Digital visual control system and method applied to food processing production line

Abstract

The invention relates to the technical field of food processing, in particular to a digital visual control system and method applied to a food processing production line. The method comprises the following steps: constructing a non-dairy creamer production line simulation model based on food processing production line design data and structure acquisition data; then, disturbance coupling instability characteristics in the machining process are recognized through the simulation model, and the pressure-bearing response dramatic increase trend is further detected so as to predict the abnormal deviation chain type growth situation of the production line; attenuation detection and working condition deviation judgment are carried out on the regulation and control efficiency, so that the quality abnormal deviation condition of the non-dairy creamer product is identified; the abnormal deviation data is traced, the defect position of the production line is positioned, targeted optimization control is executed according to the defect position, and production line optimization control data is generated; by optimizing the food processing production line, the production efficiency and quality of the non-dairy creamer food processing production line are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of food processing, and in particular to a digital visualization control system and method applied to a food processing production line. Background Art

[0002] Traditional food processing production lines rely on PLC logic control, timer triggering, and manual inspections to manage processes and respond to exceptions. This approach suffers from complex information acquisition layers, insufficient real-time response, delayed status feedback, and a lack of overall system coordination. This makes it difficult to meet the dynamic linkage and refined control requirements of modern high-precision, high-efficiency food processing. This is particularly true in the processing of non-dairy creamer powders, which involve multiple, highly dynamic processes, including raw material mixing, pipeline transportation, spray drying, hot air dehydration, powder cooling, and packaging. This requires highly coordinated equipment control, consistent processing rhythms across each stage, and precise adjustment of flow parameters. During process execution, issues such as untimely nozzle angle adjustment leading to abnormal powder size distribution, hot air temperature fluctuations causing drying deviations, and uneven packaging pressure leading to product agglomeration frequently arise, requiring higher-dimensional system monitoring to address. However, traditional digital visualization control for food processing production lines suffers from inaccurate prediction and detection of abnormal deviations from the chain growth trend. Summary of the Invention

[0003] Based on this, it is necessary to provide a digital visualization control system and method for food processing production lines to solve at least one of the above technical problems.

[0004] To achieve the above object, a digital visualization control method applied to a food processing production line comprises the following steps: Step S1: Acquire design data of a food processing production line; collect internal processing structure data of a non-dairy creamer production line based on the design data of the food processing production line; and construct a simulation model of a non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line; Step S2: determining the processing disturbance coupling instability characteristics based on the non-dairy creamer processing line simulation model; detecting the production line pressure response surge trend based on the processing disturbance coupling instability characteristics; and predicting the abnormal deviation chain growth trend based on the production line pressure response surge trend; Step S3: detecting a decrease in processing control efficiency based on the abnormal deviation from the chain growth trend; determining a production line operating condition deviation based on the processing control efficiency decrease; and determining a non-dairy creamer product quality abnormal deviation based on the production line operating condition deviation; Step S4: based on the abnormal deviation of non-dairy creamer product quality, the defect position of the non-dairy creamer production line is traced to obtain the production line defect position data; based on the production line defect position data, the non-dairy creamer processing production line is optimized and controlled to obtain the non-dairy creamer processing production line optimization control data.

[0005] The present invention realizes the accurate restoration of the structure and processing characteristics of the non-dairy creamer production line by constructing a processing simulation model, thereby improving the correspondence between the design data and the actual working conditions. On this basis, through the analysis of the disturbance coupling characteristics, it is possible to timely identify the potential unstable trend of the production system, capture the surge in pressure response and the deviation growth chain characteristics in advance, and effectively enhance the recognition accuracy of system anomalies. Combined with the control efficiency attenuation data, it further realizes the dynamic tracking of the working condition deviation, thereby accurately identifying the source of quality anomalies, providing a basis for subsequent quality tracing and correction. Based on the accurate defect position determination results, the system can implement targeted control optimization, so that the overall control strategy fits the actual load and structural fluctuation state, significantly improves the control accuracy and product consistency, and reduces the system operation energy consumption and quality fluctuation amplitude. This method improves the response speed, stability and quality closed-loop capability of the non-dairy creamer production line as a whole, and realizes the full-chain closed-loop linkage of processing structure-working condition fluctuation-product anomaly-defect positioning-optimization control. Therefore, the present invention is an optimization processing of the traditional digital visualization control applied to food processing production lines, which solves the problems of inaccurate prediction of abnormal deviation from chain growth trend and inaccurate detection of abnormal deviation from chain growth trend in the traditional digital visualization control applied to food processing production lines, and improves the accuracy of prediction of abnormal deviation from chain growth trend and the accuracy of detection of abnormal deviation from chain growth trend.

[0006] The present invention further provides a digital visualization control system for a food processing production line, which is used to execute the digital visualization control method for a food processing production line as described above. The digital visualization control system for a food processing production line comprises: A model building module is used to obtain the design data of the food processing production line; based on the design data of the food processing production line, collect the internal processing structure data of the non-dairy creamer production line; and build a simulation model of the non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line; The module predicts the chain growth trend of abnormal deviations, which is used to determine the processing disturbance coupling instability characteristics based on the non-dairy creamer processing line simulation model; detect the production line pressure response surge trend based on the processing disturbance coupling instability characteristics; and predict the chain growth trend of abnormal deviations based on the production line pressure response surge trend; The quality abnormal deviation determination module is used to detect the decline of processing control efficiency based on the abnormal deviation chain growth trend; determine the deviation of production line working condition based on the decline of processing control efficiency; and determine the abnormal deviation of non-dairy creamer product quality based on the deviation of production line working condition; The optimization control processing module is used to trace the defect location of the non-dairy creamer production line based on the abnormal deviation of the non-dairy creamer product quality, and obtain the production line defect location data; based on the production line defect location data, the non-dairy creamer processing production line is optimized and controlled to obtain the optimization control data of the non-dairy creamer processing production line.

[0007] The digital visualization control system applied to a food processing production line of the present invention can realize any digital visualization control method applied to a food processing production line of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the digital visualization control method applied to the food processing production line. The internal modules of the system cooperate with each other to realize a digital closed loop of the entire process of the non-dairy creamer processing production line from structural modeling, disturbance identification, regulation diagnosis to defect tracing and control optimization, thereby improving the operation stability of the production line and the consistency of product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A schematic flow chart of the steps of a digital visualization control method applied to a food processing production line; Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG. Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG. The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0009] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0010] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0011] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0012] To achieve this, please refer to Figures 1 to 3 , a digital visualization control method applied to a food processing production line, comprising the following steps: Step S1: Acquire design data of a food processing production line; collect internal processing structure data of a non-dairy creamer production line based on the design data of the food processing production line; and construct a simulation model of a non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line; In one embodiment of the present invention, an integrated industrial data acquisition system extracts design drawings and structural parameter information for the food processing line to which the non-dairy creamer belongs from the production design department's engineering database. This acquired data includes, but is not limited to, equipment layout diagrams, transmission path distribution, logical connections between unit modules, structural parameters of the heat treatment unit and mixing module, process flow cycle parameters, material flow time windows, temperature control zone setting parameters, and the location coordinates of key processing nodes. This information is converted to a standardized format using the industrial data interface protocol OPC-UA and then imported into the production line digital modeling platform. Based on this imported standardized design data, a 3D physical simulation tool (such as Siemens ProcessSimulate or Visual Components) is used to construct a simulation model of the internal structure of the non-dairy creamer processing line. During the modeling process, the acquired layout diagrams and module structure serve as the geometric foundation for the simulation space. A virtual device structure entity is constructed through parameter matching, and corresponding processing function attributes are assigned to each node. For example, material drying time, inlet air temperature, and hot air circulation path are set for the granulation unit; shear rate and mixing uniformity distribution data are assigned for the emulsification and mixing unit. Based on the process timeline in the design drawings, we leveraged timing simulation to align the various stages of the non-dairy creamer production process, establishing an end-to-end dynamic production process logic. The resulting data model includes information such as the 3D spatial structure, functional node characteristics, and dynamic process timing, providing the underlying structural support for subsequent disturbance detection and data analysis.

[0013] Step S2: determining the processing disturbance coupling instability characteristics based on the non-dairy creamer processing line simulation model; detecting the production line pressure response surge trend based on the processing disturbance coupling instability characteristics; and predicting the abnormal deviation chain growth trend based on the production line pressure response surge trend; In an embodiment of the present invention, after the above-mentioned simulation model is constructed, the real-time operating data of the non-dairy creamer processing process is collected through an industrial signal acquisition module (such as a Siemens S7 series PLC acquisition system), including temperature, pressure, flow rate, shear rate, motor load, current change, material concentration and other signals, and is synchronously recorded at a time interval of 5 seconds. All collected signals are denoised, normalized and processed with sampling frequency consistency through an edge data preprocessing unit, and embedded into the simulation model node as input data. By mapping the actual signal data to the processing nodes of the simulation model, the degree of offset between the simulation working conditions and the standard operating conditions in multiple dimensions is compared to determine the coupling disturbance situation. During the detection process, an interference coupling chain diagram is constructed using indicators such as the processing delay difference between nodes, the energy consumption fluctuation intensity of the unit processing module, the local temperature gradient fluctuation rate, and the shear stability change amplitude, and the unstable coupling area is marked with a threshold value of more than 0.2s for the fluctuation conduction rate between nodes. Based on the above-mentioned instability coupling chain diagram, combined with the time sequence of the occurrence of instability characteristics, the response state change trend of each functional module in the non-dairy creamer production line is judged, and the specific starting point module and conduction path of the pressure response surge trend are identified. For example: when the shear temperature inside the mixing unit is detected to rise rapidly and is transmitted to the preheating zone within 8 seconds, accompanied by a sharp increase in energy consumption and uneven material flow rate, it is marked as a high-risk coupling response area. According to the structural mapping of the surge response path, the response data of other nodes in the processing flow are tracked to detect whether there are chain trend characteristics such as continuous response lag, operation deviation from synchronous beat, and cumulative increase in load index. If there is a delayed response time of >0.5s, load ratio >1.4, and output deviation >15% for three consecutive modules, it is recorded as an abnormal deviation from the chain growth trend. The data output includes the coupling instability segment index, the response surge path diagram, and the abnormal deviation trend indicator set.

[0014] Step S3: detecting a decrease in processing control efficiency based on the abnormal deviation from the chain growth trend; determining a production line operating condition deviation based on the processing control efficiency decrease; and determining a non-dairy creamer product quality abnormal deviation based on the production line operating condition deviation; In this embodiment of the present invention, based on the abnormal deviation chain growth trend data identified in step S2, the automated control response capabilities of each production line's functional modules are further analyzed. Specifically, by accessing SCADA system logs and PLC command response data, the control behavior of each module is reversely reconstructed, analyzing the response time difference from the occurrence of the disturbance to the execution of the response command, the number of control command repetitions, the target value recovery rate, and the actual adjustment accuracy. During the determination process, a control efficiency matrix is established for each functional node, including the following four quantitative indicators: response time delay (ms), control action lag times, post-adjustment deviation amplitude (%), and re-control time window (s). If any two of these indicators continuously exceed the set threshold (for example, response delay greater than 500ms, control action repetitions more than three times), the node is marked as a control efficiency degradation unit. Based on the distribution of control efficiency degradation nodes and combined with material flow logic, the actual operating condition deviation distribution map of the entire non-dairy creamer processing process is analyzed. For example, if the shear rate deviation fluctuation rate of the emulsification control unit exceeds 20%, and the output moisture content deviation of the drying unit continuously exceeds 3%, the operating condition deviation is marked as an "emulsification-drying combined abnormal zone." Based on the aforementioned operating condition deviation distribution diagram, quality inspection data for the product (non-dairy creamer granules) at the discharge end, including particle size distribution, dry-to-wet ratio, color, and cohesive strength, is collected. Deviation comparison analysis is conducted against historical quality inspection standards. If abnormal particle size concentration, dry-to-wet ratio deviation exceeding 5%, or deterioration in cohesiveness are detected, the product quality deviation is confirmed. The operating condition deviation data is then reverse-indexed to the corresponding control efficiency reduction module, completing the data chain.

[0015] Step S4: based on the abnormal deviation of non-dairy creamer product quality, the defect position of the non-dairy creamer production line is traced to obtain the production line defect position data; based on the production line defect position data, the non-dairy creamer processing production line is optimized and controlled to obtain the non-dairy creamer processing production line optimization control data.

[0016] In this embodiment of the present invention, based on the product quality deviation status output in step S3 and its corresponding operating condition deviation node, a reverse tracing path is established for all identified control attenuation modules. By performing a time-series mapping between module execution logs, actual processing signals, and simulation model structures, the specific defect location data causing the product anomaly in the non-dairy creamer processing line is determined. During the defect location tracing process, a time-series comparison analysis method is used, using the product anomaly detection time as a reference point, to search forward to the control response log of the corresponding functional unit. This records indicators such as equipment operating status, electrical signal fluctuations, and abnormal temperature / pressure changes during the abnormal response period. For example, in the case of insufficient non-dairy creamer cohesion, it was found that the nozzle pressure in the spray drying zone fluctuated by more than 20% over multiple sampling cycles during the corresponding time period, and the air duct heating did not reach the target temperature. The spray drying unit was then identified as the defective node. Based on this defect location data, the parameter control module in the embedded control system is invoked to directly optimize the operating parameters of the defective module. This includes, but is not limited to, adjusting the nozzle angle, correcting the supply pressure curve, recalibrating the hot air temperature controller PID control coefficient, and limiting the shear rate fluctuation frequency. All optimization instructions are then written to the internal task queue of the digital controller. The structural model of the non-dairy creamer processing production line is synchronously updated in the digital visualization control platform, the defective modules are marked as red warning areas, and the optimized parameters are bound to the control chart nodes in real time, forming a simulation-reality parameter linkage view output including: defective module index, defect cause analysis, optimized parameter set and digital control task table, realizing closed-loop optimization control of the non-dairy creamer processing production line.

[0017] Preferably, step S1 includes the following steps: Step S11: Acquire food processing production line design data; In an embodiment of the present invention, a complete set of process design drawings and technical data generated during the construction phase of a food processing plant is retrieved, including but not limited to original design documents such as the "Equipment Layout Drawing," the "Process Flowchart," the "Automation Control Logic Diagram," the "Equipment Parameter Manual," and the "Conveyor System Structure Specification." A structured data conversion platform is used to perform structural interpretation and field standardization. A CAD drawing data parsing tool is used to deconstruct the AutoCAD format process layout drawings and cross-sectional drawings, extracting spatial geometric parameters such as the conveyor belt start and end nodes, processing equipment installation location information, and auxiliary structure layout information. A control logic parsing engine is used to extract control points, logical associations, start / stop signals, and chain sequence data from PLC control logic description files (.lad / .xml format). An equipment list data interface program is used to extract parameters such as equipment name, rated power, processing capacity, operating pressure, and equipment classification number from Excel or a database. All analyzed data are uniformly stored in accordance with the four major structures of "design main equipment table, power system table, control point table, and spatial coordinate table", and numbered using a unified engineering data identification and coding system to achieve structured acquisition of food processing production line design data. The output design data includes equipment layout coordinate set, process node sequence set, logical chain structure matrix, and power and energy consumption parameter list.

[0018] Step S12: collecting internal processing structure data of the non-dairy creamer production line based on the food processing production line design data; In this embodiment of the present invention, using the food processing line design data obtained in step S11 as a reference, a 3D scanning device (using a FARO Focus S70 laser scanner) is deployed on-site at the actual non-dairy creamer production line to perform a 360-degree scan of the processing site, generating 3D point cloud data of the equipment structure. A real-time image acquisition unit (Basler industrial camera + GigE interface high-speed acquisition card) is used to sample the equipment's operating status at key process stages. A wireless RFID device identification module is used to read device tags installed on each processing unit, including information such as device ID, operating hours, maintenance records, and control status. Sensor data from each processing unit, such as pressure, temperature, speed, flow rate, and liquid level, is read in real time via an industrial control bus (e.g., Modbus TCP / IP). This data is standardized and encapsulated using the OPC UA protocol and uniformly mapped to the corresponding device logical structure. A structure mapping engine is then invoked to align and match the 3D structure point cloud with the spatial coordinate table in the design drawings, establishing a one-to-one correspondence between the physical structure of the equipment and the designed structure. The collected internal processing structure data are divided into structural modules such as high-temperature spray drying unit, emulsification and stirring unit, liquid-feeding unit, gas supply system, and material premixing system according to the functional units of the equipment. A structural parameter library is constructed for each module, including processing paths, process levels, pipeline layouts, equipment coupling points, etc. The output internal processing structure data includes the spatial position of each device, processing interface relationships, control point parameters, and structural parameter matrix, which serve as the basic data support for subsequent analysis and simulation.

[0019] Step S13: determining the processing flow status of the non-dairy creamer production line based on the food processing production line design data and the internal processing structure data of the non-dairy creamer production line; In an embodiment of the present invention, an event-driven data flow path identification mechanism is constructed, and the "logical chain structure matrix" in the design data obtained in step S11 is matched item by item with the "structural parameter matrix" obtained in step S12. A processing path backtracking algorithm is used to calculate the flow direction of the entire process from the raw material input point to the finished product packaging point, thereby identifying the complete material flow path and process function chain of the non-dairy creamer production line. Specifically, this includes identifying the processing units of the entire process of "raw material storage - premixing - emulsification - spray drying - cooling - metering and packaging", and analyzing the connection relationships between each unit, such as the conveyor chain, heat exchange chain, pressure chain, and flow coupling chain. A processing node sequence restoration system is used to analyze the startup sequence, closed-loop feedback relationship, and control cycle of each node. Combined with the equipment startup log and process parameter curve under actual operating conditions, bottleneck sections in the processing process (such as frequent temperature fluctuations in the spray drying temperature control zone and unstable flow in the emulsification homogenization process) are identified, and then divided into "flow smoothing area, flow hysteresis area, and abnormal fluctuation area." At the same time, a delay evaluation method based on the transmission rate of physical quantities is introduced to calculate the response lag time of each process path, so as to obtain the dynamic working condition feature set in each processing path and output the processing process status data of the non-dairy creamer production line, including the time response function of each process, process switching nodes, abnormal jitter points and flow critical points and other parameters, providing a processing process logic basis for the subsequent simulation structure construction.

[0020] Step S14: constructing a simulation model of the non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line and the processing flow status of the non-dairy creamer production line.

[0021] In this embodiment of the present invention, the non-dairy creamer production line structural parameter matrix obtained in step S12 and the process flow status data obtained in step S13 are used as dual inputs to dynamically model the entire production line process using a modular simulation modeling tool (such as Tecnomatix PlantSimulation or FlexSim). Within the simulation platform, a process structure layout corresponding to the actual site is restored based on the spatial coordinate relationships of the equipment. Each processing unit is encapsulated as an independent simulation module, and the unit's physical parameters, processing capacity, response time, energy consumption level, coupling interface type, and other parameters are input. Subsequently, based on the process flow status data, the material flow relationship and control logic sequence between the simulation modules are defined, and the transmission path, processing cycle, feedback control closed-loop conditions, and interlocking start-stop rules are set. After establishing the connection structure, a time-series model is constructed for the entire non-dairy creamer production line, incorporating a standard process cycle as the timeline benchmark. The material residence time and conversion efficiency in each process section are set segment by segment, and fluctuation boundary conditions and disturbance threshold parameters are set based on the fluctuation amplitude in historical operating data. Combined with the historical trajectory of real-time sensor data, the flow rate mutation, thermal pressure anomaly, uneven flow and other conditions existing in the actual operation process are loaded as external disturbance sources. The simulation platform records the disturbance response timing, processing rhythm changes and material lag trends, and completes the coupling relationship mapping of the entire processing link. The output simulation model of the non-dairy creamer processing production line is saved in the simulation platform in the form of a module, and outputs simulation result data such as the structural configuration list, response time matrix, material flow map, disturbance response diagram, rhythm synchronization chart, etc., which provide a technical input basis for the instability identification and pressure response evaluation in the subsequent step S2.

[0022] Preferably, step S14 includes the following steps: Step S141: using the internal processing structure data of the non-dairy creamer production line to determine the internal physical layout of the non-dairy creamer production line with a point cloud accuracy of ±2mm; In an embodiment of the present invention, this step is based on the internal processing structure data of the non-dairy creamer production line obtained in step S12, and calls an industrial-grade three-dimensional laser scanner (such as FARO Focus S70 or Leica BLK360) to perform high-precision point cloud scanning on all key processing equipment and auxiliary facilities in the non-dairy creamer production line. The scanning process adopts a regional and layered scanning strategy to collect point cloud data for the spray drying unit, emulsification and stirring unit, conveying device and circulating air duct, etc., to ensure overall coverage without omissions. The scanner uses high-frequency laser pulse emission technology, combined with a multi-angle and multi-viewpoint scanning solution, and after fusing multiple sets of point cloud data, it uses a point cloud registration algorithm to achieve data alignment, and the error is controlled within the range of ±2 mm. Subsequently, the collected original point cloud is subjected to noise reduction, filtering and edge sharpening processing using point cloud data processing software (such as Geomagic Design X) to eliminate isolated points and noise points generated in the scanning blind area. After processing, the point cloud data is voxelized and subdivided to construct a physical space entity model in the three-dimensional digital twin, completing the precise reconstruction of the internal spatial geometric structure of the production line. The obtained internal physical layout data of the non-dairy creamer production line includes the specific three-dimensional coordinates, relative spatial relationships and dimensional parameters of the equipment, with an accuracy of ±2mm, which serves as the basic input for subsequent nozzle distribution and air duct structure identification.

[0023] Step S142: collecting processing structure nozzle position distribution data using the internal physical layout of the non-dairy creamer production line; In this embodiment of the present invention, step S142 collects specialized data on the printhead distribution based on the high-precision internal physical layout determined in step S141. Using a geometric feature recognition algorithm from the laser scanning point cloud, the printhead locations are identified through point cloud slicing analysis and shape fitting techniques. Surface reconstruction and local feature matching methods are used to locate the printhead mounting holes, nozzle shape, and spray direction. Combining high-resolution static images captured by an industrial camera, image registration techniques are used to overlay visual information with point cloud data, enabling precise extraction of the printhead surface structure and position. By precisely measuring the three-dimensional coordinates of the printhead center point, a printhead location distribution list is generated, including the printhead number, spatial coordinates (X, Y, Z), and spray direction vector. Throughout the acquisition process, the printhead location data maintains a consistent coordinate system with the physical layout data from step S141, achieving seamless spatial alignment. The resulting printhead location distribution dataset details the specific layout and orientation parameters of each printhead on the production line, providing a key basis for subsequent airflow analysis and spray velocity setting.

[0024] Step S143: Identifying the geometric structural features of the internal circulation air duct of the production line based on the internal physical layout of the non-dairy creamer production line; In an embodiment of the present invention, step S143 uses the internal physical layout data obtained in step S141 to focus on identifying the geometric structural features of the circulating air duct in the production line. By performing cross-section extraction and curve fitting on the pipeline structure in the point cloud data, a topological analysis method is used to determine the connectivity, branch nodes and pipe diameter changes of the air duct. Combined with the technical data of the air duct material and installation method, the pipeline curvature analysis algorithm is used to calculate the bending radius, length and node angle of the air duct to achieve a three-dimensional geometric representation of the spatial form of the air duct. Using CFD preprocessing technology, the air duct cross section is divided into multiple sections, the key cross-sectional area and ventilation section are extracted, and the equivalent flow area and resistance coefficient of the air duct are calculated. The geometric feature data of the air duct includes the sequence of pipeline coordinate points, cross-sectional shape parameters, branch topological structure and coordinates of the connection nodes, which serve as the geometric basis for air flow simulation. This step ensures that the complex spatial structure of the air duct is accurately quantified, which is convenient for subsequent wind speed setting and air flow direction judgment.

[0025] Step S144: using the geometric structure characteristics of the internal circulation air duct of the production line and the processing structure nozzle position distribution data to set the injection speed to 2.5m / s and determine the airflow direction inside the production line; In an embodiment of the present invention, step S144 sets the nozzle injection speed to a constant 2.5 meters per second based on the geometric structure characteristics of the air duct in step S143 and the nozzle position distribution data in step S142, spatially matches the nozzle position with the air duct node, and determines the air duct outlet or branch point corresponding to the installation position of each nozzle. A fluid dynamics calculation method based on the geometric parameters of the air duct is used, combined with the injection speed setting, to calculate the initial flow velocity vector and its direction after the air flow is ejected from the nozzle. By analyzing the topological structure and pipe diameter changes of the air duct, the continuity equation is applied to correct the air flow rate to ensure that the injection speed matches the cross-sectional area of the pipe to avoid local overflow or backflow. The node connection relationship is used to deduce the mainstream direction of the air flow in the production line, clarify the airflow circulation path in the air duct, including the coordinated flow relationship between the nozzle injection airflow and the circulating air duct. The output airflow flow direction map contains the flow velocity direction vector of each air duct section and the injection airflow distribution parameters corresponding to the nozzle, providing an airflow dynamics basis for the stability evaluation of the processing flow.

[0026] Step S145: evaluating the processing stability of the non-dairy creamer production line based on the airflow direction within the production line to obtain production line processing stability data; In an embodiment of the present invention, step S145 couples and analyzes the airflow dynamics characteristics with the processing flow status data obtained in step S13 based on the airflow flow direction inside the production line determined in step S144. Multi-parameter fusion technology is used to synchronously monitor and compare the key process parameters and airflow direction of each processing node of the production line. The degree of influence on the material conveying stability and spray uniformity of the aerodynamic disturbance caused by airflow is calculated. A mass transfer and heat transfer analysis method based on physical mechanisms is introduced, and the influence of airflow on the suspension time and coagulation time of non-dairy creamer particles is derived in combination with airflow speed and direction. By comparing time series data, unstable points in the processing link caused by airflow fluctuations are identified, such as a sudden drop in temperature in the spray cooling area or a local airflow short circuit in the emulsification and stirring link. The production line processing stability data is output, including the stability index, disturbance frequency, impact range and corresponding physical parameter change curve of each key node.

[0027] Step S146: Evaluate the stability of the finished product of the non-dairy creamer production line based on the production line processing stability data and the processing flow status of the non-dairy creamer production line; In an embodiment of the present invention, step S146 comprehensively evaluates the stability of the finished product by combining the production line processing stability data obtained in step S145 with the processing flow status data of step S13. Statistical analysis methods are used to calculate the degree of influence of processing parameter fluctuations on product quality indicators, including particle size distribution, moisture content, density uniformity and packaging integrity. In view of the stability fluctuations in key process flows, the standard deviation and deviation rate of the finished product performance indicators are calculated to form a finished product stability evaluation report. The report covers a list of stability influencing factors, an analysis of the contribution rate of process nodes to finished products, and prediction of potential quality hazards to output the stability parameters of the non-dairy creamer production line finished product, including finished product batch consistency indicators, the range of variation of key quality characteristics and their corresponding process stability-related parameters, providing a quantitative basis for production control and optimization.

[0028] Step S147: Construct a simulation model of the non-dairy creamer processing production line based on the stability of the finished product of the non-dairy creamer production line and the geometric structure characteristics of the internal circulation air duct of the production line.

[0029] In an embodiment of the present invention, step S147 uses the finished product stability data obtained in step S146 and the geometric structure characteristics of the circulating air duct in step S143 as a basis to construct a simulation model of a non-dairy creamer processing production line with an airflow-process coupling function, reconstructs the air duct geometry in the simulation environment, and uses it as the spatial boundary of the airflow simulation module, and inputs the air duct diameter, curvature, branch nodes and flow area parameters. Secondly, the finished product stability index is loaded as a dynamic constraint condition of the processing technology module to define the output quality fluctuation range of the processing unit and the corresponding process response time. Through the multi-physical field coupling calculation method, the airflow dynamics model is combined with the processing flow model to simulate the impact of airflow disturbance on the process stability and finished product quality. The simulation model calculates the airflow velocity distribution, temperature and humidity changes, and material state evolution process in real time, and outputs the key stability parameters in the process and the finished product quality prediction data to form a simulation model to realize the digital coupling of the airflow environment inside the production line and the processing technology state, and provide comprehensive simulation support and data interface for the digital visualization control platform.

[0030] Preferably, determining the machining disturbance coupling instability characteristics in step S2 includes: According to the non-dairy creamer processing line simulation model, the non-dairy creamer processing production process is simulated to obtain the non-dairy creamer processing production process simulation data; In an embodiment of the present invention, the non-dairy creamer processing production process is simulated based on the non-dairy creamer processing production line simulation model to obtain the non-dairy creamer processing production process simulation data. In the specific operation, the simulation model constructed in step S147 is used. The model is constructed based on the geometric structure characteristics of the internal circulation air duct of the production line and the nozzle position distribution data, covering key process parameters such as air flow, heat transfer and material injection in the production line. The overall operating environment of the production line is simulated by numerical simulation software, and the preset air flow velocity (2.5m / s), nozzle injection parameters and process setting conditions are input to generate time series data and spatial distribution data covering each process node. These simulation data are specifically manifested in air temperature distribution, air flow velocity field, spray particle size distribution, pressure fluctuation and local thermal environment parameters, which provide basic data support for the subsequent identification of abnormal characteristics.

[0031] Identify abnormalities in the hot air outlet guide cavity position based on the simulated data of the non-dairy creamer processing production process; In an embodiment of the present invention, abnormal conditions in the position of the hot air outlet guide cavity are identified based on simulation data of the non-dairy creamer processing production process. By analyzing the airflow pattern and temperature distribution near the hot air outlet in the simulation results, high-resolution spatial data analysis technology is used to focus on monitoring the airflow velocity, flow direction deviation and temperature gradient changes in the guide cavity. The data analysis method is used to locate the offset of the airflow trajectory and the local temperature anomaly points, and combined with the preset standard guide cavity design size and position, it is determined whether there is position offset or deformation in the guide cavity. By comparing the deviation between the actual airflow path and the ideal design path in the simulation, the abnormal position parameters of the hot air outlet guide cavity are obtained, and abnormal condition identification data is formed for subsequent production line status judgment.

[0032] Detect the lack of symmetry of the internal air duct based on the abnormal position of the hot air outlet guide cavity; In an embodiment of the present invention, the lack of symmetry in the internal air duct is detected based on the abnormal position of the hot air outlet guide cavity. Using the identified guide cavity abnormality data as input, combined with the geometric layout of the internal circulation air duct of the production line, a symmetry analysis is performed on the spatial distribution of the air flow velocity field and pressure field. Using symmetry quantitative indicators, the flow velocity difference, pressure difference and temperature distribution difference of each symmetrical part of the air duct are calculated to identify whether the air duct structure has a lack of symmetry. The degree of symmetry loss is determined based on the critical value of the numerical difference, and the air duct symmetry loss parameter is output to form the system symmetry state data, which provides a basis for subsequent local eddy current analysis.

[0033] Detect the growth of local vortex based on the lack of symmetry of the internal air duct and the abnormal position of the hot air outlet guide cavity; In an embodiment of the present invention, the internal local vortex growth condition is detected based on the lack of symmetry in the internal air duct and the abnormal position of the hot air outlet guide cavity. The data on the lack of symmetry and the abnormality of the guide cavity are input into the fluid dynamics analysis tool, focusing on the vortex interval inside the air duct. The dynamic changes of the duct vortex are analyzed by calculating the local vortex intensity, vortex scale and vortex energy accumulation. The velocity vector field and curl field analysis methods are used to capture the spatial distribution and time evolution trend of the vortex. The degree of vortex growth is judged according to the vortex energy increase, and the local vortex growth parameter is obtained. This parameter reflects the unstable growth of the airflow disturbance and the kinetic energy of the circulating wind, indicating the non-stable state of the flow field inside the production line.

[0034] Determine the temperature retention area inside the production line based on the internal local eddy current growth conditions; In an embodiment of the present invention, the temperature retention area inside the production line is determined based on the internal local eddy current growth conditions. The local eddy current growth data is used to identify the area of stagnant or slow airflow in the air duct. Combined with the temperature field data, the heat accumulation area caused by the difficulty of effective airflow circulation is located. Through three-dimensional thermal imaging data and numerical heat transfer analysis, the spatial location where the temperature abnormality continues to rise is determined and calibrated as temperature retention. Due to the poor airflow in this area, heat accumulates, forming an abnormal temperature gradient area. The temperature retention space coordinates and its temperature value range are output as key thermal environment indicators.

[0035] Evaluate the degree of angle deviation of the processing nozzle based on the simulation data of the non-dairy creamer processing production process; In an embodiment of the present invention, the degree of angle deviation of the processing nozzle is evaluated based on the simulation data of the non-dairy creamer processing production process. In combination with the nozzle position distribution and the simulated airflow data, the deviation angle between the actual nozzle spray direction and the designed standard spray direction is analyzed. By accurately measuring the physical installation angle of the nozzle, combined with the changes in the spray particle size distribution and the spray path offset in the simulation, the angle deviation value is quantitatively calculated. The nozzle angle is calibrated using a high-precision angle measuring instrument and three-dimensional scanning technology. The data is combined with the simulation deviation to obtain the processing nozzle angle deviation parameter, which provides a basis for adjusting the nozzle performance.

[0036] The nozzle hot air interference growth trend is predicted based on the degree of angle deviation of the processing nozzle in the temperature retention area of the production line, and the nozzle hot air interference data is obtained; In an embodiment of the present invention, the growth trend of nozzle hot air interference is predicted based on the degree of angle deviation of the processing nozzle in the temperature retention area within the production line, and nozzle hot air interference data is obtained. The degree of mutual interference between the nozzle hot air and the airflow is analyzed based on the temperature retention position and angle deviation parameters. The diffusion range and intensity of the nozzle hot air disturbance are calculated using aerodynamic methods, and the growth trend of the interference over time is predicted by combining historical data and dynamic simulation. A curve is obtained showing the change of nozzle hot air interference intensity as the process operation time, forming hot air interference growth trend data for subsequent prediction of nozzle particle size distribution changes.

[0037] Predict the growth trend of nozzle particle size distribution stratification based on nozzle hot air interference data; In this embodiment of the present invention, the stratification growth trend of the nozzle particle size distribution is predicted based on nozzle hot air interference data. Based on this nozzle hot air interference data, the impact of hot air on the distribution of the sprayed particle size is analyzed, focusing on the stratification and deviation of the particle size distribution. A particle size analyzer is used to measure the actual sprayed particle size, and the particle size stratification effect caused by hot air interference is estimated using numerical simulation data. The particle size mean, standard deviation, and distribution skewness are calculated to form the nozzle particle size distribution stratification growth trend parameter, which reflects the dynamic trend of nozzle spray quality as the interference changes.

[0038] The processing disturbance coupling instability characteristics are determined based on the stratified growth trend of the nozzle particle size distribution and the temperature stagnation area inside the production line.

[0039] In an embodiment of the present invention, the characteristics of process disturbance coupling instability are determined based on the stratified growth trend of the nozzle particle size distribution and the temperature retention area within the production line. By combining particle size distribution parameters and temperature retention space information, the coupling relationship between the spray particle size change and the local thermal environment is analyzed. Using a disturbance coupling analysis method, the impact of the mutual influence of the two on the stability of the processing process is evaluated to determine the characteristic parameters of the process disturbance coupling instability, reflecting the overall unstable state caused by the interaction of disturbances within the system. This characteristic is used to guide subsequent production line control and optimization plan development to ensure process stability.

[0040] Preferably, the detection of the dramatic increase trend of the pressure response of the production line in step S2 includes: The trend of increasing local disturbance deviation is predicted by using the instability characteristics of machining disturbance coupling; In an embodiment of the present invention, based on the obtained processing disturbance coupling instability characteristic parameters, the local disturbance characteristics in the current operating state of the production line are quantitatively analyzed. Using multidimensional time series data analysis technology, the historical change data of the disturbance coupling instability characteristics are used as input, and the growth rate and acceleration of each local disturbance factor are calculated through frequency domain analysis and trend inference methods. By segmenting the time series of the instability characteristic parameters, the intensified stage of the disturbance deviation change is identified. The slope calculation method and autocorrelation function in statistics are used to determine the growth trend of the local disturbance deviation to obtain trend curve data reflecting the change of the local disturbance deviation over time, providing a basis for the dynamic change of the local disturbance for subsequent steps.

[0041] Estimate the increase in particle size distribution differences based on the increasing trend of local disturbance deviations; In an embodiment of the present invention, based on the obtained local disturbance deviation increase trend data, combined with the production line nozzle particle size distribution monitoring data, the particle size distribution change analysis is performed. By calculating the particle size mean, variance and skewness changes of the nozzle sprayed particles, the diffusion and stratification degree of the particle size distribution are quantified. Real-time spray data is collected using particle size measurement equipment, and the local disturbance deviation data is compared based on the statistical characteristics of the particle size distribution to evaluate the impact of increased disturbance on the particle size distribution difference. By establishing a corresponding relationship between the deviation and the change in particle size distribution, the degree of increase in the particle size distribution difference within a certain period of time in the future is calculated to form a particle size distribution difference increase trend parameter.

[0042] Estimate the local retention of large particles in the production line based on the increasing trend of particle size distribution differences; In an embodiment of the present invention, the obtained data on the increasing trend of the particle size distribution difference is used as input to conduct a joint analysis of the material transportation and airflow state within the production line. Combined with the inertial force and deposition behavior of the material particles with increased particle size in the air flow, the retention probability and the distribution of the retention area of the large-sized particles in the air duct of the production line are calculated. The air flow velocity and direction are collected in real time using three-dimensional flow field monitoring equipment, and the particle deposition and aggregation are detected in conjunction with optical particle image analysis technology. By comparing the particle distribution images and flow field data at different time points, the severity of the local retention of large-sized particles in the production line is estimated to form a quantitative parameter for local retention.

[0043] Determine the flow blockage situation inside the production line based on the local retention of large particles in the production line; In an embodiment of the present invention, the degree of flow blockage within the production line is assessed based on acquired local retention parameters combined with measured airflow pressure drop data within the duct. High-precision pressure sensors are used to collect real-time pressure data at key locations in the duct, analyzing abnormal pressure fluctuations and pressure differential changes to determine the location and intensity of the blockage. The reduction in the effective flow area is calculated based on the particle accumulation state within the local retention area, thereby estimating the increase in fluid flow resistance. The severity of the internal flow blockage is determined based on the correlation between the pressure drop change and the resistance increase, resulting in a flow blockage state parameter.

[0044] Detect the growth degree of suction negative pressure deviation based on the flow blockage situation inside the production line; In an embodiment of the present invention, the negative pressure status of the suction system within the production line is evaluated in combination with flow blockage state parameters. Negative pressure sensors are arranged in the suction pipeline and key nodes to monitor changes in negative pressure values in real time. By comparing the normal operation negative pressure baseline data, the suction negative pressure deviation is calculated, and the trend of abnormal negative pressure growth is analyzed in combination with the blockage location data. Using data fusion technology, the negative pressure data collected by multiple sensors are spatially integrated and temporally smoothed to accurately identify the degree of negative pressure deviation growth and output the internal structure suction negative pressure deviation growth data.

[0045] Calculate the internal suction demand growth data based on the internal flow blockage of the production line and the growth degree of suction negative pressure deviation; In this embodiment of the present invention, flow blockage data and suction negative pressure deviation data are used as inputs to calculate the increase in suction demand required to maintain normal airflow circulation on the production line based on the laws of fluid mechanics. Energy conservation and flow balance calculation methods based on increased wind resistance and changes in negative pressure are employed, combined with real-time airflow monitoring data, to determine the additional energy required to increase suction under the current operating state. Energy metering instruments monitor fan power consumption in real time, and combined with flow resistance calculations, output internal suction demand growth data, reflecting the changing trend of the system's suction load.

[0046] Detect the degree of wind pressure structure power growth based on internal suction demand growth data; In this embodiment of the present invention, the overall power consumption of the wind pressure system is assessed based on the calculated suction demand growth data. A power meter and current sensor collect real-time fan operating power data, and the actual power increase is calculated based on the fan performance curve and operating point characteristics. This is compared with the baseline power value under normal operating conditions to determine the increase in wind pressure structure power. Combining the fan load characteristics and operating parameters, a data processing algorithm is used to filter out noise and anomalies, and a quantitative indicator of the degree of wind pressure structure power growth is output.

[0047] The dramatic increase trend of the production line's pressure response is detected based on the growth rate of the wind pressure structure power and the growth rate of the suction negative pressure deviation.

[0048] In an embodiment of the present invention, the wind pressure structure power growth data and the internal structure suction negative pressure deviation growth data are integrated, and a multi-parameter fusion analysis method is used to judge the overall pressure state of the production line. By synchronously monitoring the two sets of data, the changing pattern of the pressure system response is analyzed, and the growth rate of the pressure response is determined using time series cross-analysis technology. By setting multi-level thresholds, it is possible to identify whether the pressure response has entered a sharp increase stage, and output the production line pressure response sharp increase trend parameters as an important basis for subsequent maintenance and adjustment decisions.

[0049] Preferably, the abnormal deviation chain growth trend prediction in step S2 includes: Test the stress concentration trend of the internal structure of the processing according to the sharp increase trend of the pressure response of the production line; In an embodiment of the present invention, based on the data on the dramatic increase in the pressure response of the production line obtained in the previous step, real-time monitoring is performed using strain gauge sensors arranged at key stress-bearing locations of the production line. The strain gauge sensors reflect the surface strain of the structure through changes in resistance and are connected to a data acquisition unit to achieve high-frequency sampling. The strain data of each monitoring point is continuously recorded by the data acquisition system, and the stress value is converted using the stress-strain relationship. Combined with the material mechanical parameters preset by the finite element analysis (FEA), the spatial distribution mapping of the structural stress is achieved. The stress concentration area is dynamically monitored, and the stress concentration peak value and its time variation trend are calculated. Curve fitting technology is used to extract the growth rate of the stress concentration peak value, and generate time series data reflecting the stress concentration trend of the internal structure of the processing.

[0050] Detect the internal structure crack growth condition when the stress concentration trend of the internal structure during processing is greater than 50N; In an embodiment of the present invention, the obtained stress concentration peak data is used as input to screen out the time period with a peak value exceeding 50N, which corresponds to the stage when the structure is at risk of crack propagation. Ultrasonic non-destructive testing equipment is used to scan the monitoring area, and phased array ultrasonic technology is used to detect and locate micro cracks in the early stage of cracks. By analyzing the amplitude and time delay of the ultrasonic echo signal, the crack length, depth and crack propagation rate are identified. The results of multiple consecutive tests are compared and analyzed to calculate the crack growth rate and expansion trend. Combined with the stress concentration change curve, a dynamic monitoring report of the crack growth status is formed, including a crack size change curve and a position distribution map, which are output as quantitative data of the crack growth status.

[0051] Determine the internal structure aging trend based on the internal structure crack growth status and the stress concentration trend of the processed internal structure; In an embodiment of the present invention, the crack growth rate and stress concentration trend data are integrated, and a structural life prediction method is adopted to determine the internal structure aging process through fatigue damage accumulation calculation. The cumulative fatigue damage parameter is calculated based on the frequency distribution of the crack propagation rate and the stress peak. The crack growth amount is converted into a fatigue aging index of the structural material using Miners' law or equivalent cumulative damage theory. A time series analysis is performed on the aging index of each monitored area, and an aging trend curve is extracted. The aging trend curve reflects the fatigue degradation rate of the internal structural materials and the connection parts, forming an internal structure aging trend parameter, which is used as a reference benchmark for subsequent deformation measurements.

[0052] Measure the deformation degree of the internal structure of the production line according to the stress concentration trend of the internal structure during processing and the aging trend of the internal structure; In an embodiment of the present invention, the internal structural deformation of the production line is measured based on stress concentration trend and aging trend parameters. Laser displacement sensors are arranged at key nodes of the structure, and a non-contact high-precision ranging method is adopted to obtain the tiny deformation of the surface of the structure in real time. By comparing the coordinates of the reference point in the initial unstressed state with the current measurement data, the three-dimensional displacement change of the node is calculated, and then the overall structural deformation pattern is estimated. Combined with the deformation data of the stress concentration area, the difference analysis method is used to determine the degree of correlation between deformation and aging. Based on the time series deformation data, the deformation growth rate and amplitude are extracted to form a quantitative description of the degree of internal structural deformation, which serves as the input for the subsequent precision attenuation assessment.

[0053] Evaluate the internal precision degradation of the production line based on the internal structural deformation and aging trend of the production line; In an embodiment of the present invention, the overall processing accuracy attenuation of the production line is evaluated by combining the obtained data on the degree of structural deformation with the aging trend parameters. A geometric error transmission chain is established by utilizing the influence of structural deformation on the position accuracy of the processing part. The error accumulation theory is used to calculate the relative positioning error between the tool and the workpiece caused by the deformation and convert it into a percentage of reduced processing accuracy. Combined with the data on the degradation of the mechanical properties of the material with the aging trend, the rigidity coefficient of the processing part is adjusted to comprehensively evaluate the real-time attenuation value of the processing accuracy of the production line. The evaluation results are compared with the set accuracy threshold to generate a detailed report on the accuracy attenuation status, and the amplitude of the processing accuracy attenuation is quantitatively expressed as a basis for judging abnormal deviations.

[0054] When the precision attenuation inside the production line exceeds 68.2% and the degree of structural deformation inside the production line, the abnormal deviation from the chain growth trend is predicted.

[0055] In an embodiment of the present invention, a threshold judgment standard is set based on the obtained internal precision decay percentage and structural deformation amplitude. When the internal precision decay exceeds 68.2%, the prediction mechanism of the abnormal deviation chain growth trend is activated in combination with the current structural deformation data. The time correlation analysis of historical deviation data and current detection data is used to identify the chain conduction path of deviation growth. Through the differential and integral operations of continuous monitoring data, the cumulative increment and growth rate of the deviation value are calculated to form a deviation chain growth trend curve. This trend curve can quantitatively characterize the abnormal diffusion trend of the production line operation status, provide accurate abnormal deviation warning data for the digital visualization control system, and serve as a decision-making basis for subsequent operation adjustments.

[0056] Preferably, step S3 includes the following steps: Step S31: detecting the attenuation of processing control efficiency based on the abnormal deviation from the chain growth trend; In an embodiment of the present invention, the output abnormal deviation chain growth trend data is used as input, and the real-time operation data acquisition module of the production line control system is adopted to record the control instruction execution response time, sensor feedback delay and production rhythm change. Through the high-frequency data acquisition device installed in the key control unit, the actual response of the production line after the control instruction is issued is captured to form a control response timing signal. Combined with the abnormal deviation trend curve, the timing signal analysis method is used to calculate the correlation between the control instruction response delay and the output fluctuation amplitude of the production line. Through the frequency domain analysis technology, the efficiency drop of the control system under abnormal deviation conditions is extracted to form the control efficiency attenuation value of the processing production line. This value is expressed as a percentage, reflecting the actual control ability of the current control system under the deviation abnormal state, as the quantitative output of this step.

[0057] Step S32: detecting abnormality in processing uniformity of the production line according to the processing control efficiency attenuation condition; In an embodiment of the present invention, the percentage of control efficiency attenuation obtained in step S31 is used as a reference, combined with the spatial distribution data of key parameters in the processing process of the non-dairy creamer production line, such as product thickness, powder distribution density and composition uniformity. An online imaging detection system is used to perform high-resolution scanning of continuously processed products on the production line, and collect surface and cross-sectional images of the product. Through image processing algorithms, key uniformity indicators are extracted, such as thickness deviation standard deviation, particle aggregation distribution, etc. A time series correlation analysis is performed on the product uniformity index and the control efficiency data, and the uniformity anomaly index is calculated. The index is derived based on the deviation of the uniformity index and the control efficiency attenuation, and quantitatively describes the degree of abnormality in the processing uniformity of the production line. The anomaly index is output as a quantitative parameter of the abnormal uniformity of the production line to provide data support for subsequent working condition deviation analysis.

[0058] Step S33: determining the deviation of the production line working condition according to the abnormality of the production line processing uniformity and the attenuation of the processing control efficiency; In an embodiment of the present invention, the processing uniformity abnormality index in step S32 and the control efficiency attenuation percentage in step S31 are used as inputs at the same time, and a multi-parameter fusion method is used to perform working condition deviation analysis. By establishing a cross-analysis matrix of processing uniformity and control efficiency, the synergistic influence relationship between the two is determined. Based on the cross matrix, the statistical inference method is applied to classify and grade the working condition deviations, and the working condition deviation index is calculated. This index reflects the degree of deviation of the current operating state of the production line compared to the standard working condition. The specific value is obtained through statistical analysis and presented as a percentage range. The working condition deviation index combines historical operating data and equipment parameters to further refine the deviation type, such as equipment response hysteresis, material distribution abnormality, etc. The detailed numerical output of the working condition deviation is the result of this step, which serves as the basis for working condition regulation and optimization.

[0059] Step S34: determining the abnormal deviation of the non-dairy creamer product quality based on the production line working condition deviation and the abnormal processing uniformity of the production line.

[0060] In an embodiment of the present invention, the operating condition deviation index output from step S33 and the uniformity abnormality index output from step S32 are used to evaluate the abnormal deviation of the non-dairy creamer product quality. The operating condition deviation is compared with the product quality inspection data (including moisture content, particle size distribution, viscosity and color uniformity of the finished product) through the correlation analysis method. The quality inspection data comes from the online quality inspection device at the back end of the production line and is collected through a spectrometer and image recognition technology. The operating condition deviation and uniformity abnormality data are mapped to the variation range of the quality index using the deviation transfer function. The abnormal deviation amplitude of each quality index is calculated, and a comprehensive product quality abnormal deviation index is generated. The index is expressed in the form of a percentage, clearly indicating the fluctuation and abnormal range of the non-dairy creamer product quality. The quality abnormal deviation status data provides important parameters for product quality monitoring and feedback adjustment.

[0061] It is particularly important that step S31 includes the following steps: Step S311: Detecting the growth of mechanical wear of the production line based on the abnormal deviation from the chain growth trend; In this embodiment of the present invention, after detecting an abnormal deviation trend, the wear growth trend of mechanical friction components is determined by combining the rate of change in the curve slope (the rate of increase in power fluctuation amplitude per unit time), the rate of increase in contact heat (measured by a Fluke TiX580 thermal imager), and the frequency of intermittent impact noise (collected and analyzed in real time by a microphone array). If the rate of change in the wear-related signal exceeds twice the historical growth trend for three consecutive observation cycles (for example, if the vibration amplitude slope changes from 0.5 to 1.2), the system records it as a "wear growth state" and simultaneously generates a wear growth rate value. Using micron-level component displacement growth (monitored by a Keyence LK-G5000 series laser displacement sensor) as an indicator, this data packet forms the mechanical wear growth status data for invocation in step S312.

[0062] Step S312: estimating the growth probability of equipment failure based on the growth of mechanical wear on the production line; In this embodiment of the present invention, a set of pre-failure wear parameters from historically similar operating conditions (stored in a local fault database) is used for a one-by-one comparison to determine the average failure rate corresponding to the current state at a given point in time. With a 15-minute update cycle, the system generates a "failure probability growth curve" that reflects the current wear growth trend of the equipment. The slope of the curve represents the "failure probability growth value" (in % / min). This value serves as input to the next step and, along with the wear growth data for production line machinery, contributes to material consumption forecasting.

[0063] Step S313: predicting the growth of production line material consumption based on the growth of equipment failure probability and the growth of production line mechanical wear; In this embodiment of the present invention, the failure probability increase value (% / min) and wear growth status data are directly provided by the first two steps and merged into the "Equipment Operation Influencing Parameters" table on the control platform. This table is transmitted in real time to the material control unit for dynamic adjustment of material supply strategies. In food processing lines, problems such as decreased conveying pressure, unbalanced flow rates, or shifted mixing ratios due to equipment wear directly manifest as an increase in raw material required per unit of finished product. By comparing the material input per unit of product (in L / min and kg / h) for failure rates <1% and >5%, a mapping relationship for material consumption growth rates is extracted. For example, under normal conditions, a unit fill volume of 300 ml corresponds to a syrup flow rate of 60 L / h. When the failure probability increases to 7% and the wear rate of the mixer reaches 6 μm / h, the system detects that the current unit fill liquid fluctuation range has exceeded ±10%. The material compensation amount is increased to 66 L / h, resulting in a consumption growth rate of 10%. This generates a data set for "Production Line Material Consumption Growth Status."

[0064] Step S314: Analyze the degree of equipment adjustment lag based on the growth of production line material consumption and the growth of equipment failure probability; In this embodiment of the present invention, the degree of control response lag is determined by analyzing the difference in execution response time after the control system issues an adjustment command, as well as the actual feedback lag on the adjustment action's effect on material consumption control. During operation, the "Execution Response Logging Module" in the control platform is invoked to compare the start time of the action and the feedback stabilization time of the filling system's control actuators (such as control pump valves and mixed flow control valves) after a variable frequency adjustment command is issued. For pneumatic control valves, for example, the delay between command issuance and response completion is 135ms. If, combined with material consumption growth rate data, fluctuations exceeding ±10% (exceeding the normal 5% tolerance) persist after adjustment, control lag is determined. Furthermore, combined with the increased probability of device failure, the system determines whether the lag is due to mechanical action delay or sensor response distortion. Through the sensor status verification process, if the flow sensor calibration data deviates from real-time data by more than 2%, and the corresponding increased probability of failure exceeds 10%, it is inferred that mechanical control distortion is causing the lag. The conclusion will be output in the form of "device adjustment lag data", including key indicators such as response time difference, adjustment impact window delay length and post-adjustment effect offset, providing a basis for the next step of control efficiency judgment.

[0065] Step S315: Detecting the attenuation of processing control efficiency based on the degree of equipment adjustment lag and the increase in production line material consumption.

[0066] In this embodiment of the present invention, control efficiency is evaluated based on three dimensions: command response time, material ratio correction rate, and product parameter recovery time. The degree of control lag and material consumption growth serve as direct input parameters. In the control center, the "Process Correction Behavior Recording Module" is called to obtain the average time required for production indicators (such as product viscosity, pH, and solids content) to return to a stable state after response in the past five control actions. For example, for a food colloid production line, the original viscosity was set at 950 cP. Due to control lag, it fluctuated to 980 cP. After adjustment, it took 6 minutes to recover to the range of 950 ± 5 cP, far exceeding the system's expected standard control time limit of 2 minutes. If this recovery time exceeds 3 minutes for five consecutive records and the material compensation amount increases by more than 8%, it is determined that the control efficiency has declined. Analyze the overall energy consumption growth trend caused by efficiency degradation, the number of increases in adjustment frequency (times / h), and the number of manual interventions (recorded in the HMI log) to form a data packet of control efficiency degradation status, which includes quantitative indicators such as the adjustment response timeout ratio, the energy consumption coefficient change rate, and the process adjustment failure ratio. This data packet serves as the input basis for step S32 and is used to predict the degree of aggravation of production fluctuations.

[0067] It is particularly important that step S32 includes the following steps: Step S321: predicting the degree of aggravation of fluctuations in the production process based on the attenuation of processing control efficiency; In this embodiment of the present invention, a sequence of control efficiency decay values is extracted. This sequence consists of two types of data: the first is the time interval between the issuance of a control command and the corresponding execution response (i.e., the response time series); the second is the degree of return to production line status after the completion of a unit control action (e.g., temperature stabilization, motion rhythm recovery, etc.), which serves as the efficiency recovery ratio. Specifically, a response time acquisition module is installed in the core control section of the production line to record the time each control signal T1 is sent and the time the corresponding actuator begins action T2, calculating the response delay ΔT = T2 - T1. Simultaneously, vibration sensors, pressure sensors, and a temperature and humidity combined sensor are used to quantitatively score the stability of the system status within three seconds before and after the control action, generating a control effect ratio ε. The decay trend is statistically analyzed. If the control efficiency decreases by more than a set threshold of 0.6 for multiple consecutive cycles, and the vibration amplitude standard deviation and pressure fluctuation values show a synchronous amplification trend, it is determined that "production process fluctuations are increasing." This status value is transmitted to the visualization data processing platform via the internal industrial Ethernet network for subsequent input into response delay and angle deviation assessments.

[0068] Step S322: Detecting the production line response delay condition based on the degree of aggravation of the production process fluctuation; In an embodiment of the present invention, the "fluctuation intensification state identifier" obtained in the previous step is used as a trigger condition to start the response delay analysis module. The specific process includes: collecting the sending time of all current spray control actions (i.e., the timestamp of the control instruction), and the start time of the execution action recorded by the displacement encoder or the photoelectric proximity sensor, and calculating the time difference between the two as the response delay. Taking the actual batter nozzle control unit as an example, the system collects the current nozzle action data every 10 seconds. When it is detected that the start-up lag of the nozzle servo system exceeds 80 milliseconds and the stabilization time of the nozzle feedback signal exceeds 300 milliseconds, the system determines that there is a hysteresis in the cycle response. At the same time, the response delay value is combined with the fluctuation degree parameter in step S321. If the two data exceed the threshold at the same time, the current state is recorded as "abnormal response delay state", and the state value is used for subsequent angle deviation analysis.

[0069] Step S323: detecting the deviation of the nozzle angle control of the production line according to the degree of fluctuation in the production process and the response delay of the production line; In an embodiment of the present invention, the confirmed fluctuation degree status and response delay data are called to enable the angle deviation detection process. In the spraying system, the nozzle angle setting value is issued by the PLC control system, and the actual angle is fed back through the laser displacement sensor. The system obtains the nozzle deviation angle by comparing the difference between the set angle and the feedback angle. The system sets 1° as the minimum detection accuracy, continuously monitors the set angle and feedback angle of each movement of the nozzle, calculates the difference and performs sliding mean processing. If the average deviation angle exceeds 2.5° in any 10 consecutive cycles, and the fluctuation degree parameter exceeds 1.0 and the response delay is higher than 120 milliseconds, it is judged that the nozzle control in this section has an "angle control deviation state". In the actual dairy product seasoning spraying link, due to insufficient nozzle deflection angle, the seasoning liquid fails to evenly cover the target area, and the image recognition system captures obvious color deviation on the product surface.

[0070] Step S324: determining the production line ratio imbalance condition based on the production line nozzle angle control deviation condition and the production line response delay condition; In this embodiment of the present invention, the angle deviation data obtained in step S323 and the response delay data in step S322 are simultaneously fed into the spray ratio analysis module. The actual ratio deviation is calculated by calculating the difference between the nozzle's discharge volume per unit time (monitored in real time by a flow meter) and the target discharge volume. For example, the production recipe calls for a 1:4 spray ratio of seasoning liquid to whey. However, after angle deviation causes the actual discharge volume to change, testing reveals that the seasoning liquid spray mass is over 8% under-sprayed. The system then normalizes this deviation value to form a ratio imbalance coefficient. If this coefficient consistently exceeds 0.1 (i.e., a 10% deviation), the system records the current state as a "significant ratio imbalance." This ratio imbalance is not only reflected in nozzle angle control errors but also in material response delays. This lag leads to inaccurate discharge control, which in turn results in ratio deviation. This state is used as input to determine product uniformity anomalies.

[0071] Step S325: Detecting abnormal processing uniformity of the production line based on the imbalance of the production line ratio.

[0072] In an embodiment of the present invention, this step is achieved by combining image recognition technology with a numerical comparison method. A high-resolution industrial camera is deployed at the end of the production line to continuously capture images of the surface of the product that has been sprayed. The image processing algorithm analyzes indicators such as color distribution, surface structure texture, and light reflection uniformity. The image processing platform uses a color histogram analysis method to obtain color discreteness, evaluates texture continuity through a grayscale co-occurrence matrix, and uses Fourier transform to calculate the surface reflection frequency to comprehensively form a "surface uniformity scoring index." At the same time, the ratio imbalance coefficient in step S324 is called for joint analysis. In specific operations, if within a certain period of time, the product surface color deviation coefficient is higher than 15%, the texture continuity index decreases by more than 20%, and the ratio offset during this period exceeds 10%, the system will identify it as an "abnormal processing uniformity state" and display it in real time with a red alarm graphic on the visualization platform, and record it in the quality monitoring system.

[0073] Preferably, step S33 includes the following steps: Step S331: detecting the uneven thickness of the finished product crystal layer based on the abnormal uniformity of the production line and the attenuation of the processing control efficiency; In an embodiment of the present invention, this step uses the production line processing uniformity anomaly index obtained in step S32 and the control efficiency attenuation percentage in step S31 as input, and uses a high-resolution scanning electron microscope (SEM) or a laser scanning microscope to perform a microstructural scan of the product surface collected on the non-dairy creamer production line. The surface crystal layer thickness is measured at multiple sampling points collected, and a thickness distribution curve is established by measuring the crystal layer thickness distribution in the microscopic image. Automatic image analysis software is used to identify the edge of the crystal layer and calculate the thickness, and the actual values of the crystal layer thickness at different sampling points are obtained. The standard deviation and range of the thickness are calculated as quantitative indicators of the uneven thickness of the crystal layer. This indicator is compared with the uniformity anomaly and control efficiency attenuation data to confirm the spatial distribution characteristics of the crystal layer thickness differences, forming a specific numerical description of the thickness variation, and providing basic data for subsequent tension deviation determination.

[0074] Step S332: determining the surface tension deviation of the non-dairy creamer based on the different thickness of the finished product crystal layer; In an embodiment of the present invention, combined with the statistical data of the different thicknesses of the crystal layer obtained in step S331, a surface tension tester (such as a contact angle meter) is used to perform multi-point contact angle measurements on the surface of the non-dairy creamer. The surface free energy and the corresponding surface tension distribution curve are calculated using the multi-point contact angle data. The crystal layer thickness data is mapped to the corresponding contact angle measurement points, the effect of thickness changes on the surface tension is analyzed, and the tension deviation value of each point is calculated. Using a numerical integration method, the tension deviation values of each measuring point are weighted and averaged to obtain an overall surface tension deviation index. This index reflects the spatial fluctuation characteristics of the surface tension caused by the uneven thickness of the crystal layer. The index is output in the form of percentage deviation, providing a quantitative basis for the prediction of uneven agglomeration.

[0075] Step S333: predicting the uneven agglomeration of the non-dairy creamer finished product based on the surface tension deviation of the non-dairy creamer; In an embodiment of the present invention, based on the surface tension deviation index obtained in step S332, combined with the real-time dynamic video monitoring data of the non-dairy creamer particles on the production line, the particle image analysis technology is used to quantify the product agglomeration phenomenon. A high-definition industrial camera is used to capture the particle aggregation status on the product conveyor belt. The image processing system identifies the agglomeration area through edge detection and clustering algorithm, and calculates the size distribution and agglomeration frequency of the agglomerated particles. A correlation is established between the agglomeration frequency and the tension deviation index, and the influence weight of the tension deviation on the agglomeration formation is derived through statistical analysis to form an agglomeration anomaly parameter. This parameter expresses the degree of uniformity imbalance of the finished product agglomeration in numerical form, and the agglomeration anomaly data is output as a quantitative result of the uneven agglomeration situation.

[0076] Step S334: Identifying the change in depth of the fatty acid coating layer according to the abnormal uniformity of the production line; In an embodiment of the present invention, this step is based on the production line processing uniformity anomaly index provided in step S32, and uses a Fourier transform infrared spectrometer (FTIR) to perform a molecular composition depth analysis on the surface of the non-dairy creamer sample. By performing a spectral scan on the non-dairy creamer sample, the intensity and changes of the characteristic absorption peak of the fatty acid are detected, and the thickness distribution of the fatty acid coating layer is obtained by combining spectral depth profiling technology. Through layer-by-layer spectral analysis, the changes in the depth of the fatty acid coating layer are identified, and a fatty acid coating layer depth change curve is formed. Combined with the uniformity anomaly index, the relationship between the coating layer depth change and the processing uniformity anomaly is evaluated, and a quantitative description of the coating layer depth change is generated. This data serves as an input parameter for the subsequent dissolution behavior imbalance estimation.

[0077] Step S335: estimating the imbalance of the dissolution behavior of the non-dairy creamer based on the change in the depth of the fatty acid coating layer and the uneven agglomeration of the non-dairy creamer product; In an embodiment of the present invention, the fatty acid coating depth change curve of step S334 and the uneven agglomeration abnormality parameter of step S333 are used as input, and a dynamic dissolution tester is used to analyze the dissolution behavior of the non-dairy creamer sample. The dissolution rate and dissolution uniformity of samples with different coating layer thicknesses and agglomeration conditions in a standard dissolution medium are measured, and the dissolution time-concentration curve is collected. By comparing the dissolution curve under normal working conditions, the deviation of the dissolution rate and the dissolution uniformity fluctuation index are calculated. A multi-parameter weighted fusion algorithm is used to synthesize the dissolution behavior imbalance index. This index quantitatively characterizes the abnormal state during the dissolution process and outputs the dissolution behavior imbalance value as a key basis for determining the working condition deviation.

[0078] Step S336: Determine the deviation of the production line operating condition based on the imbalance of the dissolution behavior of the non-dairy creamer.

[0079] In an embodiment of the present invention, the dissolution behavior imbalance index obtained in step S335 is comprehensively analyzed with the uniformity anomaly of the preceding step S32, the agglomeration anomaly of S333, and the coating layer depth change data of S334. A weighted normalization method is used to uniformly convert the dimensions of each parameter to construct a comprehensive indicator system for operating condition deviation. Combined with historical operating condition data, a threshold comparison technique is used to determine whether the current production line operating condition deviates from the standard. The operating condition deviation is quantified and output in percentage form, reflecting the degree of deviation of the overall operating status of the production line. This comprehensive indicator provides data support for subsequent production line adjustments and optimizations, realizing digital visualization and dynamic control of the production process.

[0080] Preferably, step S4 includes the following steps: Step S41: identifying the degree of heterogeneity of fat emulsification of the non-dairy creamer based on the abnormal deviation of the non-dairy creamer product quality; In an embodiment of the present invention, this step uses a light scattering particle size analyzer (such as a dynamic light scattering (DLS)) to analyze the particle size and distribution of fat emulsion particles in the sampled non-dairy creamer, based on the quality deviation data of the non-dairy creamer product obtained in the previous step (such as particle size distribution abnormality index, agglomeration abnormality, and uniformity fluctuation value). The size distribution curve of the fat emulsion particles is obtained through repeated measurements, and the mean, median, and multimodal characteristics of the distribution are calculated to determine the uniformity of the fat emulsion structure. The number of peaks in the distribution curve and the relative difference between the peaks are used to form an emulsion heterogeneity index, with larger values indicating more uneven emulsion structure. This index is then numerically normalized and output as a quantitative description of the degree of fat emulsion heterogeneity. This data provides a basis for subsequent detection of internal particle structure damage.

[0081] Step S42: detecting the degree of damage to the particle structure inside the product based on the heterogeneity of the fat emulsification of the non-dairy creamer; In this embodiment of the present invention, the fat emulsification heterogeneity index obtained in step S41 is used as input to observe the nanoscale particle structure of non-dairy creamer samples using transmission electron microscopy (TEM). Ultrathin sections are prepared from representative samples, and high-resolution electron microscopy is used to obtain images of the internal fat particle morphology and interface integrity. Image processing algorithms are used to identify damaged or distorted particle areas, and the ratio of the damaged area to the intact particle area is measured to form a structural damage rate indicator. Combined with the heterogeneity index, the impact of uneven fat emulsification structure on internal particle damage is analyzed, and a quantitative parameter for the degree of damage is established. The degree of damage is quantitatively output as a percentage, clearly defining the deterioration of the product's internal structure and providing precise parameters for evaluating the evolution of production process deviations.

[0082] Step S43: evaluating the deviation evolution trend of the non-dairy creamer production process based on the degree of damage to the internal particle structure of the product and the degree of heterogeneity of the fat emulsification of the non-dairy creamer; In an embodiment of the present invention, this step combines the particle structure breakage rate of step S42 and the fat emulsification heterogeneity index of step S41, and adopts a time series data analysis method to track the change trend of the two indicators according to the time sequence of the production batches. The evolution rate and fluctuation range of the production process deviation are calculated using the change trend curve. Combined with the historical operation data of the production line, the deviation state of the production process is divided into stages, which are calibrated as initial deviation, moderate deviation and severe deviation states. The evolution situation is expressed in the form of a continuous variable curve, and the production process deviation value is output to reflect the change trend of the production process stability over time. This value is the basis for subsequent defect location tracing to ensure dynamic monitoring of production process changes.

[0083] Step S44: tracing the defect location of the non-dairy creamer production line based on the deviation evolution trend of the non-dairy creamer production process to obtain production line defect location data; In an embodiment of the present invention, the production process deviation evolution trend value obtained in step S43 is used, combined with the sensor collection data of each key process node of the production line, including real-time monitoring information such as temperature, humidity, pressure, stirring speed and material conveying rate, and the multi-sensor data fusion method in the fault diagnosis technology is used to identify the abnormal characteristics of each process node. By establishing a node abnormality threshold, the process node corresponding to the abnormal peak of the deviation evolution trend is located. Using time synchronization marks, the starting node and diffusion path of the defect are traced to form a defect location distribution map of the production line. The defect location data is output in the form of node number and coordinates, which clearly points out the specific process link where the abnormality exists in the production line, and provides accurate spatial positioning information for production line optimization.

[0084] Step S45: performing optimization control processing on the non-dairy creamer processing production line according to the production line defect position data to obtain optimization control data on the non-dairy creamer processing production line.

[0085] In an embodiment of the present invention, this step is based on the defect position data obtained in step S44, combined with the production line equipment parameters and process flow rules, and the process parameters are adjusted using automated control technology. For the process node corresponding to the defect position, relevant parameters such as stirring speed, heating temperature, material ratio, etc. are adjusted, and the adjustment instructions are sent to the corresponding equipment through the industrial field bus system. The adjusted production data is collected in real time, and a closed-loop feedback mechanism is used to monitor the adjustment effect, and the control parameters are continuously corrected to form an optimized control data set. The data set includes the adjustment amplitude, adjustment frequency and quality indicators after process improvement. The data is recorded in the form of a time series and used for subsequent production line status evaluation to ensure that the processing process is stable and continuously optimized.

[0086] The present invention further provides a digital visualization control system for a food processing production line, which is used to execute the digital visualization control method for a food processing production line as described above. The digital visualization control system for a food processing production line comprises: A model building module is used to obtain the design data of the food processing production line; based on the design data of the food processing production line, collect the internal processing structure data of the non-dairy creamer production line; and build a simulation model of the non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line; The module predicts the chain growth trend of abnormal deviations, which is used to determine the processing disturbance coupling instability characteristics based on the non-dairy creamer processing line simulation model; detect the production line pressure response surge trend based on the processing disturbance coupling instability characteristics; and predict the chain growth trend of abnormal deviations based on the production line pressure response surge trend; The quality abnormal deviation determination module is used to detect the decline of processing control efficiency based on the abnormal deviation chain growth trend; determine the deviation of production line working condition based on the decline of processing control efficiency; and determine the abnormal deviation of non-dairy creamer product quality based on the deviation of production line working condition; The optimization control processing module is used to trace the defect location of the non-dairy creamer production line based on the abnormal deviation of the non-dairy creamer product quality, and obtain the production line defect location data; based on the production line defect location data, the non-dairy creamer processing production line is optimized and controlled to obtain the optimization control data of the non-dairy creamer processing production line.

[0087] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present 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 present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A digital visualization control method applied to a food processing production line, characterized in that: The following steps are involved: Step S1: Acquire design data of a food processing production line; collect internal processing structure data of a non-dairy creamer production line based on the design data of the food processing production line; and construct a simulation model of a non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line; Step S2: determining the processing disturbance coupling instability characteristics based on the non-dairy creamer processing line simulation model; detecting the production line pressure response surge trend based on the processing disturbance coupling instability characteristics; and predicting the abnormal deviation chain growth trend based on the production line pressure response surge trend; Step S3: detecting a decrease in processing control efficiency based on the abnormal deviation from the chain growth trend; determining a production line operating condition deviation based on the processing control efficiency decrease; and determining a non-dairy creamer product quality abnormal deviation based on the production line operating condition deviation; Step S4: based on the abnormal deviation of non-dairy creamer product quality, the defect position of the non-dairy creamer production line is traced to obtain the production line defect position data; based on the production line defect position data, the non-dairy creamer processing production line is optimized and controlled to obtain the non-dairy creamer processing production line optimization control data.

2. The digital visualization control method for food processing production line according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Acquire food processing production line design data; Step S12: collecting internal processing structure data of the non-dairy creamer production line based on the food processing production line design data; Step S13: determining the processing flow status of the non-dairy creamer production line based on the food processing production line design data and the internal processing structure data of the non-dairy creamer production line; Step S14: constructing a simulation model of the non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line and the processing flow status of the non-dairy creamer production line.

3. The digital visualization control method for food processing production line according to claim 2 is characterized in that: Step S14 includes the following steps: Step S141: using the internal processing structure data of the non-dairy creamer production line to determine the internal physical layout of the non-dairy creamer production line with a point cloud accuracy of ±2mm; Step S142: collecting processing structure nozzle position distribution data using the internal physical layout of the non-dairy creamer production line; Step S143: Identifying the geometric structural features of the internal circulation air duct of the production line based on the internal physical layout of the non-dairy creamer production line; Step S144: using the geometric structure characteristics of the internal circulation air duct of the production line and the processing structure nozzle position distribution data to set the injection speed to 2.5m / s and determine the airflow direction inside the production line; Step S145: evaluating the processing stability of the non-dairy creamer production line based on the airflow direction within the production line to obtain production line processing stability data; Step S146: Evaluate the stability of the finished product of the non-dairy creamer production line based on the production line processing stability data and the processing flow status of the non-dairy creamer production line; Step S147: Construct a simulation model of the non-dairy creamer processing production line based on the stability of the finished product of the non-dairy creamer production line and the geometric structure characteristics of the internal circulation air duct of the production line.

4. The digital visualization control method for food processing production line according to claim 1, characterized in that: The determination of the machining disturbance coupling instability characteristics in step S2 includes: According to the non-dairy creamer processing line simulation model, the non-dairy creamer processing production process is simulated to obtain the non-dairy creamer processing production process simulation data; Identify abnormalities in the hot air outlet guide cavity position based on simulated data from the non-dairy creamer processing process; Detect the lack of symmetry of the internal air duct based on the abnormal position of the hot air outlet guide cavity; Detect the growth of local vortex based on the lack of symmetry of the internal air duct and the abnormal position of the hot air outlet guide cavity; Determine the temperature retention area inside the production line based on the internal local eddy current growth conditions; Evaluate the degree of angle deviation of the processing nozzle based on the simulation data of the non-dairy creamer processing production process; The nozzle hot air interference growth trend is predicted based on the degree of angle deviation of the processing nozzle in the temperature retention area of the production line, and the nozzle hot air interference data is obtained; Predict the growth trend of nozzle particle size distribution stratification based on nozzle hot air interference data; The processing disturbance coupling instability characteristics are determined based on the stratified growth trend of the nozzle particle size distribution and the temperature stagnation area inside the production line.

5. The digital visualization control method for food processing production line according to claim 1 is characterized in that: The production line pressure response surge trend detection in step S2 includes: The trend of increasing local disturbance deviation is predicted by using the instability characteristics of machining disturbance coupling; Estimate the increase in particle size distribution differences based on the increasing trend of local disturbance deviations; Estimate the local retention of large particles in the production line based on the increasing trend of particle size distribution differences; Determine the flow blockage situation inside the production line based on the local retention of large particles in the production line; Detect the growth degree of suction negative pressure deviation based on the flow blockage situation inside the production line; Calculate the internal suction demand growth data based on the internal flow blockage of the production line and the growth degree of suction negative pressure deviation; Detect the degree of wind pressure structure power growth based on internal suction demand growth data; The dramatic increase trend of the production line's pressure response is detected based on the growth rate of the wind pressure structure power and the growth rate of the suction negative pressure deviation.

6. The digital visualization control method for food processing production line according to claim 1, characterized in that: The abnormal deviation chain growth trend prediction in step S2 includes: Test the stress concentration trend of the internal structure of the processing according to the sharp increase trend of the pressure response of the production line; Detect the internal structure crack growth condition when the stress concentration trend of the internal structure during processing is greater than 50N; Determine the internal structure aging trend based on the internal structure crack growth status and the stress concentration trend of the processed internal structure; Measure the deformation degree of the internal structure of the production line according to the stress concentration trend of the internal structure during processing and the aging trend of the internal structure; Evaluate the internal precision degradation of the production line based on the internal structural deformation and aging trend of the production line; When the precision attenuation inside the production line exceeds 68.2% and the degree of structural deformation inside the production line, the abnormal deviation from the chain growth trend is predicted.

7. The digital visualization control method for food processing production line according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: detecting the attenuation of processing control efficiency based on the abnormal deviation from the chain growth trend; Step S32: detecting abnormality in processing uniformity of the production line according to the processing control efficiency attenuation condition; Step S33: determining the deviation of the production line working condition according to the abnormality of the production line processing uniformity and the attenuation of the processing control efficiency; Step S34: determining the abnormal deviation of the non-dairy creamer product quality based on the production line working condition deviation and the abnormal processing uniformity of the production line.

8. The digital visualization control method for food processing production line according to claim 7, characterized in that: Step S33 includes the following steps: Step S331: detecting the uneven thickness of the finished product crystal layer based on the abnormal uniformity of the production line and the attenuation of the processing control efficiency; Step S332: determining the surface tension deviation of the non-dairy creamer based on the different thickness of the finished product crystal layer; Step S333: predicting the uneven agglomeration of the non-dairy creamer finished product based on the surface tension deviation of the non-dairy creamer; Step S334: Identifying the change in depth of the fatty acid coating layer according to the abnormal uniformity of the production line; Step S335: estimating the imbalance of the dissolution behavior of the non-dairy creamer based on the change in the depth of the fatty acid coating layer and the uneven agglomeration of the non-dairy creamer product; Step S336: Determine the deviation of the production line operating condition based on the imbalance of the dissolution behavior of the non-dairy creamer.

9. The digital visualization control method for food processing production line according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: identifying the degree of heterogeneity of fat emulsification of the non-dairy creamer based on the abnormal deviation of the non-dairy creamer product quality; Step S42: detecting the degree of damage to the particle structure inside the product based on the heterogeneity of the fat emulsification of the non-dairy creamer; Step S43: evaluating the deviation evolution trend of the non-dairy creamer production process based on the degree of damage to the internal particle structure of the product and the degree of heterogeneity of the fat emulsification of the non-dairy creamer; Step S44: tracing the defect location of the non-dairy creamer production line based on the deviation evolution trend of the non-dairy creamer production process to obtain production line defect location data; Step S45: performing optimization control processing on the non-dairy creamer processing production line according to the production line defect position data to obtain optimization control data on the non-dairy creamer processing production line.

10. A digital visual control system applied to a food processing production line, characterized in that: The method for executing the digital visualization control method for a food processing production line according to claim 1, wherein the digital visualization control system for a food processing production line comprises: A model building module is used to obtain the design data of the food processing production line; based on the design data of the food processing production line, collect the internal processing structure data of the non-dairy creamer production line; and build a simulation model of the non-dairy creamer production line based on the internal processing structure data of the non-dairy creamer production line; The module predicts the chain growth trend of abnormal deviations, which is used to determine the processing disturbance coupling instability characteristics based on the non-dairy creamer processing line simulation model; detect the production line pressure response surge trend based on the processing disturbance coupling instability characteristics; and predict the chain growth trend of abnormal deviations based on the production line pressure response surge trend; The quality abnormal deviation determination module is used to detect the decline of processing control efficiency based on the abnormal deviation chain growth trend; determine the deviation of production line working condition based on the decline of processing control efficiency; and determine the abnormal deviation of non-dairy creamer product quality based on the deviation of production line working condition; The optimization control processing module is used to trace the defect location of the non-dairy creamer production line based on the abnormal deviation of the non-dairy creamer product quality, and obtain the production line defect location data; based on the production line defect location data, the non-dairy creamer processing production line is optimized and controlled to obtain the optimization control data of the non-dairy creamer processing production line.

Images