A digital visual control method for a food production line control system

Abstract

This invention relates to the field of production line control technology, and more particularly to a digital visualization control method for a food production line control system. The method includes the following steps: acquiring food production line process data; extracting the filling process, sterilization process, and impurity detection process from the food production line process data, and locating the spatial coordinates of the corresponding process operation equipment to obtain the spatial coordinates of the filling machine, sterilization vessel, and metal detector; analyzing the equipment topology of the food production line process data based on the spatial coordinates of the filling machine, sterilization vessel, and metal detector to construct a production line equipment topology diagram. This invention, through equipment spatial positioning and topology construction, achieves dynamic data integration across multiple processes and automatic anomaly response, thereby enhancing the real-time digital visualization intelligent control capability of the food production line.

Description

Technical Field

[0001] This invention relates to the field of production line control technology, and in particular to a digital visualization control method for a food production line control system. Background Technology

[0002] With the development of computer technology and network communication, food production line control systems have gradually integrated digital technologies, enabling real-time acquisition, monitoring, and analysis of production data. The visual interface of SCADA (Supervisory Control and Data Acquisition) systems allows operators to intuitively understand the production status and quickly respond to anomalies, greatly improving management efficiency and safety. Simultaneously, the integration of MES (Manufacturing Execution System) and ERP (Enterprise Resource Planning) systems promotes collaboration between production planning and execution, facilitating information sharing and process optimization.

[0003] However, existing food production line control systems have specific deficiencies in the three key processes of filling, sterilization, and impurity detection: the filling process relies heavily on single pressure monitoring, neglecting the spatial location of the filling machine and its coordination with other equipment, resulting in difficulty in quickly locating pressure anomalies and delayed response; the sterilization process lacks dynamic monitoring of spatial temperature distribution, making it difficult to detect local temperature differences and affecting sterilization effectiveness; the impurity detection process is not sufficiently linked with the filling and sterilization processes, resulting in inaccurate impurity location and delays or misjudgments in the rejection action; in addition, the system as a whole lacks comprehensive real-time updates and visualization of the status of multiple processes and multiple devices, affecting the efficiency of management decision-making. Summary of the Invention

[0004] Therefore, it is necessary to provide a digital visualization control method for food production line control systems to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a digital visualization control method for a food production line control system is provided, the method comprising the following steps:

[0006] Step S1: Obtain food production line process data; extract the filling process, sterilization process and impurity detection process from the food production line process data, and locate the spatial coordinates of the corresponding process operation equipment to obtain the spatial coordinates of the filling machine, sterilization kettle and metal detector.

[0007] Step S2: Analyze the equipment topology of the food production line process data based on the spatial coordinates of the filling machine, the sterilization vessel, and the metal detector to construct a production line equipment topology map; use the production line equipment topology map to dynamically overlay process data of the filling process, sterilization process, and impurity detection process to generate the key production process interface.

[0008] Step S3: Determine whether the filling pressure deviation and abnormal sterilization temperature gradient at the interface of the key production process exceed the preset threshold. If so, generate a filling machine pressure stabilization command or a sterilizer steam adjustment command, and drive the sorting valve to remove metal contaminant materials.

[0009] Step S4: Update the pressure gradient, temperature distribution and sorting status in the production line equipment topology diagram in real time according to the pressure stabilization command of the filling machine or the steam adjustment command of the sterilizer, so as to execute the digital visualization control operation of the food production line control system.

[0010] This invention acquires and integrates the spatial coordinates of equipment in multiple processes, including filling, sterilization, and impurity detection, to construct an equipment topology map. This enables data linkage and dynamic overlay between different processes, effectively improving the response speed and handling accuracy for anomalies in key processes (such as filling pressure fluctuations and sterilization temperature anomalies). By dynamically overlaying process data onto the production line topology map, a key process flow interface is generated, enabling real-time visual monitoring of the food production process status and enhancing operators' awareness of production status and risk changes. The introduction of intelligent control mechanisms such as "filling machine pressure stabilization commands" and "sterilizer steam adjustment commands" proactively generates control commands based on detection results and links sorting valves to reject abnormal materials, thereby preventing unqualified products from flowing into subsequent processes and ensuring product safety and production compliance. Real-time updates of pressure gradients, temperature distributions, and sorting status in the production line equipment topology map construct a digital mapping reflecting actual production conditions, laying a data foundation for subsequent intelligent optimization decisions, predictive maintenance, and equipment collaboration. Through automatic judgment and anomaly intervention of key process data, the delay and error of manual intervention are reduced, improving the stability of production line operation and product quality consistency. Therefore, this invention achieves dynamic data integration and automatic anomaly response across multiple processes through equipment spatial positioning and topology construction, thereby enhancing the real-time digital visualization and intelligent control capabilities of food production lines.

[0011] Preferably, step S1 includes the following steps:

[0012] Step S11: Obtain food production line process data; perform process identification and analysis on the food production line process data to extract the filling process, sterilization process and impurity detection process, and generate key food process data;

[0013] Step S12: Match the equipment correspondences of the key food processing data, extract the equipment information corresponding to the filling process, and generate filling machine equipment identification data; based on the filling machine equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the filling machine equipment in the production line, and generate filling machine spatial coordinate data.

[0014] Step S13: Match the equipment correspondences of the key food processing data, extract the equipment information corresponding to the sterilization process, and generate sterilization autoclave equipment identification data; based on the sterilization autoclave equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the sterilization autoclave equipment in the production line, and generate sterilization autoclave spatial coordinate data;

[0015] Step S14: Match the equipment correspondence of the key food process data, extract the equipment information corresponding to the impurity detection process, and generate metal detector equipment identification data; based on the metal detector equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the metal detector equipment in the production line, and generate metal detector spatial coordinate data.

[0016] This invention, through process identification and analysis and matching with equipment correspondence, can accurately map abstract food production process data to specific equipment entities, ensuring that each key process (such as filling, sterilization, and impurity detection) can be accurately associated with its executing equipment, significantly improving the accuracy and completeness of process identification. Each key process generates unique equipment identification data for its involved equipment, establishing a stable equipment management index system. This facilitates unified management and precise traceability for subsequent maintenance tracking, process optimization, and equipment status monitoring. Based on the equipment identification, spatial configuration fields are parsed to extract the spatial location of each piece of equipment in the production line, constructing highly accurate "filling machine spatial coordinate data," "sterilization vessel spatial coordinate data," and "metal detector spatial coordinate data," providing fundamental data support for subsequent equipment topology modeling, spatial scheduling simulation, and digital twin system construction. By managing key processes and spatial coordinate data in a structured manner, automatic linkage between production line process logic and equipment physical layout is achieved. This facilitates dynamic updates of the equipment topology diagram in situations such as production line structure adjustments and equipment replacement, improving the system's adaptability and maintainability. The detailed steps S11 to S14 provide a standardized data foundation for subsequent intelligent functions such as anomaly detection, digital visualization, and process optimization, enhancing the data-driven capability and intelligent evolution capability of the entire production control system.

[0017] Preferably, step S11 includes the following steps:

[0018] Step S111: Collect process data of the food production line, including PLC process control data, equipment status signals and timestamp sequence information. The data sampling period is set to 100ms to 500ms, and the collection time period is no less than 1.5 times the complete production batch.

[0019] Step S112: Perform structured analysis on the food production line process data and extract key equipment status tags, including the start / stop status of the filling head, the status of the sterilization temperature control valve, and the on / off status of the metal detector; among them, the filling process is determined when the filling head is continuously open for more than 30 seconds and the bottle position sensor is in a continuously changing state.

[0020] Step S113: The rule for determining the "sterilization process" is: the temperature sensor reading is stable in the range of 80°C to 125°C for more than 60 seconds, and the main pump of the corresponding sterilization device is in the "on" state; if there is a sudden temperature change or the temperature is below 70°C for more than 20 seconds, the process segment is interrupted.

[0021] Step S114: The rule for determining the impurity detection process is: the metal detector is in the detection state and the detection signal trigger frequency is greater than 0.2 times / second, or the visual inspection device continuously identifies abnormal image feature areas; any of the above conditions can be marked as an impurity detection process.

[0022] Step S115: Archive and integrate the segment numbers, start and end times, corresponding process labels and key equipment identification codes of the identified filling process, sterilization process and impurity detection process to generate food key process data.

[0023] This invention ensures sufficient time coverage and accuracy in the acquired data by setting a data sampling period (100ms to 500ms) and a collection time period (no less than 1.5 times the complete batch), providing stable and reliable data support for process identification and judgment, and avoiding misjudgments caused by omissions of key states or sampling delays. It employs multi-dimensional equipment status signals (such as temperature control valve status, sensor temperature values, detector on / off status, etc.) and timing information to establish quantifiable judgment logic (e.g., "temperature 80°C to 125°C for more than 60 seconds" is a sterilization process), making the identification standards for filling, sterilization, and impurity detection processes clear and reproducible, enhancing the system's stability and standardization capabilities. By combining PLC data with equipment status signals and leveraging multiple sources such as continuous changes in bottle position sensors, the operating status of temperature control devices, and the trigger frequency of metal detectors to collaboratively determine process status, it effectively reduces the misidentification rate caused by single signal anomalies and improves the accuracy of key process extraction. Interruption rules, such as "interrupting the sterilization process due to sudden temperature changes or a temperature drop below 70°C for 20 seconds," are set to intelligently identify and exclude non-standard or abnormal process segments. This ensures that archived data represents authentic and valid production processes, avoiding interference with subsequent process analysis and control strategy development. By archiving identified key processes according to segment number, start and end time, process label, and equipment code, structured "critical food process data" is constructed, providing highly consistent data resources for subsequent process topology modeling, production visualization, equipment linkage control, and anomaly traceability.

[0024] Preferably, the equipment topology in step S2, which analyzes the food production line process data based on the spatial coordinates of the filling machine, the sterilization vessel, and the metal detector, includes:

[0025] Calculate the spatial path distance between the spatial coordinate data of the filling machine and the spatial coordinate data of the sterilization vessel, and generate relative path data between the filling and sterilization equipment;

[0026] Calculate the spatial path distance between the spatial coordinates of the sterilization autoclave and the spatial coordinates of the metal detector, and generate relative path data between the sterilization and detection equipment.

[0027] The relative path data of filling-sterilization equipment and relative path data of sterilization-detection equipment are fused and analyzed to generate equipment logical connection relationship data.

[0028] Generate graph structure nodes from the logical connection data of the equipment, construct the topology graph node identifier for each equipment, and generate production line equipment node set data;

[0029] Assign edge relationships to the production line equipment node set data, construct a complete topological edge set, and generate production line equipment edge set data; assemble the production line equipment node set data and production line equipment edge set data into a graph structure to generate a production line equipment topology graph.

[0030] This invention achieves precise geometric modeling of equipment in physical space by calculating spatial path distances and generating relative path data between filling machines, sterilization tanks, and metal detectors. This provides a foundation for modeling real production line layouts and avoids spatial deviations caused by planar diagrams or manual assumptions. By integrating spatial path data with process sequence, the logical connection relationships between equipment (e.g., "filling → sterilization → detection") are automatically derived, achieving a dual mapping between process logic and equipment spatial distribution. This facilitates production process streamlining and process dependency analysis. The logical connection data of equipment is converted into nodes and edges in a graph structure, forming standard "production line equipment node sets" and "production line equipment edge sets." This facilitates path planning, process refactoring, bottleneck identification, and other operations through graph computing technology, supporting structured management and visualization of complex production lines. The production line equipment topology graph built based on the graph structure model can dynamically update node or edge information in scenarios such as equipment position changes and production line process adjustments. It possesses high flexibility and scalability, effectively supporting modular layout and rapid refactoring in intelligent manufacturing systems. The construction of the topology graph not only serves the visualization of equipment space, but also provides a unified carrying model for the subsequent overlay and status visualization of process nodes for data such as temperature distribution, pressure control, and abnormal response. It supports multi-dimensional data fusion and control path scheduling optimization on the graph structure.

[0031] Preferably, step S2, which involves dynamically overlaying process data for the filling process, sterilization process, and impurity detection process using a production line equipment topology diagram, includes:

[0032] Based on the filling process, sterilization process, and impurity detection process, respectively, the canning pressure data, sterilization temperature data, and metal impurity detection signals are collected.

[0033] Color gradient mapping is performed on filling pressure data to generate filling nozzle color gradient data; structural drift detection is performed on historical filling nozzle color gradient data, and microstructure displacement risks are identified through abnormal fan-shaped offset areas. Deformation fitting verification is performed in combination with the preset filling machine support arm CAD static model to generate structural-level implicit linkage risk map data.

[0034] Calculate the spatial cross-sectional interpolation of the sterilization temperature data to generate thermal map data of the sterilization autoclave cross-section;

[0035] The alarm trigger analysis is performed on the metal impurity detection signal data to control the display status of the sorting valve and generate flashing alarm data for the sorting valve.

[0036] Process location mapping processing is performed on the color gradient data of filling nozzles, thermal map data of sterilizer cross-section, and flashing alarm data of sorting valves to generate process visualization layer data.

[0037] By integrating and overlaying process visualization layer data with production line equipment topology data using structural-level implicit linkage risk map data, a key production process flow interface is generated.

[0038] This invention maps real-time process data such as filling pressure, sterilization temperature, and impurity detection signals to the production line equipment topology diagram, generating visualized layer data. This enables dynamic spatial presentation of process data, allowing operators to intuitively grasp the operational status and risk distribution of each process node. By performing structural drift analysis on historical color gradient data of the filling nozzle and combining it with deformation fitting of the CAD static model of the support arm, "invisible microstructural deformation risks" can be accurately identified, generating a "structural-level implicit linkage risk map." This significantly improves the mechanical micro-damage detection capabilities that are difficult to cover with traditional monitoring methods, aiding in equipment health management and predictive maintenance. The generation of the thermal map of the sterilization autoclave cross-section is based on spatial interpolation of temperature data, achieving accurate modeling of the internal heat distribution. This provides a visualized analytical basis for judging the uniformity of heat sterilization, identifying potential temperature control deviations or dead zones, and improving the stability of sterilization effects and product safety assurance capabilities. Once the metal impurity detection signal is triggered, the sorting valve status can be directly controlled, and visual flashing alarm data can be generated. This integrates alarm information, process location, and response action, significantly reducing operator response time and effectively improving impurity removal efficiency and system linkage capabilities. By merging the "structural-level implicit linkage risk map" with various process layers and equipment topology diagrams, a "critical production process interface" is formed. This not only provides a unified view of "real-time data + spatial topology + risk visualization," but also facilitates rapid traceability and closed-loop handling of structure-process-signal in cases of equipment malfunctions, process fluctuations, etc.

[0039] Preferably, structural drift detection is performed on historical filling nozzle color gradient data, and microstructural displacement risks are identified through abnormal fan-shaped offset regions. Deformation fitting verification is then performed in conjunction with a pre-set CAD static model of the filling machine support arm, including:

[0040] Historical filling nozzle color gradient data is resampled over time to extract the color gradient boundary change sequence within the filling nozzle's working cycle, generating filling nozzle boundary gradient evolution data.

[0041] Multi-frame boundary contour alignment is performed on the gradient evolution data of the filling nozzle boundary to identify the drift direction of the contour trajectory, construct the offset aggregation region in the sector corner domain, and generate abnormal sector offset region data.

[0042] Multidimensional geometric center drift analysis was performed on the abnormal fan-shaped offset region data to extract the joint features of offset amplitude and angle, and a microstructure displacement probability scoring matrix was constructed to generate microstructure displacement risk scoring data.

[0043] The microstructure displacement risk score data is compared with the preset CAD static model of the filling machine support arm by comparing the node constraints. The residuals are fitted according to the mechanical tolerance of the actual deformable nodes to generate deformation fitting residual data.

[0044] Based on the deformation fitting residual data, key component nodes with high strain concentration trends in the CAD static model of the filling machine support arm are identified, and structural-level implicit linkage risk map data is generated.

[0045] This invention extracts the evolution sequence of the color gradient boundary of the filling nozzle over time and performs contour alignment and fan-shaped offset analysis to identify microstructural drift trends that are invisible to the naked eye and difficult to detect with traditional sensors, greatly improving the early detection capability of micro-deformation in filling machines. By constructing an "abnormal fan-shaped offset region" and establishing a risk scoring matrix based on multi-dimensional features such as offset direction, amplitude, and angle, multi-angle, high-resolution detection of microstructural deformation is achieved, avoiding misjudgments and omissions caused by relying solely on single-point monitoring. The scoring results are then fitted with the node-level residuals of the CAD static model of the filling machine's support arm, enhancing the structural correlation of the detection results and effectively identifying the relationship between deformation trends and stress distribution in the support structure, forming a closed-loop identification path from phenomenon to structural root cause. Based on the fitting analysis of deformation residuals, key nodes in the CAD model with potential fatigue, stress concentration, or deformation accumulation risks can be accurately located, providing precise technical basis for equipment maintenance, structural optimization, and service life prediction. The final generated "structural-level implicit linkage risk map" not only identifies the existing potential structural anomalies, but can also be integrated with the process flow layer and topology diagram to achieve integrated closed-loop management of "risk-data-structure-control" for key equipment.

[0046] Preferably, the spatial cross-sectional interpolation for calculating sterilization temperature data includes:

[0047] Three-dimensional sampling point coordinate mapping is performed on the sterilization temperature data to construct a spatial temperature point set of the inner wall and core area of ​​the sterilization autoclave, and sterilization temperature point cloud data is generated.

[0048] Local thermal field density estimation is performed on the sterilization temperature point cloud data to extract the temperature distribution gradient of different cross-sectional regions and generate initial cross-sectional temperature distribution contour data.

[0049] Two-dimensional bilinear interpolation is performed on the initial cross-sectional temperature distribution profile data, and morphological constraint correction is performed based on the geometry of the sterilization vessel to generate the interpolated cross-sectional temperature grid data of the sterilization vessel.

[0050] The interpolated temperature grid data of the sterilizer section is subjected to thermal classification and pseudo-color mapping to generate thermogram data of the sterilizer section.

[0051] This invention constructs a temperature point cloud by mapping temperature sampling points in three-dimensional space, accurately reconstructing the heat distribution of the inner wall and core area of ​​the sterilization vessel. This expands thermal field perception beyond local points, enhancing the spatial insight of the temperature control system. Through local thermal field density estimation and temperature distribution gradient extraction, it identifies temperature difference distribution trends and temperature jumps in different cross-sectional areas, providing structured support for sterilization uniformity analysis, dead zone identification, and process stability assessment. While employing two-dimensional bilinear interpolation, the temperature grid is morphologically corrected based on the actual geometric boundaries of the sterilization vessel, avoiding physically impossible interpolation distortions and ensuring a high degree of consistency between the thermal map results and the actual equipment shape, thus enhancing model credibility. The final output cross-sectional thermal map, through thermal grading and pseudo-color mapping, visually displays key indicators such as temperature levels and distribution uniformity, reducing the complexity of manual identification and improving operators' ability to quickly locate potential problem areas. The cross-sectional thermal map can be used as input feedback to the control system to analyze which areas have insufficient temperature control or overheating trends, and assist in the intelligent adjustment of steam supply, valve control and other links to achieve closed-loop control of the thermal field and adaptive optimization of process parameters.

[0052] Preferably, step S3 includes the following steps:

[0053] Step S31: Perform regional stability window sliding calculation on the filling pressure data in the key production process interface to generate filling pressure deviation data; fit the steady-state drift trend of the filling pressure deviation data to identify whether there is local non-equilibrium continuous fluctuation, and generate filling pressure anomaly identification result data.

[0054] Step S32: Perform isothermal zone morphological connectivity analysis on the thermal map data of the sterilization autoclave section, extract the thermal gradient fault zone area, and generate sterilization temperature gradient anomaly layer data.

[0055] Step S33: Determine whether the abnormal filling pressure identification result data exceeds the preset threshold. If it does, generate a filling machine pressure stabilization command. Determine whether there is a high-risk area in the abnormal sterilization temperature gradient layer data. If so, generate a sterilization autoclave steam adjustment command.

[0056] Step S34: Confirm the status of the impurity signal triggered by the metal detector through the pressure stabilization command of the filling machine or the steam regulation command of the sterilizer, and generate a sorting valve rejection action command according to its position to perform the metal contaminant rejection operation.

[0057] This invention, through regional sliding window calculation and steady-state drift trend fitting of filling pressure data, not only identifies numerical anomalies but also detects trend anomalies in non-equilibrium fluctuations. This avoids the limitations of traditional monitoring methods that only focus on instantaneous exceedances while ignoring potential hazards, thus enhancing the ability to identify "gradual instability" in the process state. Utilizing the thermographic map of the sterilizer cross-section for isothermal zone connectivity and thermal gradient discontinuity identification, it detects abnormal "discontinuities" and high-temperature dead zones in the thermal field structure, effectively reflecting the uniformity and consistency of the sterilization process in different spatial regions, and enhancing the spatial perception and analysis capabilities of sterilization safety risks. Based on the anomaly identification results, it automatically generates pressure stabilization commands for the filling machine and steam adjustment commands for the sterilizer, constructing a fully automated closed-loop path from data identification to control execution. This improves the system's response efficiency and autonomous adjustment capabilities, reducing the burden of manual judgment and intervention. By integrating anomaly identification and impurity signal status confirmation mechanisms in the filling and sterilization processes, product quality fluctuations caused by unstable filling pressure or sterilization anomalies can be identified. The sorting valves are then proactively controlled to remove related materials, ensuring the quality of downstream finished products and establishing an integrated quality assurance mechanism encompassing process, detection, and control. This step forms a unified control logic chain from filling and sterilization to impurity removal, not only improving the stability, compliance, and traceability of the production line but also achieving intelligent upgrades through a multi-point linkage control mechanism. This effectively supports the digital, closed-loop, and automated evolution of food production lines.

[0058] Preferably, step S34 includes the following steps:

[0059] Step S341: Receive the pressure stabilization command from the filling machine and the steam regulation command from the sterilizer, parse the command parameters, where the pressure stabilization command has a pressure range of 0.5MPa~1.2MPa and the steam regulation command has a temperature range of 100°C~130°C, and ensure that the command execution status is "confirmed".

[0060] Step S342: Monitor the impurity signal triggered by the metal detector in real time. The impurity signal strength threshold is set to >50mV, and the signal duration is not less than 10ms to be considered a valid trigger.

[0061] Step S343: Based on the timestamp and location code of the metal detector trigger signal, and combined with the current pressure stabilization and steam status of the filling machine and sterilizer, determine whether the impurity signal is within the normal production operating range; when both the pressure stabilization command and the steam command are within the fluctuation range of ±5%, confirm that the impurity signal is valid.

[0062] Step S344: Based on the confirmed impurity signal, calculate the trigger time of the sorting valve rejection action. Considering the product transmission speed, the conveyor belt speed range is set to 0.2~1.0m / s, and the rejection response time error does not exceed ±50ms. Generate a synchronous sorting valve rejection action command; send the rejection action command to the sorting valve control unit to ensure that the sorting valve action range covers the spatial position of the material corresponding to the impurity signal, and the opening and closing action time of the sorting valve is controlled within 100ms.

[0063] Step S345: Record the rejection action execution log, including the impurity signal trigger time, the sorting valve action time, and the rejection success flag, to obtain the sorting valve response configuration data for performing the metal contaminant rejection operation.

[0064] This invention, through parameter analysis and status confirmation of the filling machine's pressure stabilization command and the sterilizer's steam adjustment command, ensures that all control commands are in a "confirmed" state and that the command parameters are within a reasonable range before executing the rejection logic. This effectively prevents misjudgments and rejections caused by erroneous commands or abnormal fluctuations, improving the overall reliability of the system. Dual threshold judgments of impurity signal intensity and duration (>50mV and ≥10ms) filter out occasional interference signals, effectively avoiding false triggers and improving the accuracy and robustness of metal detectors in identifying valid impurities on high-speed production lines. By combining the current filling pressure and sterilization temperature, it verifies whether the impurity signal is within the "stable production window" (±5% fluctuation range), enabling cross-stage linkage to determine whether impurity anomalies are genuine contamination events, preventing false alarms caused by process fluctuations, and improving the intelligence and accuracy of rejection decisions. By combining conveyor belt speed and material position coding, the system accurately calculates the rejection trigger timing of the sorting valve, controlling the rejection response error within ±50ms and the sorting action time within 100ms. This achieves precise rejection on high-speed production lines, avoiding missed or incorrect rejections and improving product qualification rates. By recording impurity signal triggers, sorting action times, and rejection results, the system generates sorting valve response configuration data and log information, providing data support for post-event traceability, system optimization, and quality auditing, thereby improving overall quality management.

[0065] Preferably, step S4 includes the following steps:

[0066] Step S41: Adjust the pressure control parameters of the filling machine according to the pressure stabilization command of the filling machine, and generate pressure adjustment execution data; adjust the steam valve opening according to the steam regulation command of the sterilizer, and generate steam flow adjustment data;

[0067] Step S42: Locate the impurity signal data triggered by the metal detector to generate impurity signal location data; combine the impurity signal location data with the sorting valve response configuration data to generate sorting valve rejection action trigger data;

[0068] Step S43: Control the sorting valve to perform the rejection operation based on the rejection action trigger data of the sorting valve, and generate rejection operation execution result data; perform feedback monitoring on the rejection operation execution result data, confirm the rejection effect, and generate rejection status confirmation data;

[0069] Step S44: Update the production line equipment topology diagram in real time based on pressure adjustment execution data, steam flow adjustment data, and rejection status confirmation data to perform digital visualization control operations of the food production line control system.

[0070] This invention converts the pressure stabilization command of the filling machine into specific pressure control parameter adjustments and maps the steam regulation command of the sterilization autoclave into steam valve opening control, ensuring that each control command is executable and supported by specific parameters. This achieves precise adjustment of the filling and sterilization equipment, improving the efficiency and process stability of the control system. By locating the trigger signal of the metal detector and linking it with the sorting valve response configuration data, highly consistent rejection action trigger data is generated, ensuring the timeliness and accuracy of impurity rejection operations, especially suitable for rejection tasks under high-speed conveyor belt conditions. It not only executes the sorting valve rejection action but also monitors and confirms the rejection status, forming a closed-loop control link of "action → result → feedback," ensuring that the effectiveness of each batch of impurity rejection can be verified, thereby guaranteeing the quality of the finished product and the traceability of the rejection operation. The filling pressure adjustment data, steam flow control data, rejection confirmation status, and other operation results are dynamically mapped to the equipment topology diagram, achieving layer-level visual linkage updates. This allows operators to view the operating status of each device and the execution of control commands in real time on a digital interface, greatly improving the visualization control capabilities and operational transparency of the food production line. By unifying data linkage and layer feedback across multiple stages such as elimination, adjustment, and topology changes, a highly collaborative and interconnected digital production line environment is constructed, providing a comprehensive, real-time, and accurate data foundation for subsequent intelligent production scheduling, production line operation, quality analysis, and early warning strategies. Attached Figure Description

[0071] Figure 1 A flowchart illustrating the steps of a digital visualization control method for a food production line control system.

[0072] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1.

[0073] Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S4.

[0074] Figure 4 A real-world flowchart of a digital visualization control method for a food production line control system.

[0075] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0076] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0077] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0078] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0079] To achieve the above objectives, please refer to Figures 1 to 4 A digital visualization control method for a food production line control system, the method comprising the following steps:

[0080] Step S1: Obtain food production line process data; extract the filling process, sterilization process and impurity detection process from the food production line process data, and locate the spatial coordinates of the corresponding process operation equipment to obtain the spatial coordinates of the filling machine, sterilization kettle and metal detector.

[0081] Step S2: Analyze the equipment topology of the food production line process data based on the spatial coordinates of the filling machine, the sterilization vessel, and the metal detector to construct a production line equipment topology map; use the production line equipment topology map to dynamically overlay process data of the filling process, sterilization process, and impurity detection process to generate the key production process interface.

[0082] Step S3: Determine whether the filling pressure deviation and abnormal sterilization temperature gradient at the interface of the key production process exceed the preset threshold. If so, generate a filling machine pressure stabilization command or a sterilizer steam adjustment command, and drive the sorting valve to remove metal contaminant materials.

[0083] Step S4: Update the pressure gradient, temperature distribution and sorting status in the production line equipment topology diagram in real time according to the pressure stabilization command of the filling machine or the steam adjustment command of the sterilizer, so as to execute the digital visualization control operation of the food production line control system.

[0084] In the embodiments of this invention, please refer to Figure 4A sensor network deployed at key nodes of the production line collects real-time process data, including fluid pressure, temperature, equipment operating status, and control commands. The collected process data is input into a data processing module, which, based on a predefined process flow structure, identifies the time series and process stages of the filling, sterilization, and impurity detection processes. High-precision 3D measuring equipment (such as a laser rangefinder or industrial vision system) is used to locate the filling machine 101, sterilizer 102, and metal detector 103 in spatial coordinates, determining the specific 3D coordinates of each device within the production line. These spatial coordinates are converted using a coordinate system and stored in a database, ensuring the accuracy and real-time updating of equipment positions. Based on the spatial coordinates obtained in step S1, a graph theory-based equipment topology diagram is constructed. First, the filling machine, sterilizer, and metal detector are treated as nodes, and the edges between nodes are determined based on physical connections and data flow direction. A complete production line equipment topology diagram is generated using topology modeling software (such as the NetworkX library or a professional process modeling platform). Based on a topology graph, dynamic data fusion technology is used to overlay real-time process data from the filling, sterilization, and impurity detection processes. Through a time synchronization mechanism, data from each process is mapped to corresponding equipment nodes and edges, forming a key production process interface. This interface reflects equipment status, process parameter changes, and process progress in real time, achieving dynamic integrated display of process data. A process monitoring system monitors the filling pressure and sterilization temperature gradient in the key production process interface in real time. Filling pressure deviation thresholds (e.g., ±0.05 MPa) and abnormal sterilization temperature gradient thresholds (e.g., exceeding 10℃ / minute) are set as judgment criteria. The system uses a sliding window algorithm to smooth sensor data, removing short-term noise and ensuring accurate judgment. When the filling pressure deviation exceeds the threshold, a pressure stabilization command is automatically generated for the filling machine, adjusting its control parameters to restore the normal pressure range. When the sterilization temperature gradient is abnormal, a steam regulation command for the sterilizer is generated, adjusting the steam valve opening to stabilize the temperature curve. Commands are sent to the corresponding equipment in real time via an industrial fieldbus. When the metal detector detects a metal contamination signal, the system synchronously drives the sorting valve to locate the contaminated material based on the equipment's spatial coordinates, generating a rejection command to accurately remove the contaminated material and prevent defective products from flowing out. Based on the pressure stabilization and steam regulation commands generated in step S3, the system updates the process parameter nodes in the production line equipment topology diagram in real time, specifically updating pressure gradients, temperature distributions, and sorting status. The update process utilizes an industrial IoT platform and the OPC UA protocol to transmit data, ensuring information synchronization and consistency. The digital visualization control system dynamically renders the status of each piece of equipment and changes in process parameters on the production line based on the updated topology diagram.The system uses a human-machine interface to display heat maps of equipment pressure and temperature distribution, as well as status indicators of sorting valves. This allows maintenance personnel to monitor the production line in real time and issue operation commands, achieving digital and intelligent control of the food production line. The system employs a two-way communication mechanism to ensure real-time uploading of control command execution feedback, forming a closed-loop control system and improving the stability and safety of the production line operation.

[0085] As an example of the present invention, reference is made to... Figure 2 As shown, in this example, step S1 includes:

[0086] Step S11: Obtain food production line process data; perform process identification and analysis on the food production line process data to extract the filling process, sterilization process and impurity detection process, and generate key food process data;

[0087] Step S12: Match the equipment correspondences of the key food processing data, extract the equipment information corresponding to the filling process, and generate filling machine equipment identification data; based on the filling machine equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the filling machine equipment in the production line, and generate filling machine spatial coordinate data.

[0088] Step S13: Match the equipment correspondences of the key food processing data, extract the equipment information corresponding to the sterilization process, and generate sterilization autoclave equipment identification data; based on the sterilization autoclave equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the sterilization autoclave equipment in the production line, and generate sterilization autoclave spatial coordinate data;

[0089] Step S14: Match the equipment correspondence of the key food process data, extract the equipment information corresponding to the impurity detection process, and generate metal detector equipment identification data; based on the metal detector equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the metal detector equipment in the production line, and generate metal detector spatial coordinate data.

[0090] In this embodiment of the invention, process data collected from the food production line is segmented and analyzed. Pre-defined process classification rules and a process coding system are used to perform process identification on the collected production line process data. The process identification module accurately extracts three key processes—filling, sterilization, and impurity detection—based on the time sequence and event triggering information in the process data. A data parsing algorithm is used to perform structured transformation on the process data, outputting corresponding key food process data. This data includes the process name, start and end times, involved operation steps, and related process parameters, ensuring the completeness and accuracy of the process data. Based on the key food process data generated in step S11, an equipment matching rule base is used to match the equipment information corresponding to the filling process. The rule base includes equipment number, equipment type, functional description, and associated process information. A precise matching algorithm extracts the filling machine equipment identifier from the key process data, ensuring a unique correspondence. Using the filling machine equipment identifier data, the spatial configuration field in the food production line process data is parsed. The spatial configuration field represents the installation position of the equipment in the production line using a three-dimensional coordinate system. The X, Y, and Z spatial parameters in the field are parsed using the coordinate parsing module. The parsed results undergo format conversion and error verification to generate spatial coordinate data for the filling machine, ensuring millimeter-level accuracy and providing a precise basis for subsequent equipment positioning and scheduling. Using a method similar to step S12, equipment correspondence matching is performed on the sterilization process within the key food processing data. This matching process relies on a sterilization equipment information database, including the sterilization kettle's equipment number, model, and process parameters, to accurately extract the sterilization kettle's equipment identification data. Based on the sterilization kettle's equipment identification data, the spatial configuration field in the food production line process data is accessed to parse the sterilization kettle's spatial location within the production line. This parsing process involves multi-sensor data fusion, combining laser ranging and visual positioning results to fuse and correct the spatial coordinate data, obtaining accurate sterilization kettle spatial coordinate data and ensuring consistency between the data and the on-site equipment layout. For the impurity detection process, equipment correspondence matching is performed, extracting metal detector equipment identification data from the key process data based on the metal detector's equipment attributes and function definitions. The equipment identification data is synchronized with the equipment asset management system to ensure real-time information updates. Using the metal detector equipment identification data, the spatial configuration field of the production line process data is parsed, and a spatial positioning module converts the identification information into spatial coordinates. Spatial coordinate analysis comprehensively considers the production line layout and equipment height information to generate spatial coordinate data for the metal detector. The coordinate data undergoes multiple verifications, including checks based on on-site measurements and equipment installation records, to ensure the accuracy and completeness of the spatial coordinates.

[0091] Preferably, step S11 includes the following steps:

[0092] Step S111: Collect process data of the food production line, including PLC process control data, equipment status signals and timestamp sequence information. The data sampling period is set to 100ms to 500ms, and the collection time period is no less than 1.5 times the complete production batch.

[0093] Step S112: Perform structured analysis on the food production line process data and extract key equipment status tags, including the start / stop status of the filling head, the status of the sterilization temperature control valve, and the on / off status of the metal detector; among them, the filling process is determined when the filling head is continuously open for more than 30 seconds and the bottle position sensor is in a continuously changing state.

[0094] Step S113: The rule for determining the "sterilization process" is: the temperature sensor reading is stable in the range of 80°C to 125°C for more than 60 seconds, and the main pump of the corresponding sterilization device is in the "on" state; if there is a sudden temperature change or the temperature is below 70°C for more than 20 seconds, the process segment is interrupted.

[0095] Step S114: The rule for determining the impurity detection process is: the metal detector is in the detection state and the detection signal trigger frequency is greater than 0.2 times / second, or the visual inspection device continuously identifies abnormal image feature areas; any of the above conditions can be marked as an impurity detection process.

[0096] Step S115: Archive and integrate the segment numbers, start and end times, corresponding process labels and key equipment identification codes of the identified filling process, sterilization process and impurity detection process to generate food key process data.

[0097] In this embodiment of the invention, an industrial-grade data acquisition device is used to connect to the PLC (Programmable Logic Controller) system in the production line to collect process control data in real time, including sensor readings, actuator status, and alarm signals. The acquisition frequency is set between 100 milliseconds and 500 milliseconds to ensure the timeliness and completeness of the data. The acquisition time covers at least 1.5 times the duration of a complete production batch to avoid data loss or incompleteness due to process fluctuations and to ensure the accuracy of subsequent process identification. All acquired data is timestamped to form a time-series database, supporting subsequent time synchronization and status correlation analysis. The acquired process data is formatted according to a preset data protocol and communication format, converting the raw data into a structured table or database format. Key equipment status tags are extracted: the start and stop status of the filling head is determined based on the start and stop signals in the PLC; continuous operation for more than 30 seconds is considered a valid filling operation stage; the status changes of the bottle position sensor are monitored; continuous sensor changes indicate that the bottle is in a flow state in the filling line; the fluid control status during the sterilization process is analyzed based on the valve opening sensor and switch signals; the activation signal of the metal detector is read to identify its detection cycle and status changes. By combining the above tags with timestamps, a multi-dimensional state-labeled dataset is formed. Temperature data collected by temperature sensors undergoes sliding window smoothing to filter sensor noise and short-term fluctuations. The condition for determining the sterilization process is that the temperature is stably maintained between 80°C and 125°C for at least 60 seconds. Simultaneously, the main pump status of the sterilization device is verified as "on," and the effectiveness of the sterilization process is confirmed through logical condition linkage. If a temperature sudden change exceeding 10°C is detected and remains below 70°C for more than 20 seconds, the current sterilization process segment is interrupted, the process segment boundary is updated, and it is marked as an abnormal interruption state. All judgment results are recorded with timestamps and relevant sensor data for easy traceability and analysis. Based on the working status signal of the metal detector, the detection trigger frequency is statistically analyzed; a trigger frequency exceeding 0.2 times per second is considered an active detection stage. Combining the output of the visual inspection system, image processing algorithms are used to extract abnormal region features, such as morphological, color, and texture anomalies. The appearance of abnormal features in consecutive frames indicates the continuity of detection. Meeting any of these conditions confirms the impurity detection process segment. All detection events include corresponding timestamps and equipment status, ensuring accurate identification on the process timeline. The segment numbers, start and end times, process labels, and equipment identification codes for each stage of the filling, sterilization, and impurity detection processes determined in steps S112 to S114 are standardized and archived. Each record includes the process type, corresponding equipment number, start and end times, and a description of the process status. Data storage uses a relational database, supporting multi-dimensional queries and analysis by time, equipment, or process type, forming a dataset of key food processing processes. This dataset provides a structured data foundation for digital monitoring of the production line and subsequent intelligent control strategies.

[0098] Preferably, the equipment topology in step S2, which analyzes the food production line process data based on the spatial coordinates of the filling machine, the sterilization vessel, and the metal detector, includes:

[0099] Calculate the spatial path distance between the spatial coordinate data of the filling machine and the spatial coordinate data of the sterilization vessel, and generate relative path data between the filling and sterilization equipment;

[0100] Calculate the spatial path distance between the spatial coordinates of the sterilization autoclave and the spatial coordinates of the metal detector, and generate relative path data between the sterilization and detection equipment.

[0101] The relative path data of filling-sterilization equipment and relative path data of sterilization-detection equipment are fused and analyzed to generate equipment logical connection relationship data.

[0102] Generate graph structure nodes from the logical connection data of the equipment, construct the topology graph node identifier for each equipment, and generate production line equipment node set data;

[0103] Assign edge relationships to the production line equipment node set data, construct a complete topological edge set, and generate production line equipment edge set data; assemble the production line equipment node set data and production line equipment edge set data into a graph structure to generate a production line equipment topology graph.

[0104] In this embodiment of the invention, the spatial straight-line distance between the filling machine and the sterilization vessel is calculated sequentially using a three-dimensional Euclidean distance calculation method, utilizing the spatial coordinate data of the filling machine and the sterilization vessel. If the spatial coordinates contain multiple key points (such as equipment inlet, outlet, or multiple installation points), the shortest path distance between the key points is calculated separately, and the shortest distance is taken as the spatial path distance between the filling machine and the sterilization vessel. The distance unit is meters, and the calculation accuracy is at least two decimal places to meet the precise spatial positioning requirements of industrial production lines. This distance data records the spatial relationship between the equipment, providing a spatial basis for subsequent path fusion analysis. Similarly, a three-dimensional spatial distance calculation is performed on the spatial coordinates of the sterilization vessel and the metal detector. If the production line has complex pipelines or conveyor belt paths, the actual transport path length is calculated in the spatial grid model using a shortest path algorithm (such as Dijkstra's algorithm) based on the equipment installation location and path constraints, serving as the relative path distance between the sterilization and detection equipment. The calculation results are used to determine the transmission delay between the equipment and the rationality of the process. By combining the relative path data between filling and sterilization equipment with that between sterilization and detection equipment, and based on the process flow sequence of the food production line, the equipment paths are analyzed. The rationality of the paths and the preset process sequence rules (filling → sterilization → impurity detection) are determined, and the sequence is matched. Through process timing control rules, spatial path relationships are mapped to logical process relationships, constructing ordered connections between equipment and forming equipment logical connection relationship data. This data includes equipment pairs, connection directions, and connection weights (such as path distance or transmission time estimates). Based on the equipment logical connection relationship data, unique topology node identifiers are assigned to equipment such as filling machines, sterilization kettles, and metal detectors. Each equipment node contains equipment type, equipment identifier code, and corresponding spatial coordinate information, constituting the production line equipment node set data. The node data structure is designed to include fields such as node ID, equipment name, spatial coordinates, and equipment status information, facilitating subsequent graph structure processing and status monitoring. For the production line equipment node set data, the edge set in the topology graph is assigned according to the equipment logical connection relationship data. The edge set contains the starting node ID, ending node ID, and edge weight information (such as spatial path distance, process priority, etc.). A directed graph structure is used, with edges representing the direction of data or material flow, ensuring clear and traceable process logic between devices. The edge set forms a complete production line equipment connection network, supporting path querying and topology analysis. The production line equipment node set data and edge set data are integrated to construct a complete production line equipment topology graph data structure. It is stored using a graph database or adjacency matrix / adjacency list, enabling fast access and updates to nodes and edges. The topology graph includes node attributes and edge attributes, supporting complex queries and dynamic updates. This topology graph serves as the basis for mapping the spatial and logical relationships of production line equipment, providing data support for subsequent dynamic overlay of process data and digital control.

[0105] Preferably, step S2, which involves dynamically overlaying process data for the filling process, sterilization process, and impurity detection process using a production line equipment topology diagram, includes:

[0106] Based on the filling process, sterilization process, and impurity detection process, respectively, the canning pressure data, sterilization temperature data, and metal impurity detection signals are collected.

[0107] Color gradient mapping is performed on filling pressure data to generate filling nozzle color gradient data; structural drift detection is performed on historical filling nozzle color gradient data, and microstructure displacement risks are identified through abnormal fan-shaped offset areas. Deformation fitting verification is performed in combination with the preset filling machine support arm CAD static model to generate structural-level implicit linkage risk map data.

[0108] Calculate the spatial cross-sectional interpolation of the sterilization temperature data to generate thermal map data of the sterilization autoclave cross-section;

[0109] The alarm trigger analysis is performed on the metal impurity detection signal data to control the display status of the sorting valve and generate flashing alarm data for the sorting valve.

[0110] Process location mapping processing is performed on the color gradient data of filling nozzles, thermal map data of sterilizer cross-section, and flashing alarm data of sorting valves to generate process visualization layer data.

[0111] By integrating and overlaying process visualization layer data with production line equipment topology data using structural-level implicit linkage risk map data, a key production process flow interface is generated.

[0112] In this embodiment of the invention, corresponding process data are collected in real time for the filling process, sterilization process, and impurity detection process, respectively. In the filling process, filling pressure data is collected, with a sampling frequency set to once every 200 milliseconds to ensure data timeliness and accuracy. In the sterilization process, internal temperature data of the sterilizer is collected using a multi-point temperature sensor array with a sampling interval of 1 second, covering key cross-sectional areas of the sterilizer. In the impurity detection process, trigger signals from the metal impurity detector are collected, also with a sampling frequency of 200 milliseconds, recording the detection signal status in real time. The collected filling pressure data is converted into color gradient data using a color gradient mapping algorithm. Specifically, the pressure values ​​are first normalized to the 0-1 range, and then, according to a preset color mapping table (e.g., low pressure corresponds to blue, high pressure corresponds to red, with a transition from yellow to orange), a color value for each filling nozzle position is generated. The color gradient data of the filling nozzles is output, including the coordinates of each filling nozzle and its corresponding color value, for subsequent structural drift analysis and visualization. The current filling nozzle color gradient data is compared with historical datasets, and structural drift detection is performed using image differencing and region statistical methods. The specific operations include converting the current and historical color gradient maps into two-dimensional matrices, calculating the color change rate of each corresponding region, and identifying abnormal fan-shaped offset regions (defined as regions with a color change rate exceeding 15% and an area greater than a preset threshold, such as 100 pixels). This method can locate microstructural displacement risk points. Combined with a preset 3D CAD static model of the filling machine support arm, deformation fitting verification is performed on the identified abnormal fan-shaped offset regions. Using finite element analysis (FEA) software, the abnormal regions are mapped onto the surface of the support arm model to simulate stress deformation. By comparing the actual offset data with the simulation results, implicit linkage risks are confirmed, and finally, structural-level implicit linkage risk map data is generated, including displacement, risk level, and location coordinates. Temperature data collected from multiple temperature sensor points inside the sterilizer are used to reconstruct the spatial temperature distribution of the sterilizer cross-section using the Kriging interpolation method. By establishing a three-dimensional temperature space model, detailed thermographic data of the sterilizer cross-section is generated. This thermographic map provides temperature values ​​at different locations on each cross-section for subsequent process location mapping and risk assessment. Real-time analysis of metal impurity detection signals is performed to determine alarm trigger events. The specific rules are as follows: a valid alarm is considered to occur when the detection signal trigger frequency exceeds 0.2 times / second, or when the duration of a single signal exceeds 50 milliseconds. Based on the alarm status, the display status of the corresponding sorting valve is controlled (e.g., the LED flashing frequency is set to 1Hz), generating sorting valve flashing alarm data, including the alarm time, duration, and valve number. The filling nozzle color gradient data, sterilizer cross-sectional heat map data, and sorting valve flashing alarm data are mapped to the production line equipment topology diagram according to their respective process flow positions. Using coordinate transformation and mapping algorithms, each data point is associated with equipment nodes and edges, forming a multi-layered process data layer.This layer data structure includes timestamps, process data types, spatial locations, and display attributes. Based on structural-level implicit linkage risk map data, multi-layer fusion is performed on the process visualization layer data and the production line equipment topology diagram. The fusion employs graphic overlay technology and a unified spatial coordinate algorithm to ensure spatial consistency and logical correlation among the data layers. The final result is a key production process flow interface data integrating pressure, temperature, alarm, and structural risk information, supporting subsequent digital monitoring and control operations.

[0113] Preferably, structural drift detection is performed on historical filling nozzle color gradient data, and microstructural displacement risks are identified through abnormal fan-shaped offset regions. Deformation fitting verification is then performed in conjunction with a pre-set CAD static model of the filling machine support arm, including:

[0114] Historical filling nozzle color gradient data is resampled over time to extract the color gradient boundary change sequence within the filling nozzle's working cycle, generating filling nozzle boundary gradient evolution data.

[0115] Multi-frame boundary contour alignment is performed on the gradient evolution data of the filling nozzle boundary to identify the drift direction of the contour trajectory, construct the offset aggregation region in the sector corner domain, and generate abnormal sector offset region data.

[0116] Multidimensional geometric center drift analysis was performed on the abnormal fan-shaped offset region data to extract the joint features of offset amplitude and angle, and a microstructure displacement probability scoring matrix was constructed to generate microstructure displacement risk scoring data.

[0117] The microstructure displacement risk score data is compared with the preset CAD static model of the filling machine support arm by comparing the node constraints. The residuals are fitted according to the mechanical tolerance of the actual deformable nodes to generate deformation fitting residual data.

[0118] Based on the deformation fitting residual data, key component nodes with high strain concentration trends in the CAD static model of the filling machine support arm are identified, and structural-level implicit linkage risk map data is generated.

[0119] In this embodiment of the invention, historically collected color gradient data of filling nozzles is first resampled according to the working cycle of the filling nozzles over time to ensure uniform sampling intervals and complete coverage of the cyclical changes. Then, the color gradient boundary changes of the filling nozzles at each time point are extracted to form a continuous sequence of color gradient boundary changes, resulting in filling nozzle boundary gradient evolution data, which describes the dynamic changes of the color gradient over time. The boundary contours of multiple frames in the filling nozzle boundary gradient evolution data are spatially aligned using image registration or contour matching algorithms to unify the boundary starting point and direction. By comparing the positional changes of the contours at different time points, the drift direction and offset path of the boundary contours are identified. Based on the contour drift trajectory, fan-shaped angular regions are divided, and regions with concentrated and continuous offsets are aggregated to identify abnormal fan-shaped offset regions. For the identified abnormal fan-shaped offset regions, multi-dimensional spatial geometric center drift analysis is performed to extract joint features of offset amplitude (offset distance) and offset angle. Based on these features, combined with a predefined drift threshold, a probability score for microstructure displacement is calculated, and a risk score matrix is ​​constructed. This score matrix comprehensively considers offset magnitude, directional consistency, and regional continuity, outputting microstructure displacement risk score data. Microstructural displacement risk scoring data is mapped onto a pre-defined static CAD model of the filling machine support arm, and comparisons are made between key nodes and connection constraints in the model. Based on the actual mechanical tolerances of the nodes (such as maximum allowable deformation or stress thresholds), residual fitting is performed on the deformation guided by the risk score. The residual between the theoretical deformation and the actual offset is calculated, generating deformation fitting residual data that reflects the difference between the actual structural offset and the design tolerance. Based on the deformation fitting residual data, key component nodes with high strain concentration in the support arm CAD model are identified, with nodes exhibiting excessive residuals and critical mechanical load-bearing capacity highlighted. By combining the residuals and risk scores of each node, a structural-level implicit linkage risk map is generated. This map visually displays the distribution of microstructural displacement risks in the support arm, assisting in early warning and maintenance decisions.

[0120] Preferably, the spatial cross-sectional interpolation for calculating sterilization temperature data includes:

[0121] Three-dimensional sampling point coordinate mapping is performed on the sterilization temperature data to construct a spatial temperature point set of the inner wall and core area of ​​the sterilization autoclave, and sterilization temperature point cloud data is generated.

[0122] Local thermal field density estimation is performed on the sterilization temperature point cloud data to extract the temperature distribution gradient of different cross-sectional regions and generate initial cross-sectional temperature distribution contour data.

[0123] Two-dimensional bilinear interpolation is performed on the initial cross-sectional temperature distribution profile data, and morphological constraint correction is performed based on the geometry of the sterilization vessel to generate the interpolated cross-sectional temperature grid data of the sterilization vessel.

[0124] The interpolated temperature grid data of the sterilizer section is subjected to thermal classification and pseudo-color mapping to generate thermogram data of the sterilizer section.

[0125] In this embodiment of the invention, the collected sterilization temperature data is mapped to its corresponding three-dimensional spatial coordinates to form a spatial temperature point set covering the inner wall of the sterilization vessel and the core heating area. By integrating these coordinates with the corresponding temperature values, complete sterilization temperature point cloud data is generated, ensuring accurate correlation between spatial and temperature information. For the sterilization temperature point cloud data, a thermal field density estimation technique is used to divide the space into multiple sections (e.g., slices along the vessel height or radial direction), calculating the density distribution and temperature gradient change of temperature points within each section to obtain initial temperature distribution contour data for each section, providing a basis for subsequent interpolation. For the initial temperature distribution contour data of each section, a two-dimensional bilinear interpolation algorithm is applied to smoothly fill the blank areas between temperature points within the section, generating detailed temperature grid data. Based on the geometry of the sterilization vessel (e.g., the curved surface of the cylindrical vessel wall, the range of the core heating area), the interpolation results are morphologically constrained to ensure that the temperature distribution grid conforms to the actual vessel shape and boundary conditions. The interpolated temperature grid data of the sterilizer cross section is divided into thermal zone levels (such as high temperature zone, medium temperature zone, and low temperature zone) according to the temperature gradient and numerical range. The temperature values ​​are visualized by combining pseudo-color mapping rules (such as gradient colors from blue to red), generating intuitive thermographic data of the sterilizer cross section, which is convenient for subsequent process monitoring and anomaly analysis.

[0126] Preferably, step S3 includes the following steps:

[0127] Step S31: Perform regional stability window sliding calculation on the filling pressure data in the key production process interface to generate filling pressure deviation data; fit the steady-state drift trend of the filling pressure deviation data to identify whether there is local non-equilibrium continuous fluctuation, and generate filling pressure anomaly identification result data.

[0128] Step S32: Perform isothermal zone morphological connectivity analysis on the thermal map data of the sterilization autoclave section, extract the thermal gradient fault zone area, and generate sterilization temperature gradient anomaly layer data.

[0129] Step S33: Determine whether the abnormal filling pressure identification result data exceeds the preset threshold. If it does, generate a filling machine pressure stabilization command. Determine whether there is a high-risk area in the abnormal sterilization temperature gradient layer data. If so, generate a sterilization autoclave steam adjustment command.

[0130] Step S34: Confirm the status of the impurity signal triggered by the metal detector through the pressure stabilization command of the filling machine or the steam regulation command of the sterilizer, and generate a sorting valve rejection action command according to its position to perform the metal contaminant rejection operation.

[0131] In this embodiment of the invention, filling pressure data collected from the interface of key production processes is analyzed using a regional stability window sliding calculation method. The pressure data is segmented by a fixed-size time window, and the pressure deviation within each window is calculated to generate filling pressure deviation data. This pressure deviation data is further processed using a steady-state region drift trend fitting technique to detect localized and persistent pressure fluctuations, identify abnormal pressure trends, and generate filling pressure anomaly identification results. For the thermographic data of the sterilization autoclave cross-section, morphological connectivity analysis of the isothermal zone is performed to identify thermal gradient fault zones in the temperature distribution. These fault zones are extracted as temperature gradient anomaly regions, forming a sterilization temperature gradient anomaly layer data to indicate potential thermal unevenness or heating failure risk points. Based on the filling pressure anomaly identification results data, a preset pressure anomaly threshold is compared. If the anomaly exceeds the threshold, a filling machine pressure stabilization command is automatically generated to adjust the filling pressure parameters and stabilize the production process. Simultaneously, based on the abnormal layer data of the sterilization temperature gradient, it is determined whether there are high-risk hot zones. If such areas are detected, a steam regulation command for the sterilization autoclave is generated to adjust the temperature distribution during the sterilization process, ensuring sterilization effectiveness and equipment safety. Using the aforementioned pressure stabilization command from the filling machine or the steam regulation command from the sterilization autoclave, combined with impurity signal data triggered by the metal detector, status confirmation is performed to ensure the accuracy and real-time nature of the detection. Based on the specific location of the impurity signal, a rejection action command for the sorting valve is generated synchronously, driving the sorting valve to accurately reject metal contaminants, ensuring the quality of materials on the production line.

[0132] Of particular importance, the isothermal morphological connectivity analysis of the thermographic data of the sterilization vessel cross-section also includes:

[0133] Pixel-level temperature grading is performed on the thermal map data of the sterilization autoclave cross section to generate temperature zone image data;

[0134] Perform isothermal region labeling operations based on temperature partition image data to generate isothermal region connectivity labeling maps;

[0135] Extract the boundaries of connected regions from the isothermal region connectivity map to generate isothermal region boundary morphology data;

[0136] Analyze the fracture edge structure in the isothermal zone boundary morphology data, identify the gradient fracture trend, and generate candidate regions for thermal gradient faults.

[0137] Perform temperature abruptness rate calculation on the candidate region of thermal gradient tomography, filter out regions with abruptness rate exceeding a set threshold, and generate temperature abruptness anomaly fragment data;

[0138] Layer projection annotation is performed on abnormal temperature change fragment data to generate sterilization temperature gradient abnormal layer data.

[0139] In this embodiment of the invention, pixel-level temperature grading processing is performed on the acquired thermogram data of the sterilization vessel cross-section. Based on a preset temperature level division (e.g., each 5°C level), the thermogram pixels are assigned to different levels according to their temperature values, generating temperature partition image data with distinguishable color bands or grayscale levels for subsequent region labeling. A connected component analysis algorithm (e.g., 8-connectivity or 4-connectivity) is used to label each isothermal region in the temperature partition image data, assigning each connected isothermal region an independent label number, generating an isothermal region connectivity labeling map, clarifying the spatial range and connectivity status of each isothermal region. In the isothermal region connectivity labeling map, boundary tracing or contour extraction algorithms (e.g., Sobel edge detection or contour fitting) are used to extract the outer edge contour line of each isothermal region, generating isothermal region boundary morphology data. This boundary data can be used to analyze the uniformity of heat distribution and the integrity of local structures. Geometric structure analysis is performed on the isothermal region boundary morphology data to identify regions with structural features such as boundary breaks, acute angle shear, and non-closed edges, marking them as candidate regions where there may be abrupt changes in thermal gradient, i.e., thermal gradient fault candidate regions. This step helps identify phenomena such as abnormal heating and insulation failure. Within the candidate region of the thermal gradient tomography, the temperature abrupt change rate (i.e., the temperature change per unit pixel distance) is calculated for the temperature values ​​on both sides of the boundary. If the abrupt change rate exceeds a set threshold (e.g., >10°C / px), it is identified as an abnormal temperature abrupt change region, and corresponding segments are selected to generate abnormal temperature abrupt change segment data. This abnormal temperature abrupt change segment data is then mapped back to the corresponding location area in the original thermal map using a layer overlay method, generating sterilization temperature gradient abnormal layer data. This layer can be used to visualize areas of uneven temperature or localized abnormalities within the sterilization autoclave, assisting in heat distribution optimization or equipment maintenance decisions.

[0140] Of particular importance, analyzing the fracture edge structure in isothermal zone boundary morphology data and identifying gradient fracture trends also includes:

[0141] High curvature point detection is performed on the boundary morphology data of the isothermal zone to extract local severe bending points in the boundary curve and generate high curvature edge point set data.

[0142] The curve bending angle distribution is calculated based on the high curvature edge point set data, and the curvature direction abrupt change segments are screened out to generate boundary abrupt change segment data.

[0143] Perform normal direction vector clustering on boundary abrupt segment data to identify continuous paths where the boundary orientation changes abruptly, and generate candidate trajectory data for fracture orientation;

[0144] By overlaying and analyzing the candidate trajectory data of the fracture direction with the gradient difference data between adjacent temperature levels, boundary segments with high gradient transition characteristics are extracted to generate gradient anomaly coupling path data.

[0145] Spatial connectivity analysis is performed on gradient anomaly coupled path data to screen out path clusters with high spatial continuity and abrupt temperature differences, generating candidate region data for thermal gradient faults.

[0146] In this embodiment of the invention, for each boundary curve in the isothermal zone boundary morphology data, a geometric analysis method based on the second derivative or curvature function is applied to calculate the curvature of its contour curve point by point, and edge points whose local curvature values ​​exceed a set threshold are extracted. Typically, fitting the rate of change of tangent slope or circular arc fitting methods can be used to locate points with strong bends, generating a high-curvature edge point set data to characterize the tendency of boundary abrupt changes. Using the high-curvature edge point set data, a local bending angle sequence of the boundary curve is constructed, and the distribution statistics of the angles between consecutive point pairs are performed to extract continuous segments where the angle jump exceeds a set threshold (e.g., ±45°), generating boundary abrupt change segment data. This step is used to mark areas of dramatic curve changes, providing input for subsequent trend trajectory analysis. For each point in the boundary abrupt change segment data, its normal direction vector is calculated, and K-means or density clustering algorithms (e.g., DBSCAN) are applied for vector direction clustering to identify boundary segments where the normal direction changes abruptly or the cluster centers switch rapidly. Combined with the curve's connectivity, candidate trajectory data for fracture directions is extracted. These trajectories reflect potential fractures in the thermal gradient boundary paths within the isothermal structure. For each point in the boundary abrupt change segment data, its normal direction vector is calculated. K-means or density clustering algorithms (such as DBSCAN) are applied for vector direction clustering to identify boundary segments where the normal direction changes abruptly or the cluster centers switch rapidly. Combined with curve connectivity, candidate trajectory data for fracture paths are extracted. These trajectories reflect potential fractures in the thermal gradient boundary paths within the isothermal structure. Connectivity analysis is performed on the gradient anomaly coupling path data. A graph structure modeling method is used to construct a path node network. Path clusters with high spatial continuity, directional consistency, and temperature difference abrupt change indices are aggregated and filtered, ultimately outputting candidate region data for thermal gradient faults. This data is used to locate areas inside the sterilization vessel where there may be abnormal heat conduction, turbulent flow fields, or heating dead zones.

[0147] Preferably, step S34 includes the following steps:

[0148] Step S341: Receive the pressure stabilization command from the filling machine and the steam regulation command from the sterilizer, parse the command parameters, where the pressure stabilization command has a pressure range of 0.5MPa~1.2MPa and the steam regulation command has a temperature range of 100°C~130°C, and ensure that the command execution status is "confirmed".

[0149] Step S342: Monitor the impurity signal triggered by the metal detector in real time. The impurity signal strength threshold is set to >50mV, and the signal duration is not less than 10ms to be considered a valid trigger.

[0150] Step S343: Based on the timestamp and location code of the metal detector trigger signal, and combined with the current pressure stabilization and steam status of the filling machine and sterilizer, determine whether the impurity signal is within the normal production operating range; when both the pressure stabilization command and the steam command are within the fluctuation range of ±5%, confirm that the impurity signal is valid.

[0151] Step S344: Based on the confirmed impurity signal, calculate the trigger time of the sorting valve rejection action. Considering the product transmission speed, the conveyor belt speed range is set to 0.2~1.0m / s, and the rejection response time error does not exceed ±50ms. Generate a synchronous sorting valve rejection action command; send the rejection action command to the sorting valve control unit to ensure that the sorting valve action range covers the spatial position of the material corresponding to the impurity signal, and the opening and closing action time of the sorting valve is controlled within 100ms.

[0152] Step S345: Record the rejection action execution log, including the impurity signal trigger time, the sorting valve action time, and the rejection success flag, to obtain the sorting valve response configuration data for performing the metal contaminant rejection operation.

[0153] In this embodiment of the invention, the system receives pressure stabilization commands from the filling machine and steam regulation commands from the sterilizer. Key parameters in the commands are analyzed, with the pressure stabilization range set to 0.5 MPa to 1.2 MPa and the steam temperature range set to 100°C to 130°C. The execution status of the commands is confirmed as "confirmed," ensuring their validity and executability. The impurity signal emitted by the metal detector is continuously monitored. A valid trigger is defined as an impurity signal strength threshold greater than 50 millivolts and a signal duration of at least 10 milliseconds. Based on the timestamp and location code of the metal detector trigger signal, combined with the current pressure stabilization status of the filling machine and the steam regulation status of the sterilizer, it is determined whether the impurity signal is within the normal production operating range. When both the pressure stabilization command and the steam regulation command remain within a fluctuation range of ±5%, the impurity signal is confirmed as valid. Based on the confirmed valid impurity signal and the product's movement speed on the conveyor belt, the trigger time for the sorting valve's rejection action is calculated. The conveyor belt speed is set between 0.2 m / s and 1.0 m / s, and the rejection response time error is controlled within ±50 milliseconds. A synchronized rejection command for the sorting valve is generated and sent to the sorting valve control unit. The operating range of the sorting valve is ensured to cover the spatial location of the material corresponding to the impurity signal, and the opening and closing time of the sorting valve is controlled within 100 milliseconds. An execution log of the sorting valve rejection action is recorded, including the impurity signal trigger time, the sorting valve action time, and the rejection success flag. Sorting valve response configuration data is generated for subsequent quality tracking and system optimization. Through the above process, accurate rejection of metal contaminants and the safe and stable operation of the production line are ensured.

[0154] As an example of the present invention, reference is made to... Figure 3 As shown, step S4 in this example includes:

[0155] Step S41: Adjust the pressure control parameters of the filling machine according to the pressure stabilization command of the filling machine, and generate pressure adjustment execution data; adjust the steam valve opening according to the steam regulation command of the sterilizer, and generate steam flow adjustment data;

[0156] Step S42: Locate the impurity signal data triggered by the metal detector to generate impurity signal location data; combine the impurity signal location data with the sorting valve response configuration data to generate sorting valve rejection action trigger data;

[0157] Step S43: Control the sorting valve to perform the rejection operation based on the rejection action trigger data of the sorting valve, and generate rejection operation execution result data; perform feedback monitoring on the rejection operation execution result data, confirm the rejection effect, and generate rejection status confirmation data;

[0158] Step S44: Update the production line equipment topology diagram in real time based on pressure adjustment execution data, steam flow adjustment data, and rejection status confirmation data to perform digital visualization control operations of the food production line control system.

[0159] In this embodiment of the invention, the pressure control parameters of the filling machine are adjusted according to the pressure stabilization command issued by the food production line control system to ensure that the filling pressure remains stable within the set range. The adjustment process is monitored in real time by a pressure sensor, generating pressure adjustment execution data that reflects the current pressure parameter changes. Simultaneously, the opening of the steam valve is adjusted according to the steam regulation command of the sterilizer to achieve precise control of the steam flow rate within the sterilizer. Steam flow adjustment data is generated synchronously with the adjustment action for subsequent monitoring and feedback. The spatial location of the impurity signal triggered by the metal detector is determined, identifying the specific detection point of the metal impurity in the production line and generating impurity signal location data. Combined with preset sorting valve response configuration data (including valve position, response time, and action range), sorting valve rejection action trigger data is generated based on the impurity signal location, ensuring accurate and timely response from the sorting valve for targeted rejection. Based on the sorting valve rejection action trigger data, the sorting valve is controlled to execute the rejection action, removing the detected metal contaminant from the production line, generating rejection operation execution result data, and recording the time, status, and success or failure of the rejection action. Sensors installed in the rejection area provide real-time feedback monitoring of the rejection action's effectiveness, confirming whether impurities have been successfully removed and generating rejection status confirmation data. This data serves as a basis for subsequent production process adjustments and anomaly alarms. Based on pressure adjustment execution data, steam flow adjustment data, and rejection status confirmation data, the system updates the equipment operating status, pressure gradient distribution, temperature parameters, and rejection action status in the production line equipment topology diagram in real time. This ensures that the topology diagram in the digital control system accurately reflects the current actual operating status of the production line. The updated topology diagram supports visualized management and intelligent decision-making in the food production line control system, improving the transparency and control precision of the production process and achieving digital closed-loop control.

[0160] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the application be incorporated into the invention.

[0161] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A digital visualization control method for a food production line control system, characterized in that, Includes the following steps: Step S1: Obtain food production line process data; Extract the filling process, sterilization process, and impurity detection process from the food production line process data, and locate the spatial coordinates of the process operation equipment for each process to obtain the spatial coordinates of the filling machine, sterilization kettle, and metal detector. Step S2: Analyze the equipment topology of the food production line process data based on the spatial coordinates of the filling machine, sterilization vessel, and metal detector to construct a production line equipment topology map. Use this production line equipment topology map to dynamically overlay process data for the filling, sterilization, and impurity detection processes, generating key production process interfaces. This dynamic overlay of process data for the filling, sterilization, and impurity detection processes using the production line equipment topology map includes: Based on the filling process, sterilization process, and impurity detection process, respectively, the canning pressure data, sterilization temperature data, and metal impurity detection signals are collected. Color gradient mapping is performed on filling pressure data to generate filling nozzle color gradient data; structural drift detection is performed on historical filling nozzle color gradient data, and microstructure displacement risks are identified through abnormal fan-shaped offset areas. Deformation fitting verification is performed in combination with the preset filling machine support arm CAD static model to generate structural-level implicit linkage risk map data. Calculate the spatial cross-sectional interpolation of the sterilization temperature data to generate thermal map data of the sterilization autoclave cross-section; The alarm trigger analysis is performed on the metal impurity detection signal data to control the display status of the sorting valve and generate flashing alarm data for the sorting valve. Process location mapping processing is performed on the color gradient data of filling nozzles, thermal map data of sterilizer cross-section, and flashing alarm data of sorting valves to generate process visualization layer data. By integrating and overlaying process visualization layer data and production line equipment topology diagrams with structural-level implicit linkage risk map data, a key production process flow interface is generated. Step S3: Determine whether the filling pressure deviation and abnormal sterilization temperature gradient at the interface of the key production process exceed the preset threshold. If so, generate a filling machine pressure stabilization command or a sterilizer steam adjustment command, and drive the sorting valve to remove metal contaminant materials. Step S4: Update the pressure gradient, temperature distribution and sorting status in the production line equipment topology diagram in real time according to the pressure stabilization command of the filling machine or the steam adjustment command of the sterilizer, so as to execute the digital visualization control operation of the food production line control system.

2. The digital visualization control method for a food production line control system according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain food production line process data; perform process identification and analysis on the food production line process data to extract the filling process, sterilization process and impurity detection process, and generate key food process data; Step S12: Match the equipment correspondences of the key food processing data, extract the equipment information corresponding to the filling process, and generate filling machine equipment identification data; based on the filling machine equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the filling machine equipment in the production line, and generate filling machine spatial coordinate data. Step S13: Match the equipment correspondences of the key food processing data, extract the equipment information corresponding to the sterilization process, and generate sterilization autoclave equipment identification data; based on the sterilization autoclave equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the sterilization autoclave equipment in the production line, and generate sterilization autoclave spatial coordinate data; Step S14: Match the equipment correspondence of the key food process data, extract the equipment information corresponding to the impurity detection process, and generate metal detector equipment identification data; based on the metal detector equipment identification data, parse the spatial configuration field in the food production line process data, extract the spatial position of the metal detector equipment in the production line, and generate metal detector spatial coordinate data.

3. The digital visualization control method for a food production line control system according to claim 2, characterized in that, Step S11 includes the following steps: Step S111: Collect process data of the food production line, including PLC process control data, equipment status signals and timestamp sequence information. The data sampling period is set to 100ms to 500ms, and the collection time period is no less than 1.5 times the complete production batch. Step S112: Perform structured analysis on the food production line process data and extract key equipment status tags, including the start / stop status of the filling head, the status of the sterilization temperature control valve, and the on / off status of the metal detector; among them, the filling process is determined when the filling head is continuously open for more than 30 seconds and the bottle position sensor is in a continuously changing state. Step S113: The rule for determining the "sterilization process" is: the temperature sensor reading is stable in the range of 80°C to 125°C for more than 60 seconds, and the main pump of the corresponding sterilization device is in the "on" state; if there is a temperature change or the temperature is below 70°C for more than 20 seconds, the process segment is interrupted. Step S114: The rule for determining the impurity detection process is: the metal detector is in the detection state and the detection signal trigger frequency is greater than 0.2 times / second, or the visual inspection device continuously identifies abnormal image feature areas; any of the above conditions can be marked as an impurity detection process. Step S115: Archive and integrate the segment numbers, start and end times, corresponding process labels and key equipment identification codes of the identified filling process, sterilization process and impurity detection process to generate food key process data.

4. The digital visualization control method for a food production line control system according to claim 1, characterized in that, The equipment topology analysis of the food production line process data based on the spatial coordinates of the filling machine, the sterilization vessel, and the metal detector in step S2 includes: Calculate the spatial path distance between the spatial coordinate data of the filling machine and the spatial coordinate data of the sterilization vessel, and generate relative path data between the filling and sterilization equipment; Calculate the spatial path distance between the spatial coordinates of the sterilization autoclave and the spatial coordinates of the metal detector, and generate relative path data between the sterilization and detection equipment. The relative path data of filling-sterilization equipment and relative path data of sterilization-detection equipment are fused and analyzed to generate equipment logical connection relationship data. Generate graph structure nodes from the logical connection data of the equipment, construct the topology graph node identifier for each equipment, and generate production line equipment node set data; Assign edge relationships to the production line equipment node set data, construct a complete topological edge set, and generate production line equipment edge set data; assemble the production line equipment node set data and production line equipment edge set data into a graph structure to generate a production line equipment topology graph.

5. The digital visualization control method for a food production line control system according to claim 1, characterized in that, Structural drift detection is performed on historical filling nozzle color gradient data. Microstructural displacement risks are identified through abnormal fan-shaped offset regions. Deformation fitting verification is performed using a pre-set CAD static model of the filling machine support arm, including: Historical filling nozzle color gradient data is resampled over time to extract the color gradient boundary change sequence within the filling nozzle's working cycle, generating filling nozzle boundary gradient evolution data. Multi-frame boundary contour alignment is performed on the gradient evolution data of the filling nozzle boundary to identify the drift direction of the contour trajectory, construct the offset aggregation region in the sector corner domain, and generate abnormal sector offset region data. Multidimensional geometric center drift analysis was performed on the abnormal fan-shaped offset region data to extract the joint features of offset amplitude and angle, and a microstructure displacement probability scoring matrix was constructed to generate microstructure displacement risk scoring data. The microstructure displacement risk score data is compared with the preset CAD static model of the filling machine support arm by comparing the node constraints. The residuals are fitted according to the mechanical tolerance of the actual deformable nodes to generate deformation fitting residual data. Based on the deformation fitting residual data, key component nodes with high strain concentration trends in the CAD static model of the filling machine support arm are identified, and structural-level implicit linkage risk map data is generated.

6. The digital visualization control method for a food production line control system according to claim 1, characterized in that, Spatial cross-sectional interpolation for calculating sterilization temperature data includes: Three-dimensional sampling point coordinate mapping is performed on the sterilization temperature data to construct a spatial temperature point set of the inner wall and core area of ​​the sterilization autoclave, and sterilization temperature point cloud data is generated. Local thermal field density estimation is performed on the sterilization temperature point cloud data to extract the temperature distribution gradient of different cross-sectional regions and generate initial cross-sectional temperature distribution contour data. Two-dimensional bilinear interpolation is performed on the initial cross-sectional temperature distribution profile data, and morphological constraint correction is performed based on the geometry of the sterilization vessel to generate the interpolated cross-sectional temperature grid data of the sterilization vessel. The interpolated temperature grid data of the sterilizer section is subjected to thermal classification and pseudo-color mapping to generate thermogram data of the sterilizer section.

7. The digital visualization control method for a food production line control system according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Perform regional stability window sliding calculation on the filling pressure data in the key production process interface to generate filling pressure deviation data; fit the steady-state drift trend of the filling pressure deviation data to identify whether there is local non-equilibrium continuous fluctuation, and generate filling pressure anomaly identification result data. Step S32: Perform isothermal zone morphological connectivity analysis on the thermal map data of the sterilization autoclave section, extract the thermal gradient fault zone area, and generate sterilization temperature gradient anomaly layer data. Step S33: Determine whether the abnormal filling pressure identification result data exceeds the preset threshold. If it does, generate a filling machine pressure stabilization command. Determine whether there is a high-risk area in the abnormal sterilization temperature gradient layer data. If so, generate a sterilization autoclave steam adjustment command. Step S34: Confirm the status of the impurity signal triggered by the metal detector through the pressure stabilization command of the filling machine or the steam regulation command of the sterilizer, and generate a sorting valve rejection action command according to its position to perform the metal contaminant rejection operation.

8. The digital visualization control method for a food production line control system according to claim 7, characterized in that, Step S34 includes the following steps: Step S341: Receive the pressure stabilization command from the filling machine and the steam regulation command from the sterilizer, parse the command parameters, where the pressure stabilization command has a pressure range of 0.5MPa~1.2MPa and the steam regulation command has a temperature range of 100°C~130°C, and ensure that the command execution status is "confirmed". Step S342: Monitor the impurity signal triggered by the metal detector in real time. The impurity signal strength threshold is set to >50mV, and the signal duration is not less than 10ms to be considered a valid trigger. Step S343: Based on the timestamp and location code of the metal detector trigger signal, and combined with the current pressure stabilization and steam status of the filling machine and sterilizer, determine whether the impurity signal is within the normal production operating range; when both the pressure stabilization command and the steam command are within the fluctuation range of ±5%, confirm that the impurity signal is valid. Step S344: Based on the confirmed impurity signal, calculate the trigger time of the sorting valve rejection action. Considering the product transmission speed, the conveyor belt speed range is set to 0.2~1.0m / s, and the rejection response time error does not exceed ±50ms. Generate a synchronous sorting valve rejection action command; send the rejection action command to the sorting valve control unit to ensure that the sorting valve action range covers the spatial position of the material corresponding to the impurity signal, and the opening and closing action time of the sorting valve is controlled within 100ms. Step S345: Record the rejection action execution log, including the impurity signal trigger time, the sorting valve action time, and the rejection success flag, to obtain the sorting valve response configuration data for performing the metal contaminant rejection operation.

9. The digital visualization control method for a food production line control system according to claim 8, characterized in that, Step S4 includes the following steps: Step S41: Adjust the pressure control parameters of the filling machine according to the pressure stabilization command of the filling machine, and generate pressure adjustment execution data; adjust the steam valve opening according to the steam regulation command of the sterilizer, and generate steam flow adjustment data; Step S42: Locate the impurity signal data triggered by the metal detector to generate impurity signal location data; combine the impurity signal location data with the sorting valve response configuration data to generate sorting valve rejection action trigger data; Step S43: Control the sorting valve to perform the rejection operation based on the rejection action trigger data of the sorting valve, and generate rejection operation execution result data; perform feedback monitoring on the rejection operation execution result data, confirm the rejection effect, and generate rejection status confirmation data; Step S44: Update the production line equipment topology diagram in real time based on pressure adjustment execution data, steam flow adjustment data, and rejection status confirmation data to perform digital visualization control operations of the food production line control system.

Images