A method and system for informationized monitoring and management of a galvanization production process

Abstract

This invention discloses an information-based monitoring and management method and system for galvanizing production processes. Utilizing a SCADA+SKF multi-parameter visualization monitoring platform, it acquires the internal and surface defect characteristics of production components corresponding to risk nodes where oxidized mixtures intrude at unit temperatures. Combining data from manually involved processes, impurity introduction, acid washing concentration, and temperature and humidity changes, it obtains an oxidized mixture distribution impact assessment model through an edge computing gateway, generates an oxidized mixture distribution matrix per unit time, and conducts impact degree analysis and impurity source tracing to calculate the galvanizing compliance rate. Furthermore, through a bidirectional categorical variable chi-square test of defects and compliance rate, it identifies key interference factors and stores the quality assessment results, achieving intelligent monitoring and optimized management of the entire galvanizing process. This invention realizes closed-loop intelligent control from data acquisition and process modeling to quality analysis, significantly improving the stability, automation level, and product compliance rate of the galvanizing process.

Description

Technical Field

[0001] This invention relates to the field of galvanizing production process supervision, and in particular to an information-based monitoring and management method and system for galvanizing production processes. Background Technology

[0002] In traditional galvanizing production processes, the risk of oxide mixture intrusion, defect identification of production components, and coating quality control remain key factors affecting product performance and production efficiency. However, existing monitoring methods mainly rely on manual sampling or low-frequency data collection, making it difficult to achieve real-time perception and intelligent analysis of the entire production process. This extensive management model not only leads to delayed identification of defective components but also fails to promptly detect the potential impact of oxide mixtures on galvanizing effects under different temperature and humidity conditions, severely restricting the consistency and compliance rate of galvanized product quality.

[0003] Existing technologies often suffer from information silos, meaning that data from different production stages lacks unified integration and intelligent processing mechanisms, making it difficult to accurately track and analyze the causes of oxidized mixture intrusion. For complex component defects (such as porosity, cracks, and color differences), the lack of systematic identification models and weighting mechanisms leads to significant subjectivity in analyzing the impact and origin of defects. Furthermore, the lack of digital quantification of impurity introduction during human intervention makes it difficult to accurately assess quality fluctuations caused by human factors in production, limiting the scope for process optimization.

[0004] Furthermore, existing galvanizing production processes typically fail to incorporate advanced multi-parameter fusion algorithmic intelligent technologies, resulting in low utilization of large-scale heterogeneous data and difficulty in effectively eliminating data noise. This leads to low accuracy and limited predictive capabilities in quality assessment models, failing to provide effective support for production management. Faced with increasingly stringent quality requirements and energy conservation and emission reduction targets, existing technologies are no longer sufficient to meet the urgent needs of modern galvanizing production for high-precision, automated, and intelligent management. Therefore, there is an urgent need to construct an information-based monitoring and management method and system for the galvanizing production process to promote the digital upgrade and intelligent optimization of the entire process. Summary of the Invention

[0005] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides an information-based monitoring and management method and system for galvanizing production processes.

[0006] Firstly, this application provides an information-based monitoring and management method for a galvanizing production process, including:

[0007] The internal and surface defect characteristics of the production components corresponding to the risk nodes of oxidation mixture intrusion at a unit temperature in the galvanizing production process are obtained by using the SCADA+SKF multi-parameter visualization monitoring platform preset in the galvanizing production process. The internal and surface defect characteristics of the production components include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the components, and the location of coating peeling.

[0008] Obtain data on impurities introduced during manual cleaning processes in the galvanizing production process under different production humidity conditions.

[0009] For the risk node of the intrusion of the oxide mixture at a unit temperature, the production information management center of the edge computing gateway obtains the oxide mixture distribution impact assessment model, oxide mixture distribution information under different acid washing concentrations, galvanizing temperature fluctuation range, and galvanizing solution concentration ratio corresponding to the oxide mixture intrusion risk node. The internal and surface defect characteristics of the production components corresponding to the oxide mixture intrusion risk node are input into the oxide mixture distribution impact assessment model to obtain the oxide mixture distribution matrix per unit time corresponding to the oxide mixture intrusion risk node. The impact degree analysis is performed on the oxide mixture distribution at a unit temperature based on the data brought in by impurities in the manual intervention process, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio.

[0010] Based on the impact degree analysis results of the distribution of the oxide mixture under the impact assessment at a unit temperature, the impurity source tracing analysis of the distribution of the oxide mixture under the impact assessment at a unit temperature is carried out to obtain the galvanizing compliance rate of the oxide mixture distribution in the galvanizing production process.

[0011] Further, the step of inputting the internal and surface defect characteristics of the production component corresponding to the oxidation mixture intrusion risk node into the oxidation mixture distribution impact assessment model to obtain the unit-time oxidation mixture distribution matrix corresponding to the oxidation mixture intrusion risk node includes:

[0012] The internal and surface defect characteristics of the production components are input into the oxidation mixture distribution influence assessment model; the oxidation mixture distribution influence assessment model includes defect type identification coefficient, porosity / bubble distribution and size correlation matrix, crack length and depth correlation matrix, component surface color difference and flatness, coating peeling location correlation matrix, weight allocation mechanism and multi-head variational map autoencoder;

[0013] The defect type identification coefficients are used to identify the type of defects in the internal and surface features of the production components at a unit temperature, respectively, to obtain the type identification result of the defects at a unit temperature, and the defects at a unit temperature are input into the corresponding correlation matrix based on the type identification result of the defects at a unit temperature.

[0014] The distribution and size correlation matrix of pores / bubbles is used to adapt the matrix elements of the received distribution and size of pores / bubbles to obtain the defect matrix coefficients.

[0015] The crack length and depth correlation matrix is ​​used to adapt the matrix elements of the received crack length and depth to obtain the compliance rate matrix coefficients.

[0016] The correlation matrix between the surface color difference and flatness, and the location of coating peeling of the component is used to adapt the matrix elements of the received surface color difference, flatness, and coating peeling location of the component to obtain the surface and coating matrix coefficients.

[0017] The weight allocation mechanism applies influence weights to the defect matrix coefficients, the compliance rate matrix coefficients, and the surface and coating matrix coefficients to obtain a weight adaptation matrix.

[0018] The multi-head variational graph autoencoder uses a preset denoising rule to denoise the weight adaptation matrix, obtains the denoising result corresponding to the weight adaptation matrix, and then uses a preset decoder to reconstruct the matrix of the denoised result to obtain the distribution matrix of the oxidized mixture per unit time.

[0019] Furthermore, the data on impurities introduced during the manual intervention process includes a dust particle change curve per unit time.

[0020] The analysis of the impact of the data on impurities introduced during the manual intervention process, the distribution information of the oxide mixture under different acid washing concentrations, the fluctuation range of the galvanizing temperature, and the concentration ratio of the galvanizing solution on the distribution of the oxide mixture per unit temperature includes:

[0021] For the dust particle change curve at a unit temperature, the dust particle size / concentration corresponding to the dust particle change curve is determined using a preset dust particle size / concentration classification rule;

[0022] For the distribution of the oxide mixture under the influence assessment at a unit temperature, the degree of influence of the defect characteristics corresponding to the distribution of the oxide mixture under the influence assessment is determined based on the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature and the galvanizing temperature fluctuation range.

[0023] For the distribution of the oxidized mixture under the influence of the unit temperature, the degree of influence of the factors affecting the non-compliance rate of the oxidized mixture distribution is determined according to the concentration ratio of the zinc plating solution.

[0024] For the distribution of the oxidized mixture under unit temperature, the influence of the oxidized mixture distribution under different acid concentrations is analyzed to determine the degree of influence on coating adhesion corresponding to the distribution of the oxidized mixture.

[0025] For the distribution of the oxide mixture under the influence assessment at a unit temperature, the influence degree analysis of defect characteristics, the influence degree analysis of non-compliance factors, and the influence degree analysis of coating adhesion corresponding to the distribution of the oxide mixture under the influence assessment are weighted using preset weight setting rules to obtain the influence degree analysis results corresponding to the distribution of the oxide mixture under the influence assessment.

[0026] Furthermore, the galvanizing temperature fluctuation range includes the temperature anomaly range corresponding to different dust particle sizes / concentrations at a unit temperature for the distribution of the oxide mixture. The analysis of the degree of influence of the defect characteristics corresponding to the distribution of the oxide mixture, based on the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature and the galvanizing temperature fluctuation range, includes:

[0027] Within the galvanizing temperature fluctuation range, determine the pollution parameters corresponding to the dust particle size / concentration of the dust particle size distribution of the oxidized mixture under the dust particle change curve at a unit temperature;

[0028] Clustering of the pollution parameters at a unit temperature yields the results of the analysis of the degree of influence of the defect characteristics.

[0029] Further, the step of determining the degree of influence of the oxide mixture distribution on the coating adhesion corresponding to the influence assessment of the oxide mixture distribution based on the distribution information of the oxide mixture under different acid concentrations includes:

[0030] Based on the distribution information of the oxidized mixture under different washing acid concentrations, determine whether the aforementioned impact on the distribution of the oxidized mixture occurs;

[0031] When the factors affecting the coating adhesion are not present, the factors affecting the coating adhesion are determined. When the factors are present, the distribution pattern of the oxide mixture under different acid concentrations is determined based on the distribution information of the oxide mixture. The pattern is then verified using a graph neural network model to obtain the analysis results of the degree of influence on the coating adhesion.

[0032] Furthermore, the method also includes:

[0033] For the risk node of the oxidized mixture intrusion at a unit temperature, a two-way classification variable chi-square test is performed on the risk node of the oxidized mixture intrusion and the internal and surface defect characteristics of the production component corresponding to the risk node of the oxidized mixture intrusion to obtain the two-way classification variable chi-square test result of the defect;

[0034] The chi-square test results of the two-way classification variables of the defects at unit temperature are subjected to a two-way classification variable chi-square test with the zinc plating compliance rate of the oxide mixture distribution to obtain the compliance rate two-way classification variable chi-square test results, and the compliance rate two-way classification variable chi-square test results are stored in the production information management center of the edge computing gateway.

[0035] Further, the step of performing a two-way chi-square test on the intrusion risk node of the oxide mixture and the internal and surface defect characteristics of the corresponding production component to obtain the two-way chi-square test result of the defect classification variable includes:

[0036] Obtain the production condition matrix corresponding to the nodes at risk of oxidized mixture intrusion;

[0037] The influence factors of internal and surface defects of the production components are calculated using a preset defect feature identification method to obtain the defect identification hazard coefficient corresponding to the defect feature;

[0038] The production condition matrix is ​​convolved with the defect identification hazard coefficient to obtain the chi-square test result of the two-way classification variable of the defect.

[0039] Further, the step of performing a two-way chi-square test on the defect bidirectional classification variable at unit temperature and the zinc plating compliance rate of the oxide mixture distribution to obtain the compliance rate two-way classification variable chi-square test result includes:

[0040] Using a pre-defined visualization chart, the chi-square test results of the two-way categorical variable of defects per unit temperature are presented as a curve to obtain the defect fluctuation curve;

[0041] The zinc plating compliance rate of the oxide mixture was plotted using a preset curve identification method to obtain a compliance rate fluctuation curve; the defect fluctuation curve and the compliance rate fluctuation curve did not show the same interfering factors.

[0042] Analyze whether the defect fluctuation curve and the compliance rate fluctuation curve are similar curves;

[0043] When the analysis results are not identical curves, the non-human factor error sources of the defect fluctuation curve and the non-human factor error sources of the compliance rate fluctuation curve are analyzed and ranked according to the degree of influence. The non-human factor error sources whose influence exceeds the preset value are identified as calibrated non-human factor error sources. The human factor error sources of the defect fluctuation curve and the human factor error sources of the compliance rate fluctuation curve are analyzed and ranked according to the degree of influence. The human factor error sources whose influence exceeds the preset value are identified as calibrated human factor error sources.

[0044] Construct a defect anomaly curve and a compliance rate anomaly curve; the names of the horizontal and vertical axes of the defect anomaly curve and the compliance rate anomaly curve are consistent; the non-human factor error sources of the defect anomaly curve are greater than the calibration non-human factor error sources; the human factor error sources of the defect anomaly curve are greater than the calibration human factor error sources; the interference factors of the defect anomaly curve are the same per unit temperature; and the interference factors of the defect anomaly curve do not appear in the defect fluctuation curve and the compliance rate fluctuation curve.

[0045] The defect identification curve is obtained by replacing the interference factor of the defect abnormality curve in the same production process with the interference factor of the unit temperature of the defect fluctuation curve, and the compliance rate identification curve is obtained by replacing the interference factor of the compliance rate abnormality curve in the same production process with the interference factor of the unit temperature of the compliance rate fluctuation curve.

[0046] By comparing the maximum / minimum values ​​of the defect identification curve and the compliance rate identification curve, the chi-square test results of the two-way categorical variable of compliance rate are obtained.

[0047] On the other hand, this application provides an information-based monitoring and management system for the galvanizing production process, including:

[0048] The defect acquisition unit is used to acquire the internal and surface defect characteristics of the production components corresponding to the risk nodes of oxidation mixture intrusion at a unit temperature in the galvanizing production process through the SCADA+SKF multi-parameter visualization monitoring platform preset in the galvanizing production process; the internal and surface defect characteristics of the production components include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the components, and the location of coating peeling.

[0049] The impurity data acquisition unit is used to acquire data on impurities introduced during the cleaning process involving human intervention in the galvanizing production process under different production humidity conditions.

[0050] The impact analysis unit is used to obtain the impact assessment model of the oxide mixture distribution corresponding to the oxide mixture intrusion risk node at a unit temperature, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio from the production information management center of the edge computing gateway. It also inputs the internal and surface defect characteristics of the production components corresponding to the oxide mixture intrusion risk node into the oxide mixture distribution impact assessment model to obtain the oxide mixture distribution matrix per unit time corresponding to the oxide mixture intrusion risk node. Furthermore, it performs an impact analysis on the distribution of the oxide mixture at a unit temperature based on the data brought in by impurities from the human intervention process, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio.

[0051] The two-way categorical variable chi-square test unit is used to perform impurity source tracing analysis on the distribution of the oxidized mixture under the influence assessment at a unit temperature based on the influence degree analysis results, and to obtain the galvanization compliance rate of the oxidized mixture distribution in the galvanizing production process.

[0052] Beneficial effects:

[0053] This application provides an information-based monitoring and management method and system for galvanizing production processes, overcoming the problems of inaccurate identification of the risk of oxide mixture intrusion, difficulty in tracing the source of impurities, reliance on manual experience for quality control, and low data utilization in existing technologies. By introducing the SCADA+SKF multi-parameter visualization monitoring platform, it achieves multi-dimensional real-time acquisition and monitoring of component defects (such as porosity, cracks, color difference, and coating peeling) in the galvanizing process, effectively compensating for the shortcomings of traditional monitoring methods, such as single data dimensions and slow response. Simultaneously, the system integrates edge computing, graph neural networks, and multi-head variational graph autoencoder technologies to model, denoise, and reconstruct the matrix of defect data, accurately generating the distribution matrix of oxide mixtures per unit time, achieving high-precision process modeling and prediction. The system also uses a bidirectional categorical variable chi-square test between defects and compliance rates, combined with analysis of error sources from both human and non-human factors, to accurately locate key factors affecting quality, significantly improving the accuracy and efficiency of impurity source tracing. Furthermore, the system employs visualized fluctuation curves and identification curve analysis methods to achieve quantitative identification and intervention of abnormal factors in the production process, reducing bias caused by subjective human judgment. This invention realizes closed-loop intelligent control from data acquisition and process modeling to quality analysis, which greatly improves the stability, automation level and product compliance rate of the galvanizing process. Attached Figure Description

[0054] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a flowchart of the method steps of the present invention;

[0056] Figure 2 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0057] It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. The following describes the application in further detail with reference to the accompanying drawings and specific embodiments.

[0058] like Figure 1 As shown in the embodiment of this application, an information-based monitoring and management method for a galvanizing production process is provided. The method includes:

[0059] Step S1: Obtain the internal and surface defect characteristics of the production components corresponding to the risk nodes of oxidation mixture intrusion at a unit temperature in the galvanizing production process through the SCADA+SKF multi-parameter visualization monitoring platform preset in the galvanizing production process; the internal and surface defect characteristics of the production components include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the components, and the location of coating peeling.

[0060] Step S2: Obtain data on impurities introduced during manual intervention in the cleaning process of the galvanizing production process under different production humidity. For the risk nodes of the intrusion of the oxide mixture at a unit temperature, obtain the oxide mixture distribution impact assessment model, oxide mixture distribution information under different acid washing concentrations, galvanizing temperature fluctuation range and galvanizing solution concentration ratio corresponding to the risk nodes of the intrusion of the oxide mixture in the edge computing gateway production information management center.

[0061] Step S3: Input the internal and surface defect characteristics of the production component corresponding to the risk node of the oxidation mixture into the oxidation mixture distribution impact assessment model to obtain the unit time oxidation mixture distribution matrix corresponding to the risk node of the oxidation mixture intrusion.

[0062] Step S4: Analyze the degree of influence of the distribution of the oxidation mixture under different acid concentrations, the galvanizing temperature fluctuation range, and the concentration ratio of the galvanizing solution on the distribution of the oxidation mixture under the influence assessment at a unit temperature.

[0063] Step S5: Based on the influence degree analysis results corresponding to the distribution of the oxide mixture under the influence assessment at a unit temperature, perform impurity source tracing analysis on the distribution of the oxide mixture under the influence assessment at a unit temperature to obtain the galvanizing compliance rate of the oxide mixture distribution in the galvanizing production process.

[0064] Step S6: For the risk nodes of the oxidation mixture intrusion at a unit temperature, perform a two-way classification variable chi-square test on the risk nodes of the oxidation mixture intrusion and the internal and surface defect characteristics of the production components corresponding to the risk nodes of the oxidation mixture intrusion, and obtain the two-way classification variable chi-square test results of the defects.

[0065] Step S7: Perform a two-way classification variable chi-square test on the defect bidirectional classification variable chi-square test result at unit temperature and the zinc plating compliance rate of the oxide mixture distribution to obtain the compliance rate two-way classification variable chi-square test result, and store the compliance rate two-way classification variable chi-square test result in the edge computing gateway production information management center.

[0066] In this embodiment, the internal and surface defect characteristics of production components corresponding to the risk nodes of oxide mixture intrusion at a unit temperature within the galvanizing production process are obtained through a pre-set SCADA+SKF multi-parameter visualization monitoring platform within the galvanizing production process. These internal and surface defect characteristics include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the component, and the location of coating peeling. Specifically, the SCADA+SKF multi-parameter visualization monitoring platform is pre-set at the risk nodes of oxide mixture intrusion at a unit temperature within the galvanizing production process.

[0067] The SCADA+SKF multi-parameter visualization monitoring platform is an industrial intelligent monitoring system that integrates SCADA (Supervisory Control and Data Acquisition) and SKF (Intelligent Operation and Maintenance Solution based on sensor and diagnostic technologies) technologies. SCADA is mainly used for real-time acquisition and control of various parameters on the production line, while the SKF system provides high-precision condition monitoring and fault prediction capabilities. Through this platform, multiple key parameters in the galvanizing production process, such as temperature, humidity, vibration, and pressure, can be centrally monitored in a multi-dimensional, visualized, and real-time manner. The advantages of this solution include: High real-time performance: The platform can monitor key parameters such as temperature and humidity during the galvanizing process in real time, helping to identify high-risk points where oxide mixtures can intrude; Multi-parameter fusion analysis: Combining data from multiple sensors in the SKF system, it can comprehensively analyze internal and surface defects of produced components, such as porosity, cracks, and color differences, improving data accuracy; Visual display: A graphical interface intuitively displays the status of each production stage and component, facilitating rapid decision-making and fault location; Integration with edge computing: The platform supports uploading key data to the edge computing gateway management center, facilitating subsequent data modeling, distribution analysis, and optimized control; Assisted quality assessment and tracking analysis: The data collected by the platform provides a foundation for subsequent defect and impurity tracking analysis, chi-square testing, and other statistical analyses, helping to improve the overall galvanizing compliance rate. In summary, the SCADA+SKF multi-parameter visualization monitoring platform not only enhances the intelligence and accuracy of process control in the galvanizing production process but also provides key data support for quality analysis and optimization.

[0068] The SCADA+SKF multi-parameter visualization monitoring platform can be fixedly installed at the risk node of oxide mixture intrusion at a unit temperature within the galvanizing production process. Alternatively, the SCADA+SKF multi-parameter visualization monitoring platform can be mounted on a drone or satellite, and carried to the imaging location of the risk node of oxide mixture intrusion at a unit temperature to obtain the internal and surface defect characteristics of the production components at the risk node of oxide mixture intrusion at a unit temperature. Understandably, the distribution and size of pores / bubbles can reveal the appearance, color, and shape of buildings, equipment, and other objects within the risk node of oxide mixture intrusion. This can be used to observe the distribution of oxide mixture at the macroscopic level, such as defects in objects or whether equipment is operating normally. The distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of components, and the location of coating peeling help to achieve comprehensive detection of the distribution of oxide mixture within the risk node of oxide mixture intrusion.

[0069] Data on impurity introduction during manual intervention in the cleaning process of the galvanizing production process under different production humidity conditions was obtained. Specifically, this data was obtained from the official website of the Bureau of Human Resource Management for Impurity Introduction during Manual Intervention. This data includes, but is not limited to, changes in temperature, humidity, wind speed, and rainfall during the cleaning process.

[0070] For the risk node of the intrusion of the oxide mixture at a unit temperature, the production information management center of the edge computing gateway obtains the oxide mixture distribution impact assessment model, oxide mixture distribution information under different acid washing concentrations, galvanizing temperature fluctuation range and galvanizing solution concentration ratio corresponding to the risk node of the intrusion of the oxide mixture. The internal and surface defect characteristics of the production components corresponding to the risk node of the intrusion of the oxide mixture are input into the oxide mixture distribution impact assessment model to obtain the oxide mixture distribution matrix per unit time corresponding to the risk node of the intrusion of the oxide mixture. The impact degree analysis is performed on the oxide mixture distribution at a unit temperature based on the data brought in by impurities in the manual intervention process, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range and the galvanizing solution concentration ratio. Specifically, firstly, the target text corresponding to the risk node of oxidation mixture intrusion is obtained from the production information management center of the edge computing gateway through the production condition matrix of the risk node. The target text includes an oxidation mixture distribution impact assessment model, oxidation mixture distribution information under different acid washing concentrations, galvanizing temperature fluctuation range, and galvanizing solution concentration ratio. The oxidation mixture distribution impact assessment model is a pre-trained deep learning model. The galvanizing temperature fluctuation range includes the temperature anomaly intervals corresponding to different dust particle sizes / concentrations at a unit temperature. The galvanizing solution concentration ratio includes an analysis of the degree of harm caused by the oxidation mixture distribution at a unit temperature. The range of motion and the concentration ratio of the zinc plating solution were both obtained experimentally. The distribution information of the oxide mixture under different acid pickling concentrations recorded the type of potential hazard and the time of occurrence of each oxide mixture distribution occurrence within the risk node. Then, the internal and surface defect characteristics of the production components corresponding to the risk node were input into the oxide mixture distribution impact assessment model to obtain the oxide mixture distribution matrix per unit time corresponding to the risk node. Finally, the impact degree of the oxide mixture distribution per unit temperature was analyzed based on the data introduced by impurities in the human intervention process, the distribution information of the oxide mixture under different acid pickling concentrations, the zinc plating temperature fluctuation range, and the concentration ratio of the zinc plating solution. It can be understood that by analyzing the impact degree of the oxide mixture distribution per unit temperature based on the data introduced by impurities in the human intervention process, the distribution information of the oxide mixture under different acid pickling concentrations, the zinc plating temperature fluctuation range, and the concentration ratio of the zinc plating solution, a multi-dimensional impact degree analysis of the oxide mixture distribution was achieved, which helps improve the accuracy of the impact degree analysis results.

[0071] Based on the impact analysis results of the distribution of the oxide mixture under the aforementioned impact assessment at a unit temperature, an impurity source tracing analysis is performed on the distribution of the oxide mixture under the aforementioned impact assessment at a unit temperature to obtain the galvanizing compliance rate of the oxide mixture distribution within the galvanizing production process. Specifically, the impact assessment distributions of the oxide mixtures corresponding to all risk nodes of oxide mixture intrusion are sorted according to the degree of impact analysis, with the oxide mixture distributions with higher scores ranked first and those with lower scores ranked last. This results in a list indicating to staff which potential hazards within the galvanizing production process are the most urgent and require priority handling. This list represents the final galvanizing compliance rate of the oxide mixture distribution.

[0072] By analyzing the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and smoothness of components, and the location of coating peeling, a comprehensive range of defect characteristics can be obtained for potential oxide mixture intrusion at risk points. These different types of defect characteristics complement each other, providing multi-dimensional data from macro to micro and from surface to interior, which helps to comprehensively identify and detect potential oxide mixture distributions. Furthermore, the method for analyzing the impact of oxide mixture distribution at unit temperature integrates data from impurities introduced during human intervention, oxide mixture distribution information at different acid concentrations, zinc plating temperature fluctuation range, and zinc plating solution concentration ratio, enabling a more accurate risk assessment of each oxide mixture distribution. This multi-dimensional impact analysis method improves the accuracy of impact analysis, helps to more scientifically determine the priority of treatment for oxide mixture distributions, and avoids potential misjudgments from traditional single-dimensional impact analysis.

[0073] Specifically, the step of inputting the internal and surface defect characteristics of the production component corresponding to the risk node of the oxidation mixture intrusion into the oxidation mixture distribution impact assessment model to obtain the unit-time oxidation mixture distribution matrix corresponding to the risk node of the oxidation mixture intrusion includes:

[0074] The internal and surface defect characteristics of the production components are input into the oxidation mixture distribution influence assessment model; the oxidation mixture distribution influence assessment model includes defect type identification coefficient, porosity / bubble distribution and size correlation matrix, crack length and depth correlation matrix, component surface color difference and flatness, coating peeling location correlation matrix, weight allocation mechanism and multi-head variational map autoencoder;

[0075] The defect type identification coefficients are used to identify the type of defects in the internal and surface features of the production components at a unit temperature, respectively, to obtain the type identification result of the defects at a unit temperature, and the defects at a unit temperature are input into the corresponding correlation matrix based on the type identification result of the defects at a unit temperature.

[0076] The distribution and size correlation matrix of pores / bubbles is used to adapt the matrix elements of the received distribution and size of pores / bubbles to obtain the defect matrix coefficients.

[0077] The crack length and depth correlation matrix is ​​used to adapt the matrix elements of the received crack length and depth to obtain the compliance rate matrix coefficients.

[0078] The correlation matrix between the surface color difference and flatness, and the location of coating peeling of the component is used to adapt the matrix elements of the received surface color difference, flatness, and coating peeling location of the component to obtain the surface and coating matrix coefficients.

[0079] The weight allocation mechanism applies influence weights to the defect matrix coefficients, the compliance rate matrix coefficients, and the surface and coating matrix coefficients to obtain a weight adaptation matrix.

[0080] The multi-head variational graph autoencoder uses a preset denoising rule to denoise the weight adaptation matrix, obtains the denoising result corresponding to the weight adaptation matrix, and then uses a preset decoder to reconstruct the matrix of the denoised result to obtain the distribution matrix of the oxidized mixture per unit time.

[0081] This method, by setting a defect type identification coefficient, enables the oxidation mixture distribution impact assessment model to identify and classify different types of defect features. This improves the accuracy of defect processing, allowing each defect feature to be included in its corresponding correlation matrix, ensuring the effectiveness and accuracy of matrix element adaptation. The correlation matrix for each defect type is specifically designed for that type, thereby maximizing the extraction of feature information for that defect type and improving the impact assessment accuracy of the oxidation mixture distribution impact assessment model.

[0082] Specifically, the data on impurities introduced during the manual intervention process includes the dust particle change curve per unit time. The analysis of the impact of the data on the distribution of the oxide mixture at different acid concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio on the distribution of the oxide mixture at a unit temperature includes:

[0083] For the dust particle change curve at a unit temperature, the dust particle size / concentration corresponding to the dust particle change curve is determined using a preset dust particle size / concentration classification rule;

[0084] For the distribution of the oxidized mixture under the influence assessment at a unit temperature, the degree of influence of defect characteristics corresponding to the distribution of the oxidized mixture under the influence assessment is determined based on the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature and the galvanizing temperature fluctuation range. Specifically, the galvanizing temperature fluctuation range includes the temperature anomaly intervals corresponding to different dust particle sizes / concentrations of the oxidized mixture distribution at a unit temperature. When determining the degree of influence of defect characteristics corresponding to the distribution of the oxidized mixture under the influence assessment is based on the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature and the galvanizing temperature fluctuation range, firstly, the pollution parameters corresponding to the dust particle size / concentration of the oxidized mixture distribution under the influence assessment is determined at the unit temperature within the galvanizing temperature fluctuation range. Then, the pollution parameters at a unit temperature are clustered to obtain the result of the degree of influence analysis of defect characteristics.

[0085] For the distribution of the oxidized mixture under the influence of the zinc plating solution at a unit temperature, the influence degree analysis of the factors affecting the non-compliance rate of the oxidized mixture distribution is determined according to the concentration ratio of the zinc plating solution; specifically, the influence degree analysis results of the distribution of the oxidized mixture under the influence of the zinc plating solution concentration ratio are used as the influence degree analysis of the factors affecting the non-compliance rate.

[0086] For the distribution of the oxidized mixture under unit temperature, the influence of the oxidized mixture distribution under different acid pickling concentrations on the coating adhesion is analyzed. Specifically, the distribution of the oxidized mixture under different acid pickling concentrations is used to determine whether the oxidized mixture distribution is present. If it is not present, the factors affecting the coating adhesion are determined. If it is present, the pattern of the oxidized mixture distribution is determined based on the distribution of the oxidized mixture under different acid pickling concentrations, and the pattern is verified using a graph neural network model to obtain the analysis results of the influence on the coating adhesion. When determining the pattern of the oxidized mixture distribution under different acid pickling concentrations, a target historical oxidized mixture distribution that is the same as the oxidized mixture distribution is identified in the oxidized mixture distribution information under different acid pickling concentrations. The number of times the target historical oxidized mixture distribution occurs is determined in the oxidized mixture distribution information under different acid pickling concentrations. The ratio between the number of times the target historical oxidized mixture distribution occurs and the total number of times the oxidized mixture distribution occurs in the oxidized mixture distribution information under different acid pickling concentrations is used as the pattern.

[0087] For the distribution of the oxide mixture under the influence assessment at a unit temperature, the influence degree analysis of defect characteristics, the influence degree analysis of non-compliance factors, and the influence degree analysis of coating adhesion corresponding to the distribution of the oxide mixture under the influence assessment are weighted using preset weight setting rules to obtain the influence degree analysis results corresponding to the distribution of the oxide mixture under the influence assessment.

[0088] This method enables multi-dimensional analysis of the impact of the oxidized mixture distribution on the impact assessment, which helps to improve the accuracy of the impact analysis results.

[0089] Specifically, the method further includes:

[0090] For the risk node of the oxidized mixture intrusion at a unit temperature, a two-way classification variable chi-square test is performed on the risk node of the oxidized mixture intrusion and the internal and surface defect characteristics of the production component corresponding to the risk node of the oxidized mixture intrusion to obtain the two-way classification variable chi-square test result of the defect;

[0091] The chi-square test results of the two-way classification variables of the defects at unit temperature are subjected to a two-way classification variable chi-square test with the zinc plating compliance rate of the oxide mixture distribution to obtain the compliance rate two-way classification variable chi-square test results, and the compliance rate two-way classification variable chi-square test results are stored in the production information management center of the edge computing gateway.

[0092] The step of performing a two-way chi-square test on the risk nodes of the oxidation mixture intrusion and the internal and surface defect characteristics of the corresponding production components to obtain the two-way chi-square test results for the defect classification variables includes:

[0093] Obtain the production condition matrix corresponding to the nodes at risk of oxidized mixture intrusion;

[0094] The influence factors of internal and surface defects of the production components are calculated using a preset defect feature identification method to obtain the defect identification hazard coefficient corresponding to the defect feature;

[0095] The production condition matrix is ​​convolved with the defect identification hazard coefficient to obtain the chi-square test result of the two-way classification variable of the defect.

[0096] The step of performing a two-way chi-square test on the defect bidirectional classification variable at unit temperature and the zinc plating compliance rate of the oxide mixture distribution to obtain the compliance rate two-way classification variable chi-square test result includes:

[0097] Using a pre-defined visualization chart, the chi-square test results of the two-way categorical variable of defects per unit temperature are presented as a curve to obtain the defect fluctuation curve;

[0098] The zinc plating compliance rate of the oxide mixture was plotted using a preset curve identification method to obtain a compliance rate fluctuation curve; the defect fluctuation curve and the compliance rate fluctuation curve did not show the same interfering factors.

[0099] Analyze whether the defect fluctuation curve and the compliance rate fluctuation curve are similar curves;

[0100] When the analysis results are not identical curves, the non-human factor error sources of the defect fluctuation curve and the non-human factor error sources of the compliance rate fluctuation curve are analyzed and ranked according to the degree of influence. The non-human factor error sources whose influence exceeds the preset value are identified as calibrated non-human factor error sources. The human factor error sources of the defect fluctuation curve and the human factor error sources of the compliance rate fluctuation curve are analyzed and ranked according to the degree of influence. The human factor error sources whose influence exceeds the preset value are identified as calibrated human factor error sources.

[0101] Construct a defect anomaly curve and a compliance rate anomaly curve; the names of the horizontal and vertical axes of the defect anomaly curve and the compliance rate anomaly curve are consistent; the non-human factor error sources of the defect anomaly curve are greater than the calibration non-human factor error sources; the human factor error sources of the defect anomaly curve are greater than the calibration human factor error sources; the interference factors of the defect anomaly curve are the same per unit temperature; and the interference factors of the defect anomaly curve do not appear in the defect fluctuation curve and the compliance rate fluctuation curve.

[0102] The defect identification curve is obtained by replacing the interference factor of the defect abnormality curve in the same production process with the interference factor of the unit temperature of the defect fluctuation curve, and the compliance rate identification curve is obtained by replacing the interference factor of the compliance rate abnormality curve in the same production process with the interference factor of the unit temperature of the compliance rate fluctuation curve.

[0103] By comparing the maximum / minimum values ​​of the defect identification curve and the compliance rate identification curve, the chi-square test results of the two-way categorical variable of compliance rate are obtained.

[0104] This method uses a two-way chi-square test to examine the internal and surface defect characteristics of the oxidation mixture intrusion into the risk node and its corresponding production component, obtaining the two-way chi-square test results for defects. Furthermore, it performs a two-way chi-square test between the two-way chi-square test results for defects at unit temperature and the zinc plating compliance rate of the oxidation mixture distribution, thus realizing a systematic correlation between the data. This allows relevant information to be accurately and systematically stored and managed in the production information management center of the edge computing gateway. This systematic data management approach improves the efficiency of data retrieval and provides a solid data foundation for subsequent analysis and decision-making.

[0105] like Figure 2 As shown, an information-based monitoring and management system for a galvanizing production process includes:

[0106] The defect acquisition unit is used to acquire the internal and surface defect characteristics of the production components corresponding to the risk nodes of oxidation mixture intrusion at a unit temperature in the galvanizing production process through the SCADA+SKF multi-parameter visualization monitoring platform preset in the galvanizing production process; the internal and surface defect characteristics of the production components include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the components, and the location of coating peeling.

[0107] The impurity data acquisition unit is used to acquire data on impurities introduced during the cleaning process in the galvanizing production process under different production humidity conditions, specifically during manual intervention.

[0108] The impact analysis unit is used to obtain the impact assessment model of the oxide mixture distribution corresponding to the oxide mixture intrusion risk node at the edge computing gateway production information management center, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio for the oxide mixture intrusion risk node at a unit temperature. The internal and surface defect characteristics of the production components corresponding to the oxide mixture intrusion risk node are input into the oxide mixture distribution impact assessment model to obtain the oxide mixture distribution matrix at the unit time corresponding to the oxide mixture intrusion risk node. The unit also performs an impact analysis on the distribution of the oxide mixture at the unit temperature based on the data brought in by impurities from the human intervention process, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio.

[0109] The two-way categorical variable chi-square test unit is used to perform impurity source tracing analysis on the distribution of the oxidized mixture under the influence assessment at a unit temperature based on the influence degree analysis results, and to obtain the galvanization compliance rate of the oxidized mixture distribution in the galvanizing production process.

[0110] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system described above can be referred to the corresponding process in the aforementioned embodiment of an information-based monitoring and management method for galvanizing production processes, and will not be repeated here.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for information-based monitoring and management of a galvanizing production process, characterized in that, The method includes: Step S1: Obtain the internal and surface defect characteristics of the production components corresponding to the risk nodes of oxidation mixture intrusion at a unit temperature in the galvanizing production process through the SCADA+SKF multi-parameter visualization monitoring platform preset in the galvanizing production process; the internal and surface defect characteristics of the production components include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the components, and the location of coating peeling. Step S2: Obtain data on impurities introduced during manual intervention in the cleaning process of the galvanizing production process under different production humidity. For the risk nodes of the intrusion of the oxide mixture at a unit temperature, obtain the oxide mixture distribution impact assessment model, oxide mixture distribution information under different acid washing concentrations, galvanizing temperature fluctuation range and galvanizing solution concentration ratio corresponding to the risk nodes of the intrusion of the oxide mixture in the edge computing gateway production information management center. Step S3: Input the internal and surface defect characteristics of the production component corresponding to the risk node of the oxidation mixture into the oxidation mixture distribution impact assessment model to obtain the unit time oxidation mixture distribution matrix corresponding to the risk node of the oxidation mixture intrusion. Step S4: Analyze the impact of the data on impurities introduced during the manual intervention process, the distribution information of the oxidized mixture under different acid washing concentrations, the fluctuation range of the galvanizing temperature, and the concentration ratio of the galvanizing solution on the distribution of the oxidized mixture at a unit temperature. Step S5: Based on the influence degree analysis results corresponding to the distribution of the oxide mixture under the influence assessment at a unit temperature, perform impurity source tracing analysis on the distribution of the oxide mixture under the influence assessment at a unit temperature to obtain the galvanizing compliance rate of the oxide mixture distribution in the galvanizing production process. Step S3 includes: The internal and surface defect characteristics of the production components are input into the oxidation mixture distribution influence assessment model; the oxidation mixture distribution influence assessment model includes defect type identification coefficient, porosity / bubble distribution and size correlation matrix, crack length and depth correlation matrix, component surface color difference and flatness, coating peeling location correlation matrix, weight allocation mechanism and multi-head variational map autoencoder; The defect type identification coefficients are used to identify the type of defects in the internal and surface features of the production components at a unit temperature, respectively, to obtain the type identification result of the defects at a unit temperature, and the defects at a unit temperature are input into the corresponding correlation matrix based on the type identification result of the defects at a unit temperature. The distribution and size correlation matrix of pores / bubbles is used to adapt the matrix elements of the received distribution and size of pores / bubbles to obtain the defect matrix coefficients. The crack length and depth correlation matrix is ​​used to adapt the matrix elements of the received crack length and depth to obtain the compliance rate matrix coefficients. The correlation matrix between the surface color difference and flatness of the component and the location of coating peeling is used to adapt the matrix elements of the received surface color difference and flatness and the location of coating peeling to obtain the surface and coating matrix coefficients. The weight allocation mechanism applies influence weights to the defect matrix coefficients, the compliance rate matrix coefficients, and the surface and coating matrix coefficients to obtain a weight adaptation matrix. The multi-head variational graph autoencoder uses a preset denoising rule to denoise the weight adaptation matrix, obtains the denoising result corresponding to the weight adaptation matrix, and then uses a preset decoder to reconstruct the matrix of the denoised result to obtain the distribution matrix of the oxidized mixture per unit time.

2. The method for information-based monitoring and management of a galvanizing production process according to claim 1, characterized in that, The method also includes: Step S6: For the risk nodes of the oxidation mixture intrusion at a unit temperature, perform a two-way classification variable chi-square test on the risk nodes of the oxidation mixture intrusion and the internal and surface defect characteristics of the production components corresponding to the risk nodes of the oxidation mixture intrusion, and obtain the two-way classification variable chi-square test results of the defects. Step S7: Perform a two-way classification variable chi-square test on the defect bidirectional classification variable chi-square test result at unit temperature and the zinc plating compliance rate of the oxide mixture distribution to obtain the compliance rate two-way classification variable chi-square test result, and store the compliance rate two-way classification variable chi-square test result in the edge computing gateway production information management center.

3. The method for information-based monitoring and management of a galvanizing production process according to claim 2, characterized in that, Step S6 includes: Obtain the production condition matrix corresponding to the nodes at risk of oxidized mixture intrusion; The influence factors of internal and surface defects of the production components are calculated using a preset defect feature identification method to obtain the defect identification hazard coefficient corresponding to the defect feature; The production condition matrix is ​​convolved with the defect identification hazard coefficient to obtain the chi-square test result of the two-way classification variable of the defect.

4. The method for information-based monitoring and management of a galvanizing production process according to claim 2, characterized in that, Step S7 includes: Using a pre-defined visualization chart, the chi-square test results of the two-way categorical variable of defects per unit temperature are presented as a curve to obtain the defect fluctuation curve; The zinc plating compliance rate of the oxide mixture was plotted using a preset curve identification method to obtain a compliance rate fluctuation curve; the defect fluctuation curve and the compliance rate fluctuation curve did not show the same interfering factors. Analyze whether the defect fluctuation curve and the compliance rate fluctuation curve are similar curves; When the analysis results are not identical curves, the non-human factor error sources of the defect fluctuation curve and the non-human factor error sources of the compliance rate fluctuation curve are analyzed and ranked according to the degree of influence. The non-human factor error sources whose influence exceeds the preset value are identified as calibrated non-human factor error sources. The human factor error sources of the defect fluctuation curve and the human factor error sources of the compliance rate fluctuation curve are analyzed and ranked according to the degree of influence. The human factor error sources whose influence exceeds the preset value are identified as calibrated human factor error sources. Construct a defect anomaly curve and a compliance rate anomaly curve; the names of the horizontal and vertical axes of the defect anomaly curve and the compliance rate anomaly curve are consistent; the non-human factor error sources of the defect anomaly curve are greater than the calibration non-human factor error sources; the human factor error sources of the defect anomaly curve are greater than the calibration human factor error sources; the interference factors of the defect anomaly curve are the same per unit temperature; and the interference factors of the defect anomaly curve do not appear in the defect fluctuation curve and the compliance rate fluctuation curve. The defect identification curve is obtained by replacing the interference factor of the defect abnormality curve in the same production process with the interference factor of the unit temperature of the defect fluctuation curve, and the compliance rate identification curve is obtained by replacing the interference factor of the compliance rate abnormality curve in the same production process with the interference factor of the unit temperature of the compliance rate fluctuation curve. By comparing the maximum / minimum values ​​of the defect identification curve and the compliance rate identification curve, the chi-square test results of the two-way categorical variable of compliance rate are obtained.

5. The method for information-based monitoring and management of a galvanizing production process according to claim 1, characterized in that, The data on impurities introduced during the manual intervention process includes the dust particle change curve per unit time. The analysis of the impact of the data on impurities introduced during the manual intervention process, the distribution information of the oxide mixture under different acid washing concentrations, the fluctuation range of the galvanizing temperature, and the concentration ratio of the galvanizing solution on the distribution of the oxide mixture per unit temperature includes: For the dust particle change curve at a unit temperature, the dust particle size / concentration corresponding to the dust particle change curve is determined using a preset dust particle size / concentration classification rule; For the distribution of the oxide mixture under the influence assessment at a unit temperature, the degree of influence of the defect characteristics corresponding to the distribution of the oxide mixture under the influence assessment is determined based on the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature and the galvanizing temperature fluctuation range. For the distribution of the oxidized mixture under the influence of the unit temperature, the degree of influence of the factors affecting the non-compliance rate of the oxidized mixture distribution is determined according to the concentration ratio of the zinc plating solution. For the distribution of the oxidized mixture under unit temperature, the influence of the oxidized mixture distribution under different acid concentrations is analyzed to determine the degree of influence on coating adhesion corresponding to the distribution of the oxidized mixture. For the distribution of the oxide mixture under the influence assessment at a unit temperature, the influence degree analysis of defect characteristics, the influence degree analysis of non-compliance factors, and the influence degree analysis of coating adhesion corresponding to the distribution of the oxide mixture under the influence assessment are weighted using preset weight setting rules to obtain the influence degree analysis results corresponding to the distribution of the oxide mixture under the influence assessment.

6. The method for information-based monitoring and management of a galvanizing production process according to claim 5, characterized in that, The galvanizing temperature fluctuation range includes the temperature anomaly range corresponding to different dust particle sizes / concentrations of the oxide mixture at a unit temperature. The method for determining the degree of influence of the defect characteristics corresponding to the distribution of the oxide mixture under the influence assessment based on the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature and the galvanizing temperature fluctuation range includes: determining the pollution parameters corresponding to the dust particle size / concentration corresponding to the dust particle change curve at a unit temperature within the galvanizing temperature fluctuation range; and clustering the pollution parameters at a unit temperature to obtain the result of the degree of influence analysis of the defect characteristics.

7. The method for information-based monitoring and management of a galvanizing production process according to claim 5, characterized in that, The step of determining the degree of influence of the distribution of the oxide mixture under different acid concentrations on the coating adhesion based on the distribution information of the oxide mixture includes: Based on the distribution information of the oxide mixture under different acid concentrations, determine whether the distribution of the oxide mixture affecting the coating adhesion occurs; if it does not occur, determine the factors affecting the coating adhesion; if it occurs, determine the pattern of the distribution of the oxide mixture affecting the coating adhesion based on the distribution information of the oxide mixture under different acid concentrations, and verify the pattern using a graph neural network model to obtain the analysis results of the degree of influence on the coating adhesion.

8. An information-based monitoring and management system for galvanizing production processes, characterized in that, This system is applied to the information-based monitoring and management method for a galvanizing production process as described in claim 1, comprising: The defect acquisition unit is used to acquire the internal and surface defect characteristics of the production components corresponding to the risk nodes of oxidation mixture intrusion at a unit temperature in the galvanizing production process through the SCADA+SKF multi-parameter visualization monitoring platform preset in the galvanizing production process; the internal and surface defect characteristics of the production components include the distribution and size of pores / bubbles, the length and depth of cracks, the surface color difference and flatness of the components, and the location of coating peeling. The impurity data acquisition unit is used to acquire data on impurities introduced during the cleaning process involving human intervention in the galvanizing production process under different production humidity conditions. The impact analysis unit is used to obtain the impact assessment model of the oxide mixture distribution corresponding to the oxide mixture intrusion risk node at a unit temperature, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio from the production information management center of the edge computing gateway. It also inputs the internal and surface defect characteristics of the production components corresponding to the oxide mixture intrusion risk node into the oxide mixture distribution impact assessment model to obtain the oxide mixture distribution matrix per unit time corresponding to the oxide mixture intrusion risk node. Furthermore, it performs an impact analysis on the distribution of the oxide mixture at a unit temperature based on the data brought in by impurities from the human intervention process, the oxide mixture distribution information under different acid washing concentrations, the galvanizing temperature fluctuation range, and the galvanizing solution concentration ratio. The bidirectional categorical variable chi-square test unit is used to perform impurity source tracing analysis on the distribution of the oxide mixture under the influence assessment at a unit temperature based on the influence degree analysis results corresponding to the distribution of the oxide mixture under the influence assessment at a unit temperature, to obtain the galvanizing compliance rate of the oxide mixture distribution in the galvanizing production process, and to perform a bidirectional categorical variable chi-square test on the oxide mixture intrusion risk node at a unit temperature and the internal and surface defect characteristics of the production component corresponding to the oxide mixture intrusion risk node, to obtain the defect bidirectional categorical variable chi-square test result, and to perform a bidirectional categorical variable chi-square test on the defect bidirectional categorical variable chi-square test result at a unit temperature and the galvanizing compliance rate of the oxide mixture distribution, to obtain the compliance rate bidirectional categorical variable chi-square test result, and store the compliance rate bidirectional categorical variable chi-square test result in the edge computing gateway production information management center.

Images