity. Specifically, because the specific heat capacity of molten zinc differs from that of the steel substrate, and the thermal conductivity varies at inclusions, the heat dissipation rate of defective areas differs from that of normal areas in the initial cooling phase. If a region has a thick layer of molten zinc residue, its heat storage is large, and its cooling is slow, appearing as a hot spot relative to the surrounding area in the thermal image; if a region has inclusions or an excessively thin zinc layer, its heat dissipation is rapid, appearing as a cold spot. For example, when analyzing thermal imaging data, the system sets a temperature gradient threshold. When the temperature difference between a local area of ​​the irregular section and the adjacent normal area exceeds 5°C and lasts for more than 2 seconds, it is marked as a temperature field anomaly. By identifying temperature field anomaly information, the material adhesion on the surface of the irregular section can be verified from a thermal perspective, complementing changes in flow resistance.

[0060] Step 3: During the cooling stage of the steel pipe, capture the micro-vibration signal generated by the release of residual stress in the irregular section when the steel pipe passes through, and analyze the abnormal amplitude of the micro-vibration. The micro-vibration anomaly amplitude refers to the mechanical vibration intensity index caused by uneven mass distribution in the irregularly shaped section (such as center of gravity shift due to zinc residue) or concentrated release of internal stress during the cooling and shrinking process of the steel pipe. This signal is acquired by a high-sensitivity accelerometer installed on the base or bracket of the conveyor roller in the cooling section. Its function is to reveal the mechanical stability and mass distribution uniformity of the internal structure of the irregularly shaped section. Specifically, when the steel pipe passes through the cooling section, the vibration spectrum of a normal straight pipe section is relatively stable, while if there is severe zinc residue or slag inclusion in the irregularly shaped section, it will cause an increase in the mass of that section and an asymmetrical distribution, which will then generate specific low-frequency shaking or high-frequency impact vibration due to the sudden release of residual stress during movement. The system performs Fast Fourier Transform (FFT) analysis on the acquired vibration signal and extracts the energy peak value in a specific frequency band as the micro-vibration anomaly amplitude. For example, when the vibration amplitude detected in the frequency range of 50Hz-200Hz exceeds three times the background noise reference value, it is determined to be a micro-vibration anomaly. By analyzing the abnormal amplitude of micro-vibrations, it is possible to further confirm whether there is substantial defect accumulation in the irregular segment from the perspective of mechanical response.

[0061] Step 4: Based on the combined information of changes in flow resistance, abnormal temperature field, and abnormal amplitude of micro-vibration, a comprehensive defect risk index for irregular sections is obtained through weighted fusion, and the defect type is determined to be zinc liquid residue or slag accumulation.

[0062] The comprehensive defect risk index for irregularly shaped segments is a numerical indicator that quantitatively characterizes the overall health status of irregularly shaped geometric segments, typically ranging from 0 to 1. This index is obtained by multiplying the normalized values ​​of flow resistance change, temperature field anomaly information, and micro-vibration anomaly amplitude obtained from the aforementioned three steps by preset weighting coefficients and then summing them. Its function is to achieve deep fusion of multimodal sensing data, thereby improving the accuracy and robustness of defect identification. Specifically, the weighting coefficients can be set based on historical data statistics or expert experience; for example, a weight of 0.4 is assigned to flow resistance change, 0.3 to temperature field anomaly information, and 0.3 to micro-vibration anomaly amplitude. When determining the defect type, the system uses logical rule matching: if the flow resistance increases significantly and is accompanied by hot spot characteristics, it tends to be identified as zinc liquid residue; if the flow resistance increases significantly and is accompanied by cold spot characteristics and a large micro-vibration amplitude, it tends to be identified as slag inclusion accumulation. Through this weighted fusion strategy, false alarms from a single sensor can be corrected by normal data from other sensors, thereby ensuring that the final generated comprehensive defect risk index for irregularly shaped sections can truly reflect the quality status of the production site and effectively solve the technical problem of difficult online accurate detection of complex pipe fittings.

[0063] Example 8: In another optional embodiment, this application provides a method for generating and outputting an intelligent early warning report for process parameters, the method comprising the following steps: Step 1: Based on the risk characteristic information set of different regions, perform multi-segment comprehensive risk assessment on different sections of each steel pipe to determine the comprehensive health level of each steel pipe; This step aims to integrate the generated dispersed regional risk data into a single quantitative indicator characterizing the overall quality status of the steel pipe. The regional risk characteristic information set includes the risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of irregularly shaped sections. These indicators reflect the potential defect probability of different geometric sections of the steel pipe. The multi-segment risk comprehensive judgment process employs priority weighting logic. Given that irregularly shaped geometric sections (such as elbows and diameter changes) have the highest probability and greatest hazard of slag inclusions or residues, the system sets the comprehensive defect risk index of this region as the highest priority weight. Specifically, when the comprehensive defect risk index of an irregularly shaped section exceeds a preset high-risk threshold, regardless of the risk status of the straight pipe section and the end section, the overall health level of the steel pipe is directly determined to be high-risk. When the irregularly shaped section is not in a high-risk state, the combined status of the straight pipe inner wall risk level and the end thickness deviation index is further combined, and the comprehensive health level is determined to be either a general concern level or a warning level through a lookup table method or fuzzy reasoning algorithm. For example, if the risk index of an irregularly shaped section of a steel pipe is 0.3 (low risk), but the risk level of the inner wall of the straight pipe is medium risk and the end thickness deviation index shows that it is too thick, the system calculates a weighted overall health level as "general concern," indicating that although the steel pipe has no fatal defects, subsequent process fluctuations need to be monitored. This determined overall health level provides a clear decision-making basis for matching subsequent treatment strategies.

[0064] Step 2: Based on the overall health level, match the corresponding recommended handling strategy from the preset handling strategy library. The recommended handling strategies include maintaining the current parameter operation, suggesting adjustment of specific process parameters, triggering secondary auxiliary treatment, or offline testing. The preset handling strategy library consists of rule mapping tables pre-stored in the workshop control server or cloud platform database, establishing a one-to-one correspondence between the overall health level and specific actions. The specific content of the recommended handling strategies is distributed in a tiered manner according to the severity of the risk level: for steel pipes with a warning level overall health, the recommended handling strategy is to maintain the current parameters and only need to be recorded in the log; for those with a general concern level, the strategy is to suggest adjusting specific process parameters, such as automatically calculating and issuing instructions to fine-tune the zinc immersion speed or internal blowing pressure setting to compensate for detection deviations; for steel pipes with local defect risks, the strategy is to trigger secondary auxiliary treatment, such as activating a dedicated pulse purging device for elbows or performing localized supplementary blowing at the ends; and for steel pipes determined to be high-risk, an offline inspection strategy is matched, immediately triggering an audible and visual alarm and controlling a robotic arm to divert the steel pipe to a re-inspection station to prevent non-conforming products from flowing into the next process. For example, when the system determines that a steel pipe with a 90-degree bend has a high overall health level, it automatically retrieves the offline inspection instruction from the strategy library and controls the sorting cylinder to operate via a PLC output signal. Simultaneously, it sends specific instructions to the operating terminal to divert the steel pipe to the inspection platform for endoscopic examination. This tiered matching mechanism achieves a seamless connection from risk identification to automated intervention, reducing the lag and misjudgment rate of manual judgment.

[0065] Step 3: Integrate the comprehensive health level, risk details of each region contained in the risk characteristic information set of each region, and recommended treatment strategies to generate an intelligent early warning report of process parameters, and output it to the workshop control visualization interface.

[0066] The intelligent early warning report for process parameters is a structured data message. Its content not only includes the final comprehensive health level conclusion but also fully retains detailed risk assessment data for the straight pipe inner wall, end sections, and irregularly shaped geometric sections (such as specific temperature difference values, pressure curve characteristic values, vibration amplitude, etc.), ensuring that managers can trace the root cause of defects. The workshop control visualization interface can be a Weintek touchscreen, industrial monitor, or cloud management screen deployed on the production site, supporting graphical display of report content. When generating the report, the system encapsulates the above comprehensive judgment results, risk details text for each segment, and the matched recommended handling strategy instructions into a JSON or XML format data packet, and pushes it to the visualization interface in real time via industrial Ethernet. For example, after the report is generated, the visualization interface immediately pops up a window displaying: Steel pipe number #20240520-08, comprehensive level: high risk, main risk source: zinc liquid residue in irregularly shaped sections, recommended measures: offline inspection, allowing operators to quickly execute corresponding operations. This step bridges the final gap between data analysis and production management, making early warning information transparent and timely, and improving the workshop's response speed to sudden quality anomalies.

[0067] Example 9: In another embodiment, this application provides a multi-segment risk comprehensive judgment method, which further includes specific steps for prioritizing and classifying the risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular segment.

[0068] Step 1: Prioritize the risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular section, with the comprehensive defect risk index of the irregular section having the highest priority. Priority ranking refers to the logical process by which the system, after receiving a set of risk characteristic information for different regions, pre-determines judgment weights based on the severity of the impact of defects in different geometric segments on the overall quality of the steel pipe. The comprehensive defect risk index for irregularly shaped segments is set with the highest priority because zinc residue or slag accumulation in irregularly shaped geometric segments such as elbows and diameter changes often leads to complete local blockage or structural failure of the steel pipe. The harm caused by this is irreversible and global, far exceeding the minor zinc plating or end thickness deviation in straight pipe sections. The risk level of the straight pipe inner wall reflects the possibility of airflow obstruction, while the end thickness deviation index characterizes the degree of deviation in the uniformity of the anti-corrosion layer; both are secondary in the judgment logic. This priority setting based on critical item veto ensures that the system can prioritize and respond to the most critical process defects during decision-making. For example, when the system detects that the comprehensive defect risk index of an irregularly shaped section of a steel pipe is 0.85 (high risk), while the risk level of the straight pipe inner wall is low risk and the end thickness deviation index is within the normal range, the sorting logic will immediately lock the high-risk status of the irregularly shaped section and ignore the low-risk signals of the other two sections, thereby preventing the risk from being diluted by averaging. This step aims to establish the core criteria for risk assessment, providing clear logical input for subsequent hierarchical decisions and effectively preventing the situation where the normality of secondary areas masks serious defects in major areas.

[0069] Step 2: When the comprehensive defect risk index of the irregular segment is in a high-risk state, the comprehensive health level is directly determined to be high-risk; Direct judgment refers to the system's action of skipping the combined analysis of other risk indices and forcibly outputting the highest-level warning signal when specific triggering conditions are met. A high-risk state typically corresponds to the comprehensive defect risk index of the irregularly shaped section exceeding a preset danger threshold (e.g., an index value greater than 0.7), indicating an extremely high probability of zinc liquid solidification blockage or severe slag inclusion in that area. At this point, regardless of whether the inner wall of the straight pipe is unobstructed or the zinc layer at the end is uniform, the overall usability of the steel pipe has been compromised, and continued flow will lead to damage to subsequent process equipment or the generation of scrap. The main body executing this step is the early warning decision module, whose function is to implement a rapid melting mechanism to ensure that high-risk steel pipes can be intercepted immediately. For example, if a sudden increase in flow resistance accompanied by obvious cold spots is detected in the irregularly shaped section, and the calculated risk index reaches 0.82, the system does not need to wait for the final stabilization analysis of the blowing pressure curve in the straight pipe section; it immediately marks the comprehensive health level of the steel pipe as high-risk. This mechanism shortens the time delay from defect identification to alarm output, reflecting a zero-tolerance strategy for major quality hazards. As a result, the production line can immediately trigger a shutdown interlock or automatic diversion command to prevent defective steel pipes from entering the cooling bed or packaging area, thereby reducing rework costs and safety hazards.

[0070] Step 3: When the comprehensive defect risk index of the irregular section is not in a high-risk state, the comprehensive health level is determined to be either a general concern level or a warning level based on the combination of the risk level of the straight pipe inner wall and the end thickness deviation index.

[0071] In this context, a "non-high-risk" state refers to a situation where the comprehensive defect risk index of the irregularly shaped section is below the danger threshold, indicating that there is no serious blockage or slag accumulation at the irregular structure, and the steel pipe has the basic conditions for continued processing. Under this premise, the system's judgment logic switches to a refined evaluation mode, focusing on the risk combination of the straight pipe section and the end section. The combination state can refer to a matrix matching of the risk level of the straight pipe inner wall (e.g., low, medium, high) and the end thickness deviation index (e.g., too thin, too thick, normal). The "general concern" level usually corresponds to a moderate risk of zinc plating in the straight pipe section or a slight thickness deviation at the end. Although it does not affect immediate production safety, process parameters need to be adjusted to prevent deterioration. The "warning" level corresponds to both sections being in a low-risk or slightly deviation state, requiring only routine recording. For example, when the risk index of the irregularly shaped section is 0.3 (non-high risk), while the risk level of the straight pipe inner wall shows medium risk (the peak internal blowing pressure is slightly delayed), and the end thickness deviation index shows a slight over-thickness, the system will determine the comprehensive health level as "general concern," suggesting that the operator fine-tune the internal blowing pressure or check the zinc liquid temperature. Conversely, if both indicators are within the ideal range, it is classified as a warning level, indicating that the current process is operating smoothly. This step enables tiered management of non-fatal defects, avoiding production disruptions caused by excessive alarms, while ensuring continuous monitoring of potential quality fluctuations.

[0072] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0073] Figure 3 This application provides a schematic diagram of the structure of an intelligent early warning system for process parameters in a galvanizing workshop, as shown in one embodiment. Figure 3 As shown in this embodiment, an intelligent early warning system 300 for process parameters in a galvanizing workshop includes: a steel pipe segmentation module 301, a risk characteristic module 302, and an intelligent early warning module 303.

[0074] The steel pipe segmentation module 301 is used to acquire a multi-source process parameter set of the steel pipe to be galvanized, and based on the multi-source process parameter set, to perform steel pipe geometric feature identification and physical segment division, and generate a steel pipe segmentation feature information set including straight pipe segments, end segments and irregular geometric segments; The risk feature module 302 is used to perform dynamic blockage risk assessment on the inner wall of the straight pipe section based on the segment feature information set of the steel pipe, perform zinc layer thickness distribution offset prediction on the end section, and perform zinc liquid residue and slag accumulation risk assessment on the irregular geometric section, and generate regional risk feature information set. The intelligent early warning module 303 is used to perform multi-segment risk comprehensive judgment and early warning level decision based on the regional risk characteristic information set, and generate and output intelligent early warning reports of process parameters.

[0075] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A method for intelligent early warning of process parameters in a galvanizing workshop, characterized in that, include: A multi-source process parameter set of the steel pipe to be galvanized is obtained. Based on the multi-source process parameter set, the geometric features of the steel pipe are identified and the physical segments are divided to generate a steel pipe segment feature information set including straight pipe segments, end segments and irregular geometric segments. Based on the segmented feature information set of the steel pipe, a dynamic blockage risk assessment of the inner wall is performed on the straight pipe segment, a zinc layer thickness distribution offset prediction is performed on the end segment, and a risk assessment of zinc liquid residue and slag accumulation is performed on the irregular geometric segment, generating a risk feature information set for different regions. Based on the aforementioned regional risk characteristic information set, multi-segment risk comprehensive judgment and early warning level decision are performed, and intelligent early warning reports of process parameters are generated and output.

2. The method according to claim 1, characterized in that, The process of generating the segmented feature information set of the steel pipe includes: Read the digital control database of the galvanizing process to obtain the set of physical parameters of the galvanized steel pipe; By using a sensor array arranged in the zinc pot, internal and external blowing stations and cooling section, real-time process parameter streams including zinc liquid temperature, steel pipe temperature, internal blowing pressure, external blowing pressure, zinc immersion angle, zinc immersion speed, residence time, pull-out speed and multi-point temperature of steel pipe surface are collected and combined to form the multi-source process parameter set. Based on the physical parameter set of the specifications, the geometric features of the steel pipe are identified and the physical segments are divided. Each steel pipe is divided along its length into straight pipe segments, end segments, and irregular geometric segments including bends, right angles, and diameter change areas, thereby generating the steel pipe segment feature information set.

3. The method according to claim 2, characterized in that, The process of identifying the geometric features of the steel pipe and dividing it into physical segments includes: Based on the physical parameter set of the specifications, the bend angle, bending radius, starting position of diameter change and diameter change ratio of the galvanized steel pipe are extracted and marked as the irregular geometric segment; Using the two ends of the irregular geometric segment as the dividing points, the remaining pipe body is divided into straight pipe segments along the length direction, and each predefined length range at both ends of the straight pipe segment is separately marked as the end segment.

4. The method according to claim 2, characterized in that, The process of generating the regional risk characteristic information set includes: Based on the segmented feature information set of the steel pipe, perform independent area risk assessments in parallel: For the straight pipe section, assess the risk of zinc nodules and pipe blockage on the inner wall to generate a risk level for the inner wall of the straight pipe; For the end segment, the offset between the end zinc layer thickness and the middle thickness is predicted, and an end thickness deviation index is generated; For the irregular geometric segment, assess the risk of zinc residue and slag accumulation, and generate a comprehensive defect risk index for the irregular segment; The risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular section are combined to generate the risk feature information set of the sub-regions.

5. The method according to claim 4, characterized in that, The process of generating the risk level of the inner wall of the straight pipe includes: Analyze the real-time temperature difference between the zinc liquid temperature and the steel pipe temperature to determine the probability level of premature solidification of the zinc liquid at the front end of the straight pipe inner wall. During the zinc immersion stage of steel pipes, non-periodic changes in the zinc pot liquid level fluctuations are monitored to infer the degree of local blockage on the inner wall; During the internal blowing stage, the rise time and peak stabilization time of the internal blowing pressure curve are analyzed to identify the obstruction strength of the zinc nodule on the airflow and obtain the characteristics of the internal blowing pressure curve. Based on the combined characteristics of the temperature difference probability level, the degree of local blockage of the inner wall, and the internal blowing pressure curve, the risk of the straight pipe inner wall is divided into three levels: low risk, medium risk, and high risk, thus generating the risk level of the straight pipe inner wall.

6. The method according to claim 4, characterized in that, The process of generating the end thickness deviation index includes: Based on the zinc immersion angle, the zinc immersion speed, the dwell time, and the pull-out speed, and in conjunction with the zinc bath temperature, a theoretical zinc immersion time distribution along the length of the steel pipe is established. In the initial cooling stage after the steel pipe exits the zinc pot, the change in the amount of zinc liquid adhering to the end due to the difference in heat dissipation rate is identified based on the temperature of multiple points on the surface of the steel pipe. During the external blowing stage, the difference in external blowing pressure between the end and the middle is analyzed to determine the local air resistance difference at the end caused by the change in the thickness of the zinc liquid. Based on the theoretical zinc immersion time distribution, the changes in zinc liquid adhesion, and the local air resistance differences, the offset direction and offset level of the end zinc layer thickness relative to the middle are determined by setting a thinner threshold and a thicker threshold, and the end thickness deviation index is generated.

7. The method according to claim 4, characterized in that, The process of generating the comprehensive defect risk index for the irregular segment includes: During the zinc immersion stage of steel pipe, the change in flow resistance of zinc liquid as it flows through the irregular section is calculated in real time. When the resistance suddenly increases, it is determined that the zinc liquid solidifies prematurely or is blocked by slag. In the initial cooling stage after the steel pipe exits the zinc pot, cold spots or hot spots in the temperature field caused by zinc liquid residue or slag inclusions are identified to obtain information on temperature field anomalies. During the cooling stage of the steel pipe, the micro-vibration signal generated by the release of residual stress in the irregular section when the steel pipe passes through is captured, and the abnormal amplitude of micro-vibration is analyzed. Based on the combined information of the flow resistance change, the abnormal temperature field, and the abnormal amplitude of the micro-vibration, a comprehensive defect risk index for the irregular segment is obtained through weighted fusion, and the defect type is determined to be zinc liquid residue or slag accumulation.

8. The method according to claim 4, characterized in that, The process of generating and outputting the intelligent early warning report for the process parameters includes: Based on the aforementioned regional risk characteristic information set, multi-segment risk comprehensive judgment is performed on different sections of each steel pipe to determine the comprehensive health level of each steel pipe; Based on the comprehensive health level, a corresponding recommended treatment strategy is matched from the preset treatment strategy library. The recommended treatment strategy includes maintaining the current parameter operation, suggesting the selection of at least one target process parameter from the multi-source process parameter set for adjustment, triggering secondary auxiliary processing, or offline detection. The target process parameters include at least one of the following: zinc bath temperature, steel pipe temperature, internal blowing pressure, external blowing pressure, zinc immersion angle, zinc immersion speed, residence time, and pull-out speed. The comprehensive health level, the risk details of each region included in the regional risk characteristic information set, and the recommended treatment strategy are integrated to generate the intelligent early warning report of process parameters, and output to the workshop control visualization interface.

9. The method according to claim 8, characterized in that, The multi-segment risk comprehensive assessment includes: The risk level of the straight pipe inner wall, the end thickness deviation index, and the comprehensive defect risk index of the irregular section are prioritized, with the comprehensive defect risk index of the irregular section having the highest priority. When the comprehensive defect risk index of the irregular segment is in a high-risk state, the comprehensive health level is directly determined to be high-risk. When the comprehensive defect risk index of the irregular segment is in a non-high-risk state, the comprehensive health level is determined to be either a general concern level or a warning level based on the combination of the risk level of the straight pipe inner wall and the end thickness deviation index.

10. A process parameter intelligent early warning system for galvanizing workshops, characterized in that, The method applied to any one of claims 1-9 includes: The steel pipe segmentation module is used to acquire a multi-source process parameter set of the steel pipe to be galvanized, and based on the multi-source process parameter set, to perform steel pipe geometric feature identification and physical segment division, and generate a steel pipe segmentation feature information set including straight pipe segments, end segments and irregular geometric segments; The risk feature module is used to perform dynamic blockage risk assessment on the inner wall of the straight pipe section based on the segment feature information set of the steel pipe, perform zinc layer thickness distribution offset prediction on the end section, and perform zinc liquid residue and slag accumulation risk assessment on the irregular geometric section, and generate regional risk feature information set. The intelligent early warning module is used to make comprehensive judgments on multi-segment risks and make early warning level decisions based on the risk feature information set of the sub-regions, and generate and output intelligent early warning reports of process parameters.

Images