Plastic product production line visual management method and system and medium

Abstract

The invention relates to the technical field of visual management, in particular to a visual management method and system for a plastic product production line and a medium. The method comprises the following steps of scanning spectral characteristics of raw materials, analyzing the spectral characteristics to generate multi-dimensional characteristics, performing thermal stability evaluation, constructing a raw material topological unit, acquiring environment data of a production line to construct a virtual environment field, performing virtual reconstruction on the raw material topological unit, generating a visual production line, and displaying the visual production line. Further obtaining historical production records, performing thermal topology tracking, constructing a production line temperature change field, performing material heating simulation based on the field and a raw material topological unit, recording a heating response behavior, combining with calculation to obtain a heat influence coefficient, and finally performing defect development prediction according to the heat influence coefficient and the temperature field. And visual warning in advance is realized, so that the production efficiency and the quality management are improved. The visual management method for the plastic product production line is more efficient and more stable.

Description

Technical Field

[0001] The present invention relates to the technical field of visual management, and particularly to a visual management method, system and medium for a plastic product production line. Background Art

[0002] Traditional production line management methods cannot meet the requirements of real-time monitoring of production processes and data analysis, resulting in low production efficiency and uneven product quality. Especially under the influence of complex environmental factors, the stability of the production line and product consistency face huge challenges. Therefore, an advanced management method is needed to improve the visualization and intelligent level of plastic product production, and ensure the high efficiency, stability and safety of the production process. Currently, many production lines still rely on manual inspection and experience judgment, lacking systematic data analysis tools, making it difficult to quickly respond and handle problems such as fluctuations in raw material quality, environmental changes or equipment failures. The influence of environmental factors such as temperature and humidity on material properties is often ignored. In addition, traditional methods cannot effectively integrate historical data and real-time data, resulting in information island phenomena in the production process, limiting the ability to predict and warn of potential problems and making it difficult to achieve true intelligent management. Summary of the Invention

[0003] Based on this, it is necessary to provide a visual management method, system and medium for a plastic product production line to solve at least one of the above technical problems.

[0004] To achieve the above object, the visual management method for a plastic product production line includes the following steps:

[0005] Step S1: Scan the spectral characteristics of the raw materials of plastic products; perform multi-dimensional feature analysis based on the raw material spectral characteristics to generate multi-dimensional characteristics of the raw materials; conduct thermal stability evaluation on the multi-dimensional characteristics of the raw materials, and construct a raw material topological unit based on the evaluation results;

[0006] Step S2: Collect production line environment data, and construct a virtual environment field based on the production line environment data; perform virtual reconstruction of the production line on the raw material topological unit according to the virtual environment field to generate a virtual visual production line;

[0007] Step S3: Obtain historical records of the plastic product production line; perform thermal topological tracking on the virtual visual production line according to the historical records of the plastic product production line, and construct a production line temperature change field based on the thermal topology;

[0008] Step S4: Perform material heating simulation based on the production line temperature change field and the raw material topological unit, and collect the heating response behaviors corresponding to each heating degree; perform long-term influence calculation based on the material heating response behaviors to obtain the production line heat influence coefficient;

[0009] Step S5: Predict the defect development of the raw material topological unit according to the production line heat influence coefficient and the production line temperature field, and give an early visual warning based on the defect development trend.

[0010] Through scanning the spectral characteristics of the raw materials, the present invention realizes a comprehensive analysis of the raw material characteristics, making the understanding of the raw materials deeper. The generated multi-dimensional characteristics of the raw materials provide a solid basis for subsequent thermal stability evaluation. The raw material topological unit constructed based on the evaluation results provides a scientific framework for production. Collecting the production line environment data and constructing the virtual environment field improve the analysis ability of the production line environment adaptability. The virtual reconstruction of the production line based on the virtual environment field realizes the visualization of the production process, effectively improving the production efficiency. Obtaining the historical records of the plastic product production line and performing thermal topology tracking makes the analysis of the past production conditions more systematic. The temperature change field of the production line constructed based on the thermal topology provides the necessary data support for understanding the temperature influence in the production process. The record of the material heat simulation and the heat response behavior deeply reveals the physical characteristics of the material during the heating process, providing a specific basis for the long-term influence calculation. The acquisition of the production line heat influence coefficient enhances the scientificity and effectiveness of the production line heat management. The defect development prediction based on the heat influence coefficient and the production line temperature field provides a forward-looking guidance for problem prevention in the production process. The early visual warning mechanism effectively reduces the risks brought by potential defects, thus ensuring the stability of the production line and the quality of the products.

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

[0012] Step S11: Scan the spectral characteristics of the plastic product raw materials in the wavelength range of 200nm - 2500nm to obtain the raw material spectral characteristics;

[0013] Step S12: Analyze the molecular structure of the raw material spectral characteristics to generate the raw material molecular topological structure;

[0014] Step S13: Statistically analyze the molecular weight distribution based on the raw material molecular topological structure, and calculate the raw material hydrodynamic parameters based on the molecular weight distribution data, where the polydispersity coefficient is set to ≤1.8;

[0015] Step S14: Evaluate the thermal stability of the plastic product raw materials according to the raw material hydrodynamic parameters, with the 5% mass loss temperature ≥280°C as the thermal degradation threshold;

[0016] Step S15: Perform matching mapping on the thermal degradation threshold based on the preset additive data, and construct the raw material topological unit based on the mapping result.

[0017] Through spectral feature scanning of plastic product raw materials within a wide wavelength range, the present invention realizes the comprehensive identification and analysis of raw materials. The generated raw material molecular topological structure provides a scientific basis for in-depth understanding of material properties. The molecular weight distribution statistics and the calculation of hydrodynamic parameters enhance the ability to evaluate the processing performance of raw materials. The set polydispersity coefficient standard ensures the consistency of fluid characteristics, thereby improving the fluidity and mixing efficiency in the production process. The thermal stability evaluation provides a quantitative analysis for determining the heat resistance of raw materials. Using the set thermal degradation threshold can timely identify potential material degradation risks. The matching mapping based on the preset additive data and the thermal degradation threshold provides an effective material selection reference for improving the thermal stability of raw materials. The constructed raw material topological unit provides a structured basis for subsequent production design and process optimization, effectively improving the overall management level of the production line, the material usage efficiency, and the quality of the final product.

[0018] Preferably, the collecting of production line environment data in step S2 and the construction of a virtual environment field based on the production line environment data include:

[0019] Collect physical environment parameters of the production line;

[0020] Map the physical constraints of the production line through the physical environment parameters of the production line;

[0021] Construct a geometric framework of the production line based on the physical environment parameters of the production line;

[0022] Conduct environmental interaction response analysis on the geometric framework of the production line according to the physical constraints of the production line, and construct an environmental response field based on the environmental interaction response data;

[0023] Conduct virtual environment field simulation on the geometric framework of the production line according to the environmental response field.

[0024] The present invention realizes the comprehensive monitoring and evaluation of the operation status of the production line by collecting the physical environment parameters of the production line. The mapped physical constraints provide a scientific basis for understanding the limiting conditions of various operations in the production process. The constructed geometric framework of the production line lays a solid foundation for subsequent environmental response analysis. Through environmental interaction response analysis, the actual impact of the physical environment on the production process can be captured, thereby optimizing the production line design and layout. The environmental response field constructed based on the environmental response data enhances the dynamic monitoring ability of various environmental factors. The realization of virtual environment field simulation provides a visual basis for the performance evaluation of the production line under different operating conditions, overall improving the understanding and control ability of the adaptability to the production environment, and having obvious advantages in coping with potential risks and optimizing production efficiency.

[0025] Preferably, the virtual reconstruction of the raw material topological unit for the production line according to the virtual environment field in step S2 includes:

[0026] Triangulate the raw material topological unit and perform coordinate system projection to obtain vertex coordinates;

[0027] Construct a local surface derivative field based on the vertex coordinates;

[0028] Perform vertex embedding on the local surface derivative field to generate a space - restricted envelope;

[0029] Perform collision detection on the space - restricted envelope and filter collision - free data;

[0030] Based on the collision - free data, perform boundary constraint detection on the space - restricted envelope, mark abnormal points and critical points, adjust the projection of abnormal points, and perform local geometric optimization on critical points to generate adjusted vertex coordinates;

[0031] Based on the adjusted vertex coordinates, model the flow characteristics of plastic products to obtain plastic flow simulation data;

[0032] Simulate the plastic product virtual entity according to the plastic flow simulation data;

[0033] Perform temperature - stress coupling analysis on the plastic product virtual entity and identify the forming characteristics of the plastic product;

[0034] Predict the deformation error of the plastic product forming characteristics and pre - regulate the adjusted vertex coordinates based on the predicted deformation error to generate preventive error vertex coordinates;

[0035] Project the preventive error vertex coordinates into the virtual environment field to construct a virtual visual production line.

[0036] The present invention realizes the accurate acquisition of vertex coordinates by triangulating the raw material topological unit and performing coordinate system projection, laying a foundation for subsequent modeling. The construction of the local surface derivative field improves the understanding of material morphological changes and provides a scientific basis for the analysis of the injection molding flow process. The generated space - restricted envelope effectively constrains the material flow range. Collision detection and collision - free data screening ensure the safety and accuracy of the simulation process. The abnormal points and critical points marked based on boundary constraint detection enhance the sensitivity to potential defects. Projection adjustment and local geometric optimization realize the refinement and stability of the design. The simulation data obtained from the plastic flow characteristics modeling provides guidance for the optimization of the production process. The forming characteristics identified by temperature - stress coupling analysis provide data support for product quality control. The realization of the deformation error prediction link provides a forward - looking solution for subsequent adjustments. The generation of preventive error vertex coordinates greatly reduces the error risk in the forming process. The projection in the virtual environment field constructs a visual production line, overall improving the intelligent level and efficiency of production, and providing a solid technical foundation for the high - quality production and management of plastic products.

[0037] Preferably, step S3 includes the following steps:

[0038] Step S31: Obtain the historical production line records of plastic products; perform time series decomposition on the historical production line records of plastic products, and identify the periodic temperature fluctuation patterns;

[0039] Step S32: Perform anomaly detection on the periodic temperature fluctuation patterns, and eliminate the abnormal fluctuations to generate the normal operating temperature characteristics;

[0040] Step S33: Based on the normal operating temperature characteristics, perform spatial registration on the virtual visualization production line to generate a thermal mapping grid; perform radial basis function interpolation on the thermal mapping grid, and construct a heat flux vector field;

[0041] Step S34: Extract the thermal topology skeleton of the heat flux vector field; perform time evolution simulation on the thermal topology skeleton to obtain the dynamic thermal topology;

[0042] Step S35: Perform stability analysis on the dynamic thermal topology, and extract the critical heat flux nodes based on the stability data; construct a production line temperature change field based on the critical heat flux nodes and the thermal mapping grid.

[0043] The present invention realizes the comprehensive analysis and evaluation of the production process by obtaining the historical production line records. The time series decomposition technology helps to identify the periodic temperature fluctuation patterns, providing a basis for subsequent data processing and optimization. The anomaly detection and elimination process ensure that the generated temperature characteristics have high accuracy, thus laying a foundation for the temperature control management of the production line. The spatial registration and the construction of the thermal mapping grid enhance the understanding of the heat distribution, providing the necessary spatial data support for the production line optimization. The radial basis function interpolation method improves the accuracy and usability of the heat flux vector field. The extracted thermal topology skeleton provides an intuitive basis for the visualization of the temperature change trend. The generation of the dynamic thermal topology supports the in-depth research and analysis of the heat conduction mechanism. The stability analysis and the extraction of the critical heat flux nodes provide an effective tool for identifying potential risks. The construction of the temperature change field provides a scientific basis for formulating the temperature control strategy during the production process, comprehensively improving the refined management level and intelligent decision-making ability of the production process.

[0044] Preferably, the material heating simulation based on the production line temperature change field and the raw material topology unit in step S4, and the collection of the heating response behaviors corresponding to each heating degree include:

[0045] Perform multi-physical field coupling on the production line temperature change field and the raw material topology unit to obtain a heat-material interaction field;

[0046] Based on the heat-material interaction field, perform material heating simulation on the plastic products to obtain simulated heating data;

[0047] Identify the heat distribution of the simulated heat data and mark the corresponding heat level of the plastic product based on the heat distribution;

[0048] Analyze the heat response behavior corresponding to each heat level according to the heat level of the plastic product.

[0049] The present invention realizes the effective integration of the production line temperature change field and the raw material topological unit through multi-physical field coupling. The generated thermal-material interaction field provides a dynamic perspective for material property analysis. The material heat simulation carried out based on this interaction field generates detailed heat data, enabling the quantification basis for the performance change of plastic products under different temperature conditions. The heat distribution identification technology ensures the accurate marking of the heat level of each part of the product, providing a clear reference for subsequent performance evaluation. Analyzing the heat response behavior corresponding to each heat level reveals the sensitivity and reaction mechanism of the material to heat effects, overall enhancing the understanding of the material's thermal properties, effectively guiding the temperature control and material selection in the production process, and providing solid data support and theoretical basis for the quality control and process optimization of plastic products.

[0050] Preferably, the long-term impact estimation according to the material heat response behavior described in step S4 includes:

[0051] Perform time series extrapolation on the heat response behavior corresponding to each heat level to obtain the predicted cumulative heat effect;

[0052] Perform thermal fatigue mapping according to the predicted cumulative heat effect to generate a thermal fatigue coefficient;

[0053] Perform matching links on the heat response behavior based on the thermal fatigue coefficient, and estimate the material life data at each heat level based on the matching links;

[0054] Quantify the heat impact according to the material life estimation data to obtain the production line heat impact coefficient.

[0055] The present invention realizes the accurate prediction of the heat effect of the material under long-term impact by performing time series extrapolation on the heat response behavior corresponding to each heat level. The generated predicted cumulative heat effect provides a quantitative basis for subsequent thermal fatigue mapping. The generation of the thermal fatigue coefficient provides a scientific basis for the evaluation of the material service life. By performing matching links on the heat response behavior, detailed material life estimation data are provided for each heat level. The completion of heat impact quantification enables the theoretical support and practical reference for obtaining the production line heat impact coefficient. The entire process comprehensively enhances the understanding of the material durability and its performance in complex environments, enhances the reliability and operation efficiency of the production line, and provides an effective guarantee for long-term stable production.

[0056] Preferably, step S5 includes the following steps:

[0057] Step S51: Reconstruct the tensor field of the heat influence coefficient of the production line to obtain the spatial distribution matrix of thermal influence;

[0058] Step S52: Perform spectral density conversion on the temperature field of the production line to obtain the temperature frequency domain characteristics;

[0059] Step S53: Quantify the thermal stress effect on the spatial distribution matrix of thermal influence and the temperature frequency domain characteristics to obtain the comprehensive thermal stress effect;

[0060] Step S54: Identify the material vulnerable areas of the raw material topological units;

[0061] Step S55: Perform Monte Carlo simulation iteration N = 10,000 times on the material vulnerable areas based on the comprehensive thermal stress effect, calculate the defect state transition probability P(T|T-1), and generate the defect evolution path data matrix based on the state transition matrix S(t,t+Δt) to generate the defect evolution path, where T is the number of iterations of the Monte Carlo simulation, that is, the current state number reached by the simulation, t is the physical time parameter, and Δt is the increment of physical time, that is, in the process of defect state evolution, the time step for each simulation calculation;

[0062] Step S56: Perform time-axis extrapolation on the defect evolution path, set the prediction step T = 50 time units, and calculate the defect growth rate G(t) and cumulative damage degree D(t) at the future time T to obtain the defect development trend;

[0063] Step S57: Construct a visualization interface based on the defect development trend and give an early warning based on the visualization interface.

[0064] Through the tensor field reconstruction of the heat influence coefficient of the production line, the present invention realizes the generation of the spatial distribution matrix of thermal influence, providing a basis for analyzing the thermal influence distribution in the production process. The spectral density conversion of the temperature field reveals the frequency domain characteristics of temperature changes, enabling a deeper understanding of temperature fluctuations. The quantification of the comprehensive thermal stress effect provides clear thermal stress data for subsequent material property analysis. Identifying the material vulnerable areas enhances the ability to predict potential risks in the production line. The Monte Carlo simulation iteration realizes a comprehensive analysis of the defect state transition probability, providing reliable data support for the construction of the material defect evolution path. The combination of the generation of the defect evolution path and time-axis extrapolation provides a scientific basis for predicting the defect development trend. The set prediction step can accurately calculate the future defect growth rate and cumulative damage degree, providing early warning information for maintenance and management in the production process. The constructed visualization interface improves the intuitiveness and practicality of data interpretation, overall improving the intelligent management level of the production line and enhancing the ability to control and predict the service life of materials.

[0065] The present invention also provides a visual management system for a plastic product production line, which is used to execute the visual management method for a plastic product production line as described above. The visual management system for a plastic product production line includes:

[0066] A raw material spectrum analysis module, which is used to scan the raw material spectrum characteristics of plastic products; perform multi-dimensional feature analysis based on the raw material spectrum characteristics to generate multi-dimensional characteristics of raw materials; conduct thermal stability evaluation on the multi-dimensional characteristics of raw materials, and construct a raw material topological unit based on the evaluation results;

[0067] A production line virtual reconstruction module, which is used to collect production line environment data and construct a virtual environment field based on the production line environment data; perform production line virtual reconstruction on the raw material topological unit according to the virtual environment field to generate a virtual visual production line;

[0068] A thermal topology tracking module, which is used to obtain historical records of plastic product production lines; perform thermal topology tracking on the virtual visual production line according to the historical records of plastic product production lines, and construct a production line temperature change field based on the thermal topology;

[0069] A material heat simulation module, which is used to perform material heat simulation based on the production line temperature change field and the raw material topological unit, and collect heat response behaviors corresponding to each heat level; perform long-term impact calculation based on the material heat response behaviors to obtain a production line heat impact coefficient;

[0070] A defect prediction and warning module, which is used to predict the defect development of the raw material topological unit according to the production line heat impact coefficient and the production line temperature field, and give an early visual warning based on the defect development trend.

[0071] Through the raw material spectrum analysis module, the present invention realizes a comprehensive understanding of the characteristics of raw materials. The generated multi-dimensional characteristics of raw materials provide data support for scientific decision-making in production. The thermal stability evaluation provides a reliable basis for judging the adaptability of raw materials. The established raw material topological unit provides an important reference for the overall architecture of the production line. The production line virtual reconstruction module improves the dynamic monitoring ability of environmental impacts by real-time collecting environmental data and constructing a virtual environment field. The generation of the virtual visual production line makes the production process more intuitive. The thermal topology tracking module can systematically analyze historical production line records. The constructed production line temperature change field provides a scientific basis for evaluating temperature fluctuations and their impacts in the production process. The material heat simulation module promotes an in-depth understanding of material quality changes through accurate simulation of material heat. The production line heat impact coefficient obtained from the long-term impact calculation provides practical data support for subsequent production optimization. The defect prediction and warning module realizes the forward-looking management of the production process through the analysis of defect development trends. The early visual warning mechanism provides time guarantee for the identification of potential risks, thereby overall improving the efficiency, stability and product quality of the production line.

[0072] The present invention also provides a computer-readable storage medium storing a computer program, and when the computer program is executed, the visual management method for a plastic product production line as described in any one of the above is implemented.

[0073] The present invention provides an automated implementation of production line management by storing a computer program. The execution of the program makes the spectral feature analysis of raw materials more efficient and accurate. The generation of multi-dimensional characteristics of raw materials provides intuitive data support for subsequent decision-making. The integration of thermal stability evaluation improves the dynamic evaluation ability of the adaptability of raw materials. The constructed raw material topology unit provides a structured information basis for the production line. The real-time collection of stored environmental data and the construction of a virtual environment field improve the reproduction degree of the actual production scenario. The generation of a virtual visual production line enhances the transparency of the production process and the visualization degree of operations. The integration of historical records and the thermal topology tracking function provide a necessary basis for analyzing the temperature impact during the production process. The recording of material heat simulation and the corresponding heat response behavior deepens the understanding of material properties. The derivation of long-term impact calculation and heat impact coefficient promotes the scientificization of thermal management strategies. The integration of the defect prediction and warning module realizes the warning mechanism for potential production problems, thus significantly improving production efficiency and product quality and providing a feasible technical basis for intelligent production management. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a schematic flowchart of the steps of a visual management method for a plastic product production line;

[0075] Figure 2 It is a schematic flowchart of the detailed implementation steps of step S3;

[0076] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0077] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0078] In addition, the attached drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities may be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0079] It should be understood that although terms such as "first" and "second" may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed related items.

[0080] To achieve the above object, please refer to Figures 1 to 2 , a method for visual management of a plastic product production line, comprising the following steps:

[0081] Step S1: Scan the spectral characteristics of the raw materials of the plastic products; perform multi-dimensional feature analysis based on the raw material spectral characteristics to generate multi-dimensional characteristics of the raw materials; perform thermal stability evaluation on the multi-dimensional characteristics of the raw materials, and construct a raw material topological unit based on the evaluation results;

[0082] Step S2: Collect production line environment data, and construct a virtual environment field based on the production line environment data; perform virtual reconstruction of the production line on the raw material topological unit according to the virtual environment field to generate a virtual visual production line;

[0083] Step S3: Obtain the historical records of the plastic product production line; perform thermal topology tracking on the virtual visual production line according to the historical records of the plastic product production line, and construct a production line temperature change field based on the thermal topology;

[0084] Step S4: Perform material heat simulation based on the production line temperature change field and the raw material topological unit, and collect the heat response behaviors corresponding to each heat level; perform long-term influence calculation according to the material heat response behaviors to obtain the production line heat influence coefficient;

[0085] Step S5: Predict the defect development of the raw material topological unit according to the production line heat influence coefficient and the production line temperature field, and perform early visual warning based on the defect development trend.

[0086] The present invention realizes a comprehensive analysis of the characteristics of raw materials by scanning the spectral characteristics of the raw materials, making the understanding of the raw materials deeper. The generated multi-dimensional characteristics of the raw materials provide a solid basis for subsequent thermal stability evaluation. The raw material topological unit constructed based on the evaluation results provides a scientific framework for production. Collecting the production line environment data and constructing the virtual environment field enhance the analysis ability of the production line environment adaptability. The virtual reconstruction of the production line based on the virtual environment field realizes the visualization of the production process, effectively improving the production efficiency. Obtaining the historical records of the plastic product production line and conducting thermal topology tracking make the analysis of the past production conditions more systematic. The production line temperature change field constructed based on the thermal topology provides the necessary data support for understanding the temperature influence in the production process. The record of the material heat simulation and the heat response behavior deeply reveals the physical characteristics of the material during the heating process, providing a specific basis for the long-term influence calculation. The acquisition of the production line heat influence coefficient enhances the scientificity and effectiveness of the production line heat management. The defect development prediction based on the heat influence coefficient and the production line temperature field provides forward-looking guidance for problem prevention in the production process. The early visualization warning mechanism effectively reduces the risks brought by potential defects, thus ensuring the stability of the production line and the quality of the products.

[0087] In an embodiment of the present invention, the visualization management method for the plastic product production line includes the following steps:

[0088] Step S1: Scan the spectral characteristics of the raw materials of the plastic product; perform multi-dimensional feature analysis based on the spectral characteristics of the raw materials to generate multi-dimensional characteristics of the raw materials; conduct thermal stability evaluation on the multi-dimensional characteristics of the raw materials, and construct a raw material topological unit based on the evaluation results;

[0089] In this embodiment, when scanning the spectral characteristics of the raw materials of plastic products, a Fourier Transform Infrared Spectrometer (FTIR) is required. The raw materials of plastic products are placed on the sample stage of the spectrometer, and spectral measurements are carried out in transmission mode or Attenuated Total Reflectance (ATR) mode. The wavelength range of spectral scanning is set to 400 nm to 2500 nm, the spectral resolution is set to 2 cm-1, and the number of scans is 32 times to improve the signal-to-noise ratio. After the obtained spectral data is subjected to baseline correction and denoising processing, the characteristic peaks are extracted by using the Principal Component Analysis (PCA) method, and are matched with the established spectral characteristic database of plastic products to determine the main chemical components of the sample, and the spectral characteristic data of the raw materials are obtained. When performing multi-dimensional characteristic analysis based on the raw material spectral characteristics, the Raman Spectroscopy method is used. The laser light source selects a semiconductor laser with a wavelength of 785 nm, the laser power is set to 50 mW, a Charge-Coupled Device (CCD) detector is used to collect the scattered light signal, and the integration time is set to 10 s to obtain sufficient signal intensity. The collected Raman spectral data is analyzed by the peak fitting method to resolve the molecular vibration mode, and the two-dimensional correlation spectroscopy analysis method (2D-COS) is combined to analyze the interaction between molecules, and the multi-dimensional characteristic data of the raw materials are obtained, including functional group distribution, chemical bond type and its spatial configuration information. When evaluating the thermal stability of the multi-dimensional characteristics of the raw materials, a Thermogravimetric Analyzer (TGA) is used. The sample mass is set to 10 mg, the heating rate is set to 10 °C / min, and it is heated from room temperature to 600 °C in a nitrogen atmosphere. The mass loss curve is recorded, and the 5% mass loss temperature (T_d5) is extracted. T_d5 is compared with the preset thermal degradation standard to judge the thermal stability of the raw materials. Based on the evaluation results, the raw material topological unit is constructed by using Molecular Dynamics Simulation. The simulation conditions are set to the NVT ensemble (constant volume and constant temperature), the temperature is set to 300 K, the calculation step is 1 fs, and the simulation time is 10 ns. Finally, the raw material topological unit structure data is obtained.

[0090] Step S2: Collect the production line environment data, and construct a virtual environment field based on the production line environment data; perform virtual reconstruction of the raw material topological unit according to the virtual environment field to generate a virtual visualization production line;

[0091] In this embodiment, when collecting the production line environment data, environmental sensors are deployed on the plastic product production line, including temperature sensors, humidity sensors, air pressure sensors, air velocity sensors, and infrared thermal imagers. The sampling frequency of the sensors is set to 1 Hz, and the data is transmitted to the central data server in real time through an Industrial Internet of Things (IIoT) gateway. The server uses a Time-Series Database to store the environmental data and combines the Kalman Filter method for data denoising to remove random errors and abnormal data. The processed data is used to construct a virtual environment field. The modeling of the virtual environment field uses the Finite Element Analysis (FEA) method. Taking the collected environmental data as boundary conditions, a three-dimensional numerical model is established, the grid cells are divided, and the steady-state heat conduction is solved to calculate the temperature distribution in each region, and finally a virtual visualization production line is generated.

[0092] Step S3: Obtain the historical records of the plastic product production line; perform thermal topology tracking on the virtual visualization production line according to the historical records of the plastic product production line, and construct a production line temperature change field based on the thermal topology;

[0093] In this embodiment, when obtaining the historical records of the plastic product production line, the production process data is extracted from the database of the Manufacturing Execution System (MES), including production batches, equipment operation parameters, historical temperature distributions, etc. The data format is stored in CSV (Comma-Separated Values) files. Based on the Time Series Analysis method, the historical data is matched with the current virtual visualization production line, the temperature change patterns under similar working conditions are compared, and the thermal topology tracking algorithm (Thermal Topology Tracking) is used to analyze the heat transfer path in the production line. The heat conduction equation is discretized by the Finite Difference Method (FDM) to calculate the temperature gradient field of the production line and generate a production line temperature change field, and the evolution process of the temperature change over time is recorded.

[0094] Step S4: Perform material heating simulation based on the production line temperature change field and the raw material topology unit, and collect the heating response behaviors corresponding to each heating degree; perform long-term influence estimation according to the material heating response behaviors to obtain the production line heat influence coefficient;

[0095] In this embodiment, when performing material heating simulation based on the production line temperature change field and the raw material topological unit, the finite element thermal analysis method (FEM, Finite Element Method) is used to construct a heat model of the plastic product. The model boundary conditions are set as the production line temperature change field data, the material properties are set as the thermal conductivity, specific heat capacity, and density of the topological unit, the calculation time step is set as 0.1 s, the total simulation duration is set as 600 s. Through numerical solution, the temperature response behavior corresponding to each heating degree is recorded. The Arrhenius equation is used to describe the thermal decomposition kinetic process, the decomposition rate at different temperatures is calculated, the long-term influence is extrapolated based on the decomposition rate data, the thermal stress accumulation effect is calculated, and the regression analysis method is used to fit the production line heat influence coefficient to ensure that the fitting error is less than 5%.

[0096] Step S5: Predict the defect development of the raw material topological unit according to the production line heat influence coefficient and the production line temperature field, and perform early visualization warning based on the defect development trend.

[0097] In this embodiment, when predicting the defect development of the raw material topological unit according to the production line heat influence coefficient and the production line temperature field, the finite element damage mechanics method (Damage Mechanics) is used. Based on the thermo-mechanical coupling model (Thermo-Mechanical Coupling Model), the thermal stress distribution is calculated, and combined with the molecular dynamics simulation data, the molecular chain fracture behavior induced by heating is analyzed. The crack propagation model adopts the Paris-Erdogan equation, the crack propagation rate parameters are set as n = 3.5, C = 1.2×10-12, the crack initiation size is set as 0.1 mm. Through the dynamic simulation with a time step of 1 s, the defect evolution trend of the plastic product under different production conditions is predicted. Based on the prediction results, using the color-coded visualization algorithm, a real-time defect warning map is generated on the virtual visualization production line, marking the high-risk areas to achieve early visualization warning.

[0098] Preferably, step S1 includes the following steps:

[0099] Step S11: Perform spectral feature scanning on the plastic product raw material in the wavelength range of 200 nm - 2500 nm to obtain the raw material spectral features;

[0100] Step S12: Analyze the molecular structure of the raw material spectral features to generate the raw material molecular topological structure;

[0101] Step S13: Perform molecular weight distribution statistics based on the raw material molecular topological structure, and calculate the raw material hydrodynamic parameters based on the molecular weight distribution data, where the polydispersity coefficient is set ≤ 1.8;

[0102] Step S14: Evaluate the thermal stability of the plastic product raw materials according to the raw material hydrodynamic parameters, and use the 5% mass loss temperature ≥ 280 °C as the thermal degradation threshold;

[0103] Step S15: Perform matching mapping on the thermal degradation threshold based on the preset additive data, and construct a raw material topological unit based on the mapping result.

[0104] In this embodiment, when performing spectral feature scanning on the raw materials of plastic products, an ultraviolet-visible-near-infrared spectrophotometer (UV-Vis-NIR Spectrophotometer) and a Fourier transform infrared spectrometer (FTIR, Fourier Transform Infrared Spectroscopy) are required. The light source of the ultraviolet-visible-near-infrared spectrophotometer selects a xenon lamp, and the detectors use a photomultiplier tube (PMT, Photomultiplier Tube) and an indium gallium arsenide detector (InGaAs Detector). The wavelength range of spectral scanning is set to 200 nm to 2500 nm, the scanning interval is set to 2 nm, the solution method is used for sample preparation, the solvent selects tetrahydrofuran (THF, Tetrahydrofuran), the sample concentration is set to 5 mg / mL, and a 1 cm quartz cuvette is used for measurement. The Fourier transform infrared spectrometer adopts the attenuated total reflectance mode (ATR, Attenuated Total Reflectance), the crystal material selects diamond (Diamond), the scanning resolution is set to 2 cm-1, and the cumulative scanning times are set to 32 times to improve the signal-to-noise ratio. After the collected spectral data is denoised by baseline correction and Savitzky-Golay Smoothing, the characteristic peak information is extracted to generate the spectral feature data of the raw materials. When performing molecular structure analysis on the spectral features of the raw materials, a Raman spectrometer is required. The laser light source selects a solid-state laser with a wavelength of 532 nm, the laser power is set to 20 mW, a CCD (Charge-Coupled Device, charge-coupled device) detector is used to collect the scattered light signal, the integration time is set to 10 s, and the sample is directly measured in a solid state. Background subtraction is performed before measurement to remove fluorescence interference. The collected Raman spectral data is used to analyze the molecular vibration mode by the peak fitting method, and the two-dimensional correlation spectroscopy (2D-COS, Two-Dimensional Correlation Spectroscopy) method is combined to calculate the co-variation relationship of functional groups. Furthermore, based on molecular dynamics simulation (Molecular Dynamics Simulation), the molecular topological structure of the raw materials is established. The simulation conditions are set to the NPT ensemble (constant pressure and constant temperature), the temperature is set to 300 K, the pressure is set to 1 atm, the calculation step size is set to 1 fs, and the simulation time is set to 10 ns. Finally, a molecular topological structure model of the raw materials is generated. When performing molecular weight distribution statistics based on the molecular topological structure of the raw materials, a gel permeation chromatograph (GPC, Gel Permeation Chromatography) is required. The mobile phase selects tetrahydrofuran (THF), and the flow rate is set to 1.0 mL / min. The detectors used are a refractive index detector (RID) and an ultraviolet detector (UV Detector). The column temperature is set at 35 °C, the sample dissolution concentration is set at 2 mg / mL, the injection volume is set at 100 μL, and calibration is performed using a polystyrene (PS) standard. The measured molecular weight distribution data includes the weight-average molecular weight (Mw), number-average molecular weight (Mn), and polydispersity index. The polydispersity index is calculated as Mw / Mn. When processing the data, the polydispersity index is set to ≤1.8. Based on the measured molecular weight distribution data, the hydrodynamic parameters of the raw material are calculated by combining molecular dynamics simulation, including the cohesive energy density (CED), viscosity coefficient (η), and diffusion coefficient (D). The Poiseuille Equation is used to describe the viscous flow behavior of the fluid during the calculation process, and the hydrodynamic parameters are fitted through regression analysis. When evaluating the thermal stability of the plastic product raw material based on the hydrodynamic parameters of the raw material, a thermogravimetric analyzer (TGA) is required. The sample mass is set at 10 mg, the heating rate is set at 10 °C / min, and it is heated from room temperature to 600 °C under a nitrogen atmosphere, and the mass loss curve is recorded. When processing the data, the 5% mass loss temperature (T_d5) is extracted and compared with the thermal degradation threshold of 280 °C. If T_d5 is greater than 280 °C, it is determined that the thermal stability of the raw material meets the requirements; otherwise, it is considered not to meet the standard. In addition, the derivative thermogravimetry (DTG) is combined to calculate the decomposition rate curve, and the Arrhenius Equation is used to describe the thermal decomposition kinetic behavior to analyze the degradation characteristics of the raw material in a high-temperature environment. When performing matching mapping on the thermal degradation threshold based on the preset additive data, an additive database needs to be established. The database includes categories such as antioxidants, ultraviolet absorbers, and heat stabilizers. Each additive data includes chemical composition, molecular structure, mechanism of action, and applicable temperature range. The thermal degradation threshold is compared with the additive data in the database using a pattern matching algorithm to screen the additive types suitable for the current raw material, and a raw material topological unit is constructed based on the mapping results. The construction of the topological unit uses the molecular dynamics simulation method. The simulation conditions are set as the NVT ensemble (constant volume and constant temperature), the temperature is set at 300 K, the simulation step size is set at 1 fs, and the simulation time is set at 10 ns. The stability of the additive at the molecular level and its interaction with the raw material matrix are calculated to finally determine the optimized raw material topological unit model.

[0105] Preferably, the acquisition of the production line environment data in step S2 and the construction of the virtual environment field based on the production line environment data include:

[0106] Collect the physical environment parameters of the production line;

[0107] Map the physical constraints of the production line through the physical environment parameters of the production line;

[0108] Construct a geometric framework of the production line based on the physical environment parameters of the production line;

[0109] Perform environmental interaction response analysis on the geometric framework of the production line according to the physical constraints of the production line, and construct an environmental response field based on the environmental interaction response data;

[0110] Perform virtual environment field simulation on the geometric framework of the production line according to the environmental response field.

[0111] In this embodiment, when collecting the physical environment parameters of the production line, an environmental sensor network is required for multi-point data collection. The collection points are arranged at the key nodes of the production line, including the extruder, injection molding machine, cooling system, conveying system, and storage unit. The temperature sensor uses a thermocouple, with a measurement range set from -50°C to 400°C and a sampling interval set to 5 seconds. The humidity sensor uses a capacitive humidity sensor, with a measurement range set from 0%RH to 100%RH and a sampling interval set to 10 seconds. The noise sensor uses a MEMS microphone, with a measurement range set from 30dB to 120dB and a sampling interval set to 1 second. The vibration sensor uses a tri-axial accelerometer, with a measurement range set to ±16g and a sampling frequency set to 1kHz. The air quality sensor uses a PID photoionization detector, and the measured parameter includes the concentration of volatile organic compounds (VOC), with a sampling interval set to 30 seconds. The data of all sensors are transmitted to the data processing center through a wireless network. A time synchronization algorithm is used to align the time series of the data of each sensor. When mapping the physical constraints of the production line through the physical environment parameters of the production line, a three-dimensional coordinate system needs to be constructed based on the production line layout information. A LiDAR (Light Detection and Ranging) is used to scan the spatial structure of the production line, with a scanning accuracy set to ±2mm. The point cloud data is registered through the ICP (Iterative Closest Point) algorithm to generate a three-dimensional point cloud model of the production line. Combining the collected environmental parameters such as temperature, humidity, noise, and vibration, parameter mapping is carried out in the point cloud model. The nearest neighbor interpolation method is used to extend the data of discrete sampling points to the entire spatial model to construct a spatial distribution function. The spatial distribution function is used to define the temperature gradient field, humidity gradient field, vibration influence area, and noise influence range. Further, based on the motion characteristics of the production line equipment, the interaction relationship between the equipment motion path and the environmental parameters is calculated to generate a motion-environment constraint matrix. When constructing the geometric framework of the production line based on the physical environment parameters of the production line, computer-aided design (CAD,Use (Computer-Aided Design) software to build a geometric model of the production line. The geometric model includes production links such as extrusion molding, injection molding, mold cooling, product handling, storage and sorting, etc. The geometric structure of each link is obtained by converting point cloud data. Use the Delaunay Triangulation algorithm to perform mesh division on the point cloud data to generate a continuous surface model. The model accuracy is set to 0.5mm. Combine the structural dimensions and motion trajectories of the production line equipment to calculate the spatial relationship between production equipment. Use the Bounding Box method to calculate the minimum distance between equipment. Use Connectivity Analysis to identify the effective channels between equipment. Based on Finite Element Analysis (FEA) to calculate the force on the equipment, determine the stability of the support structure, and finally form a complete production line geometric framework. When performing environmental interaction response analysis on the production line geometric framework according to the production line physical constraints, it is necessary to calculate the impact of environmental factors such as temperature, humidity, vibration, and noise on the production line based on the multi-physics field coupling simulation method. Use Computational Fluid Dynamics (CFD) to simulate the air flow situation. The mesh division uses unstructured meshes, and the mesh size is set to 5mm. The boundary condition is set to an outlet flow rate of 1m, 3 / s. The temperature field simulation uses Heat Transfer Analysis to calculate heat conduction, heat convection, and heat radiation effects. The humidity field simulation uses a multiphase flow model to calculate the air humidity distribution. The vibration analysis uses the Modal Analysis method to calculate the natural frequency and vibration mode of the equipment. The noise propagation uses the Ray Tracing Algorithm to simulate the sound wave propagation path in the production line environment. All environmental interaction response data is processed by the data fusion method. Use the Kalman Filter to remove noise interference and finally generate an environmental response field. When performing virtual environment field simulation on the production line geometric framework according to the environmental response field, it is necessary to build a Digital Twin model. Use a 3D modeling tool to convert the production line geometric framework into an interactive 3D environment. The model material parameters are set to real physical materials, and the lighting parameters are set to meet the standard lighting conditions of the workshop environment. Use the Physics Simulation Engine to calculate the motion trajectory and force on the equipment. Use Real-Time Rendering technology to achieve environmental interaction visualization. The change of environmental parameters is input into the simulation model in real time through the data stream. Use the parallel computing method to improve the simulation efficiency. The simulation accuracy is set to the sub-millimeter level, and finally generate an interactive virtual environment field.

[0112] Preferably, the virtual reconstruction of the production line for the raw material topological unit according to the virtual environment field in step S2 includes:

[0113] Triangulate the raw material topological unit and perform coordinate system projection to obtain vertex coordinates;

[0114] Construct a local surface derivative field based on the vertex coordinates;

[0115] Perform vertex embedding processing on the local surface derivative field to generate a space constraint envelope;

[0116] Perform collision detection on the space constraint envelope and filter collision-free data;

[0117] Based on the collision-free data, perform boundary constraint detection on the space constraint envelope, mark abnormal points and critical points, adjust the projection of the abnormal points, and perform local geometric optimization on the critical points to generate adjusted vertex coordinates;

[0118] Based on the adjusted vertex coordinates, model the flow characteristics of the plastic product to obtain plastic flow simulation data;

[0119] Simulate the plastic product virtual entity according to the plastic flow simulation data;

[0120] Perform temperature-stress coupling analysis on the plastic product virtual entity and identify the forming characteristics of the plastic product;

[0121] Predict the deformation error of the plastic product forming characteristics, and pre-adjust the adjusted vertex coordinates based on the predicted deformation error to generate preventive error vertex coordinates;

[0122] Project the preventive error vertex coordinates onto the virtual environment field to construct a virtual visualization production line.

[0123] In this embodiment, the raw material topological unit is triangulated and coordinate system projection is performed to obtain vertex coordinates. First, the raw material topological unit is discretized based on the Delaunay triangulation method, and the topological unit is divided into several non-overlapping triangular units to ensure that no other vertices are contained inside the circumcircle of any triangle. Then, a three-dimensional coordinate system is established in the Euclidean space, and the vertex coordinates after triangulation are converted to the set coordinate system using a perspective projection model. The projection is calculated using a matrix transformation method. The set projection matrix P is a 4×4 homogeneous coordinate transformation matrix, which includes a rotation matrix R(3×3) and a translation vector T(3×1). Let the original three-dimensional coordinates be (X, Y, Z, 1), then the converted coordinates (X', Y', W') are calculated as follows:

[0124] X' = P11X + P12Y + P13Z + P14, Y' = P21X + P22Y + P23Z + P24, W' = P31X + P32Y + P33*Z + P34. After normalization, the final projection coordinates (Xc, Yc) are obtained: Xc = X' / W', Yc = Y' / W', which are obtained through matrix calculation. According to the vertex coordinates, a local surface derivative field is constructed, and the bilinear interpolation method is used to calculate the local gradient information of each vertex in three-dimensional space. First, the normal vector is constructed using the coordinate information of adjacent vertices, and the calculation formula is N = (P2 - P1) × (P3 - P1), where P1, P2, and P3 are adjacent vertices. The calculated normal vector is used for the differential analysis of the local surface, and then the directional derivative values in the surface derivative field are calculated. The partial derivatives of each vertex are calculated using the finite difference method and stored in the local surface derivative field matrix D. The vertex embedding process is performed on the local surface derivative field to generate a space-limited envelope. First, the curvature value of each vertex is calculated based on the normal vector information in the local derivative field, and the space-limited range of the vertex is determined according to the curvature size. The octree segmentation method is used to partition the entire surface, and a space envelope is created for each sub-interval. The envelope is stored using the BoundingVolume Hierarchy (BVH, boundary volume hierarchy). Each envelope generates a space constraint box through its vertex coordinates and local curvature information and is hierarchically organized. Collision detection is performed on the space-limited envelope, and collision-free data is screened. The AABB (Axis-Aligned Bounding Box) method is used for preliminary collision detection. The minimum circumscribed rectangles of all envelopes are traversed to calculate whether there is an intersection. For envelopes with an intersection, the GJK (Gilbert-Johnson-Keerthi) algorithm is further used to calculate the exact collision points, and the data of the collided envelopes is removed. The collision-free envelopes are screened and stored in the collision-free dataset. Based on the collision-free data, boundary constraint detection is performed on the space-limited envelope, the abnormal points and critical points are marked, and the projection of the abnormal points is adjusted. Local geometric optimization is performed on the critical points to generate adjusted vertex coordinates. The Laplacian Smoothing method is used to adjust the projection of the abnormal points, and the calculation formula is V_i' = (1 - λ)V_i + λ(ΣV_j / n), where V_i is the current abnormal point coordinate, V_j is the adjacent vertex coordinate, n is the number of adjacent vertices, and λ is the smoothing coefficient (0.1 to 0.5), for the critical points, the coordinates are optimized by least squares fitting surface reconstruction to conform to the local geometric constraints. Based on the adjusted vertex coordinates, the flow characteristics of the plastic product are modeled to obtain plastic flow simulation data. First, the finite volume method (FVM) is used to discretize the plastic flow process. The plastic product model is divided into finite volume elements. For each element, the velocity field, pressure field, and temperature field in the Navier-Stokes equations are calculated. The SIMPLE (semi-implicit pressure coupling equation) algorithm is used to solve the pressure-velocity coupling problem, generate plastic flow simulation data and store it. According to the plastic flow simulation data, the plastic product simulation entity is simulated. The VOF (Volume of Fluid) method is used to track the free interface of the plastic flow. Based on the computational fluid dynamics (CFD) software, the flow simulation calculation is carried out. The flow simulation data is converted into three-dimensional grid data, and the marching cubes algorithm is used to generate the plastic product simulation entity model. The temperature-stress coupling analysis of the plastic product simulation entity is carried out, and the forming characteristics of the plastic product are identified. The finite element method (FEM) is used to establish the thermal-mechanical model of the plastic product, and the material parameters are set, such as density (900 - 1400 kg / m. 3 ), thermal conductivity (0.15 - 0.4 W / m·K), elastic modulus (500 - 3000 MPa). The residual stress generated during the cooling process of the plastic product is analyzed, and the warping, shrinkage and other characteristics caused by the temperature gradient during the forming process of the plastic product are identified. The deformation error of the plastic product forming characteristics is predicted, and the adjusted vertex coordinates are pre-controlled based on the predicted deformation error to generate vertex coordinates for preventing errors. The BP neural network is used to predict the deformation error. The input variables include temperature distribution, cooling rate, and stress field data, and the output variable is the deformation deviation value of the plastic product. The training data is constructed based on the plastic product simulation dataset, and the prediction error is controlled within ±5%. Based on the prediction error, the adjusted vertex coordinates are compensated and corrected. The vertex coordinates for preventing errors are projected into the virtual environment field to construct a virtual visual production line. The OpenGL graphics rendering engine is used for three-dimensional visual modeling, and the rendering efficiency is improved based on GPU parallel computing. The ray tracing method is used to achieve high-precision lighting effects. The adjusted vertex coordinates are mapped into the virtual production line environment and real-time dynamic interaction is carried out.

[0125] Preferably, step S3 includes the following steps:

[0126] Step S31: Obtain the historical production line records of plastic products; perform time series decomposition on the historical production line records of plastic products, and identify the periodic temperature fluctuation patterns;

[0127] Step S32: Conduct anomaly detection on the periodic temperature fluctuation patterns, and eliminate the abnormal fluctuations to generate the normal operating temperature characteristics;

[0128] Step S33: Perform spatial registration on the virtual visualization production line based on the normal operating temperature characteristics to generate a thermal mapping grid; perform radial basis function interpolation on the thermal mapping grid, and construct a heat flux vector field;

[0129] Step S34: Extract the thermal topological skeleton of the heat flux vector field; perform time evolution simulation on the thermal topological skeleton to obtain the dynamic thermal topology;

[0130] Step S35: Conduct stability analysis on the dynamic thermal topology, and extract the critical heat flux nodes based on the stability data; construct a production line temperature change field based on the critical heat flux nodes and the thermal mapping grid.

[0131] In this embodiment, when obtaining the historical production line records of plastic products, it is necessary to extract the production data of the past two consecutive years from the production management system. The data storage format is a CSV (Comma-Separated Values) file. Each data point contains information such as timestamp, mold temperature, ambient temperature, cooling time, injection pressure, etc. The time resolution is set to 10 seconds. After the data is loaded, the LOESS (Locally Estimated Scatterplot Smoothing) method is used for smoothing processing to reduce the influence of noise. For the smoothed data, the EMD (Empirical Mode Decomposition) algorithm is used for time series decomposition to split the original time series into multiple Intrinsic Mode Functions (IMFs). Then, the FFT (Fast Fourier Transform) is used to analyze the frequency characteristics of each IMF to identify the periodic temperature fluctuation pattern. The period threshold is set to 24 hours, and the non-periodic components are filtered out. Finally, the periodic temperature fluctuation pattern of the plastic product production line is generated. When performing anomaly detection on the periodic temperature fluctuation pattern and removing the abnormal fluctuations to generate the normal operating temperature characteristics, it is necessary to use the BoxPlot to analyze the distribution of the periodic temperature data, calculate the first quartile (Q1), the third quartile (Q3), and the interquartile range (IQR). The data points exceeding Q3 + 1.5×IQR or below Q1 - 1.5×IQR are marked as abnormal points. The DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is used to further detect local anomalies, with the parameters set as the minimum number of samples 5 and the neighborhood radius 1°C. After removing the abnormal points, the Moving Average Filter method is used to calculate the normal temperature curve, and the window size is set to 60 minutes. Finally, the normal operating temperature characteristics of the production line are obtained. When performing spatial registration on the virtual visualization production line based on the normal operating temperature characteristics to generate the thermal mapping grid, it is necessary to use the vertex coordinates of the three-dimensional production line model and adopt the ICP (Iterative Closest Point) algorithm for spatial registration. The maximum number of iterations is set to 100 times during the registration process, and the convergence threshold is set to 0.001 mm. After the registration is completed, the normal operating temperature characteristic data is mapped to the grid vertices of the three-dimensional production line model to generate a thermal mapping grid. The thermal mapping grid uses 8×8×8 three-dimensional cubic cells, and each cell stores the temperature value. The data format is 16-bit floating-point numbers. To improve data continuity, the RBF (Radial Basis Function) interpolation method is used for temperature interpolation, and the kernel function is set to Gaussian (Gaussian kernel). Finally, a heat flux vector field is constructed. Each vector consists of three-dimensional coordinates and temperature gradients. When extracting the thermal topological skeleton of the heat flux vector field, the hierarchical topological analysis method is needed to calculate the dominant direction of the heat flux vector. The heat flux field is decomposed into rotational and gradient components using the method based on Hodge Decomposition, and only the gradient component is retained to obtain the main heat flux path. Subsequently, principal component analysis (PCA) is performed on the heat flux path to extract the main heat flux direction, and the Dijkstra shortest path algorithm is used to connect the high-temperature region and the low-temperature region on the three-dimensional grid. Finally, a thermal topological skeleton is generated. When performing a time evolution simulation on the thermal topological skeleton, the finite difference method (FDM) is used to calculate the heat diffusion process that changes over time. The time step is set to 1 second, and the total simulation duration is 24 hours. The material thermal conductivity and environmental temperature changes are considered during the simulation to obtain a dynamic thermal topology. When performing a stability analysis on the dynamic thermal topology and extracting critical heat flux nodes based on the stability data, the change rate of the heat flux vector at each time step needs to be calculated, and the Kendall correlation coefficient (Kendall’s Tau) is used to evaluate the stability of the heat flux direction. The stability threshold is set to 0.9. When the change rate of the heat flux direction exceeds 5% and the Kendall correlation coefficient is lower than 0.9, the corresponding heat flux node is marked as a critical heat flux node. After the critical heat flux nodes are extracted, the temperature gradient of the critical nodes is calculated based on the thermal mapping grid, and the Marching Cubes algorithm is used to reconstruct the temperature isosurface in three-dimensional space. Finally, a production line temperature change field is constructed. The production line temperature change field is visualized using the volume rendering method.

[0132] Preferably, the material heating simulation based on the production line temperature change field and the raw material topological unit in step S4, and the acquisition of the heating response behavior corresponding to each heating degree include:

[0133] Perform multi-physical field coupling on the production line temperature change field and the raw material topological unit to obtain a heat-material interaction field;

[0134] Perform material heating simulation on plastic products based on the thermal-material interaction field to obtain simulated heating data;

[0135] Identify the heating distribution of the simulated heating data, and mark the corresponding heating degree of the plastic products based on the heating distribution;

[0136] Analyze the heating response behaviors corresponding to each heating degree according to the corresponding heating degree of the plastic products.

[0137] In this embodiment, when performing multi-physics field coupling on the temperature change field of the production line and the topological unit of the raw material, first obtain the temperature change field data from the thermodynamic model of the production line. The data includes the temperature distribution of each area of the production line. On this basis, establish a thermal-material interaction field model based on the Finite Element Method (FEM). Set the thermal conductivity of the material to 0.5 W / m·K, set the grid size of the simulation area to 1 mm×1 mm×1 mm, and couple through the temperature gradient and the heat conduction characteristics of the material. Use multi-physics field coupling simulation software to perform heat transfer simulation on plastic products. Consider the thermal expansion coefficient of the plastic to be 5×10^-5 per 1℃ increase in temperature. Set the maximum simulation time to 600 seconds, and the time step of each unit in the model to 1 second. The coupling between the temperature field and the material field is updated in real time based on heat conduction and temperature gradient. Finally, obtain the thermal-material interaction field, which includes data such as the temperature distribution, material deformation, and thermal stress of each unit. When performing material heating simulation on plastic products based on the thermal-material interaction field, select a specific type of plastic material. Assume the material is Polypropylene (PP), with a thermal expansion coefficient of 8×10^-5 per 1℃ increase in temperature, a heat capacity of 2000 J / kg·℃, and a thermal conductivity of 0.25 W / m·K. Input these parameters into the multi-physics coupling model, set the temperature of the injection molding machine to 220 °C. During the simulation, use fixed boundary conditions and internal heat flow for simulation. Specifically, in each time step, update the heat-receiving state of the plastic product in real time according to the temperature and thermal stress distributions calculated in the thermal-material interaction field. During the simulation, assume that the outermost layer temperature of the plastic product is 215 °C and the inner layer temperature is 180 °C, and the simulation time step is 5 seconds. Finally, obtain the heat-receiving data of the plastic product at different time points, including the surface and internal temperatures of the material, thermal stress distribution, and strain data. When identifying the heat-receiving distribution of the simulated heat-receiving data, use the clustering analysis method (K-means Clustering) to perform spatial distribution analysis on the temperature data of each time step during the simulation. The temperature data is divided into a low-temperature zone (<150 °C), a medium-temperature zone (150 °C - 200 °C), and a high-temperature zone (>200 °C) according to the value range. Local clustering is performed within each zone, and the center point of the clustering is set as the average temperature of each zone. After clustering, mark each zone of the plastic product. The marking results include the zones within different temperature ranges, the zones of thermal stress distribution, and the strain state of each zone. Further calibrate the heat-receiving degree of each zone. For example, if the temperature is greater than 200 °C and the strain is greater than 5%, it is marked as a high heat-receiving degree zone. When analyzing the heat-receiving response behavior corresponding to each heat-receiving degree based on the plastic product's corresponding heat-receiving degree, use a data-driven response behavior prediction model. Based on historical heat-receiving data, establish a plastic deformation model corresponding to each heat-receiving degree. The model inputs include temperature field data, strain data, and the mechanical properties of the material. Use the linear regression model (Linear Regression) to fit the heat-receiving response within each temperature range. Set the target variable of the fitting as the plastic deformation amount, and the input parameters as the temperature gradient, thermal stress value, and material properties. After fitting, obtain the prediction formula for the heat-receiving response behavior. When the temperature exceeds 200 °C and the stress is greater than the material's yield strength, the plastic product will undergo large deformation, and as the temperature increases, the deformation amount increases. Analyze the deformation characteristics of each heat-receiving zone, and finally obtain the deformation and stress response behavior data of the plastic product under different heat-receiving degrees.

[0138] Preferably, the long-term influence calculation according to the heat-receiving response behavior of the material in step S4 includes:

[0139] Perform time series extrapolation on the heat-receiving response behavior corresponding to each heat-receiving degree to obtain the predicted cumulative thermal effect;

[0140] Perform thermal fatigue mapping based on the predicted cumulative thermal effect to generate a thermal fatigue coefficient;

[0141] Perform matching links on the heat-receiving response behavior based on the thermal fatigue coefficient, and estimate the material life data under each heat-receiving degree based on the matching links;

[0142] Quantify the heat impact based on the estimated data of the material life to obtain the heat impact coefficient of the production line.

[0143] In this embodiment, when performing time series extrapolation on the heat response behaviors corresponding to different degrees of heat exposure, first, historical temperature and material reaction data under different degrees of heat exposure are obtained and subjected to time series analysis to identify the thermal effect fluctuation patterns in each thermal cycle. Then, a time series prediction algorithm is used to extrapolate based on the trend of historical data to obtain the predicted cumulative thermal effect within a certain period in the future. The ARIMA (Autoregressive Integrated Moving Average) model or the long short-term memory network (LSTM) model is used for extrapolation to output the predicted data of the cumulative thermal effect. The prediction period is usually set as a complete operation cycle of the production line, and thus the long-term thermal effect change trend under each degree of heat exposure is obtained. The specific time parameters can be set to be updated hourly or daily according to the requirements of the production cycle. When performing thermal fatigue mapping based on the predicted cumulative thermal effect, first, the predicted cumulative thermal effect is combined with the thermal fatigue model. The thermal fatigue analysis method is adopted, and through the experimental data of historical materials, a thermal fatigue damage model is established. The model combines thermal stress with parameters such as the thermal expansion coefficient and thermal conductivity of the material, and the fatigue damage situation of the material under different thermal effects is obtained through simulation experiments. The predicted cumulative thermal effect is used as the input, and numerical simulation of thermal fatigue is carried out according to the model to generate a thermal fatigue coefficient. The calculation result of the thermal fatigue coefficient represents the fatigue strength and durability of the material under the current thermal effect. During simulation, finite element analysis (FEA) software such as ANSYS or ABAQUS can be used for simulation calculation, and specific thermal fatigue constants can be set according to the actual operation to obtain the specific thermal fatigue coefficient. When performing matching and linking of the heat response behaviors based on the thermal fatigue coefficient, first, the heat response behaviors of the material under different degrees of heat exposure are compared with the generated thermal fatigue coefficient. Data mining techniques such as K-means clustering analysis or support vector machine (SVM) methods are used to perform matching between the heat response behaviors and the fatigue coefficient. By calculation, the behavior patterns related to the thermal fatigue coefficient under each degree of heat exposure are obtained. The matching and linking can identify which heat conditions will cause greater damage to the material by comparing the material property change trends under different thermal effects, and obtain the performance degradation behaviors of the material under different thermal stress states. When estimating the material life prediction data under each degree of heat exposure based on the matching and linking, a service life prediction model is used for estimation. Combining the thermal fatigue coefficient of the material with the mechanical property data of the material, a model based on the damage accumulation theory such as the Miner's rule is used to calculate the material life under each degree of heat exposure. The key parameters during calculation are set as the stress-strain curve, thermal expansion coefficient, fatigue life curve, etc. The thermal fatigue coefficient is combined in the way of damage equivalence to obtain the life prediction values of the material under each heat state. A common implementation method is to use numerical simulation methods combined with experimental data to inversely deduce the predicted service life of the material under different thermal effects. When quantifying the heat impact based on the material life prediction data, first, the impact degree is calculated according to the cumulative thermal effect of each thermal cycle and its impact on the material life.The life prediction data is converted into a heat impact coefficient that quantifies the impact through normalization. The heat impact coefficient represents the heat burden on each component in the production line under the action of thermal stress for a certain period of time. When calculating, a statistical model is combined with a thermal effect model to obtain the heat impact coefficient. Specifically, according to the relationship between the magnitude of the cumulative thermal effect and its impact on life, a quantitative calculation is performed through an integration method to obtain the total heat impact coefficient of the production line.

[0144] Preferably, step S5 includes the following steps:

[0145] Step S51: Reconstruct the tensor field of the heat impact coefficient of the production line to obtain the spatial distribution matrix of the thermal impact;

[0146] Step S52: Perform spectral density conversion on the temperature field of the production line to obtain the temperature frequency domain characteristics;

[0147] Step S53: Quantify the action of thermal stress on the spatial distribution matrix of the thermal impact and the temperature frequency domain characteristics to obtain the comprehensive action of thermal stress;

[0148] Step S54: Identify the material vulnerable areas of the raw material topological units;

[0149] Step S55: Perform Monte Carlo simulation iteration N = 10,000 times on the material vulnerable areas based on the comprehensive action of thermal stress, calculate the defect state transition probability P(T|T-1), and generate a defect evolution path data matrix based on the state transition matrix S(t,t+Δt) to generate a defect evolution path, where T is the number of iterations of the Monte Carlo simulation, that is, the state number of the current simulation, t is the physical time parameter, and Δt is the increment of physical time, that is, the time step in each simulation calculation during the defect state evolution process;

[0150] Step S56: Extrapolate the defect evolution path along the time axis, set the prediction step T = 50 time units, and calculate the defect growth rate G(t) and the cumulative damage degree D(t) at the future T moment to obtain the defect development trend;

[0151] Step S57: Construct a visualization interface based on the defect development trend and give an early warning based on the visualization interface.

[0152] In this embodiment, when performing tensor field reconstruction on the heat influence coefficient of the production line, first collect the heat influence data of each area of the production line, including the cumulative value of the thermal effect and its distribution in each part of the production line. Use the finite element analysis (FEA) method to establish a heat field model of the production line. Regard the heat influence coefficient as a tensor field, process each heat distribution area, and reconstruct the spatial distribution of the heat influence by solving the heat conduction equation to obtain the heat influence spatial distribution matrix. The elements of this matrix represent the cumulative influence degree of the thermal effect at a specific position. The parameters used in the reconstruction include the geometric structure of the production line, material properties, and thermal conductivity, etc. Calculate the heat influence amount of each area through numerical calculation. When performing spectral density conversion on the temperature field of the production line, first obtain the real-time temperature data of each monitoring point on the production line and perform Fourier Transform on it to obtain the frequency domain characteristics of the temperature change over time. Conduct spectral density analysis of the temperature signal, use the fast Fourier transform (FFT) algorithm to convert the temperature data from the time domain to the frequency domain, calculate the power spectral density of the temperature signal, and extract the main frequency components of the temperature change. Specifically, by setting the sampling frequency to 10 Hz and the sampling duration to 10 minutes, the frequency domain characteristics can be expressed as the temperature fluctuation intensity at each frequency. Further use these characteristic values to identify the periodic characteristics of the temperature fluctuation of the production line. When quantifying the thermal stress effect on the heat influence spatial distribution matrix and temperature frequency domain characteristics, first, according to the heat influence coefficient matrix and temperature frequency domain characteristics, combined with the thermal expansion coefficient and thermal conductivity of the material, use the thermal stress formula to quantify the thermal stress of each area. The calculation process includes coupling the heat influence matrix with the frequency domain characteristics of the temperature field to obtain the thermal stress response of each area. By setting a thermal stress threshold of 100 MPa, calibrate the thermal stress action area under this threshold. When quantifying, properties such as the linear expansion coefficient and elastic modulus of the material need to be considered, and the stress distribution of each production line area under the action of the thermal effect is obtained through analytical solution. When identifying the material vulnerable areas of the raw material topology units, first analyze the heat history data of each topology unit, combine the thermal stress effect, determine the thermal fatigue strength of the material, use the finite element model to simulate each topology unit to obtain the relationship between thermal stress and material reaction, and identify the vulnerable areas of the material based on the thermal fatigue analysis results. The vulnerable areas usually appear in the parts with stress concentration or faster material aging. Use a certain calibration standard, such as the damaged area when the temperature change exceeds 50 °C, to obtain the area with uneven heat absorption, and calibrate these areas as vulnerable areas. When performing Monte Carlo simulation on the material vulnerable areas based on the comprehensive action of thermal stress, first establish a damage state model for each vulnerable area, set the initial damage state to zero, and use the Monte Carlo method to randomly simulate the future damage state of each vulnerable area. The simulation process reacts to the change of thermal stress according to each time step (for example, every 10 seconds), and through iteration 10,000 times, simulate the transition probability of the material from the initial state to the damaged state under different thermal stress states, calculate the defect state transition probability P(T|T-1), obtain the defect evolution path at different time steps through Monte Carlo simulation, generate the defect evolution path data matrix through the state transition matrix S(t,t+Δt), and finally obtain the defect evolution path of each vulnerable area, where T is the number of iterations of the Monte Carlo simulation, that is, the current state number reached by the simulation, t is the physical time parameter, and Δt is the increment of physical time. When extrapolating the defect evolution path along the time axis, first, according to the simulated defect evolution path, set a prediction step T of 50 time units, and extrapolate the defect state at each time step. Use historical data and the evolution path to calculate the future defect growth rate G(t) and cumulative damage degree D(t). After setting the step size, predict the defect growth in each vulnerable area in the future T time. Extrapolate the defect progress in each area through a numerical model, calculate the defect growth rate at each moment, and accumulate to obtain the cumulative damage degree, and finally obtain the defect development trend of each area. When constructing a visualization interface based on the defect development trend, first visualize the defect development trend data through data processing tools (such as Matplotlib or Tableau in Python), design the thermal stress distribution map of the production line, and mark the defect development of each vulnerable area on the interface, and display the defect growth rate, cumulative damage degree, and thermal stress intensity in real time. Show the state changes of each component through the interface, set threshold alarms, and automatically issue early warnings when the defects in some areas reach the critical value, prompting relevant personnel to perform maintenance or replace components. The visual design of the entire interface needs to be adjusted based on the specific layout and process requirements of the production line.,

[0153] The present invention also provides a visual management system for a plastic product production line, which is used to execute the visual management method for a plastic product production line as described above. The visual management system for a plastic product production line includes:

[0154] A raw material spectrum analysis module, which is used to scan the raw material spectrum characteristics of plastic products; perform multi-dimensional feature analysis based on the raw material spectrum characteristics to generate raw material multi-dimensional characteristics; perform thermal stability evaluation on the raw material multi-dimensional characteristics, and construct a raw material topological unit based on the evaluation results;

[0155] A production line virtual reconstruction module, which is used to collect production line environment data and construct a virtual environment field based on the production line environment data; perform production line virtual reconstruction on the raw material topological unit according to the virtual environment field to generate a virtual visual production line;

[0156] A thermal topology tracking module, which is used to obtain historical records of the plastic product production line; perform thermal topology tracking on the virtual visual production line according to the historical records of the plastic product production line, and construct a production line temperature change field based on the thermal topology;

[0157] A material heat simulation module, which is used to simulate the heat of the material based on the temperature change field of the production line and the raw material topology unit, and collect the heat response behaviors corresponding to each heat level; and calculate the long-term influence based on the material heat response behaviors to obtain the heat influence coefficient of the production line.

[0158] A defect prediction and warning module, which is used to predict the defect development of the raw material topology unit according to the heat influence coefficient of the production line and the temperature field of the production line, and perform early visual warning based on the defect development trend.

[0159] The present invention realizes a comprehensive understanding of the characteristics of raw materials through the raw material spectrum analysis module. The generated multi-dimensional characteristics of the raw materials provide data support for scientific decision-making in production. The thermal stability evaluation provides a reliable basis for judging the adaptability of raw materials. The established raw material topology unit provides an important reference for the overall architecture of the production line. The production line virtual reconstruction module improves the dynamic monitoring ability of environmental impacts by collecting environmental data in real time and constructing a virtual environment field. The generation of the virtual visual production line makes the production process more intuitive. The thermal topology tracking module can systematically analyze the historical production line records. The constructed temperature change field of the production line provides a scientific basis for evaluating the temperature fluctuations and their impacts during the production process. The material heat simulation module promotes an in-depth understanding of the material quality changes through the accurate simulation of the material heat. The heat influence coefficient of the production line obtained by the long-term influence calculation provides practical data support for subsequent production optimization. The defect prediction and warning module realizes the forward-looking management of the production process through the analysis of the defect development trend. The early visual warning mechanism provides time guarantee for the identification of potential risks, thereby overall improving the efficiency, stability and product quality of the production line.

[0160] The present invention also provides a computer-readable storage medium storing a computer program, and when the computer program is executed, it implements the plastic product production line visualization management method described in any one of the above.

[0161] The present invention provides an automated implementation of production line management by storing a computer program. The execution of the program makes the spectral feature analysis of raw materials more efficient and accurate, generates multi-dimensional characteristics of raw materials, provides intuitive data support for subsequent decision-making, the integration of thermal stability assessment improves the dynamic assessment ability of the adaptability of raw materials, the constructed raw material topological unit provides a structured information basis for the production line, the real-time collection of stored environmental data and the construction of a virtual environment field improve the reproduction degree of the actual production scenario, the generation of a virtual visual production line enhances the transparency of the production process and the visualization degree of operations, the integration of historical records and the thermal topology tracking function provide a necessary basis for analyzing the temperature impact in the production process, the recording of material heat simulation and the corresponding heat response behavior deepen the understanding of material properties, the derivation of long-term impact calculation and heat impact coefficient promotes the scientificization of thermal management strategies, and the integration of the defect prediction and warning module realizes the warning mechanism for potential production problems, thus significantly improving production efficiency and product quality and providing a feasible technical basis for intelligent production management.

[0162] Therefore, in any aspect, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed within the present invention.

[0163] The above description is only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A visualization management method for a plastic product production line, characterized in that, Including the following steps: Step S1: Scan the spectral characteristics of the raw materials of plastic products; Perform multi-dimensional feature analysis based on the raw material spectral characteristics to generate raw material multi-dimensional characteristics; conduct thermal stability evaluation on the raw material multi-dimensional characteristics, and construct a raw material topological unit based on the evaluation results; Step S2: Collect production line environment data, and construct a virtual environment field based on the production line environment data; perform virtual reconstruction of the production line on the raw material topological unit according to the virtual environment field to generate a virtual visual production line; Step S3: Obtain the historical production line records of plastic products; perform thermal topology tracking on the virtual visual production line according to the historical production line records of plastic products, and construct a production line temperature change field based on the thermal topology; Step S4: Perform material heating simulation based on the production line temperature change field and the raw material topological unit, and collect the heating response behaviors corresponding to each heating degree; perform long-term influence calculation based on the material heating response behaviors to obtain the production line heat influence coefficient; Step S5: Predict the defect development of the raw material topological unit according to the production line heat influence coefficient and the production line temperature field, and perform early visual warning based on the defect development trend.

2. The visual management method for a plastic product production line according to claim 1, characterized in that Step S1 includes the following steps: Step S11: Scan the spectral characteristics of the plastic product raw materials in the wavelength range of 200nm - 2500nm to obtain the raw material spectral characteristics; Step S12: Perform molecular structure analysis on the raw material spectral characteristics to generate the raw material molecular topological structure; Step S13: Perform molecular weight distribution statistics based on the raw material molecular topological structure, and calculate the raw material hydrodynamic parameters based on the molecular weight distribution data, where the polydispersity coefficient is set to ≤1.8; Step S14: Perform thermal stability evaluation on the plastic product raw materials according to the raw material hydrodynamic parameters, with the 5% mass loss temperature ≥280°C as the thermal degradation threshold; Step S15: Perform matching mapping on the thermal degradation threshold based on the preset additive data, and construct a raw material topological unit based on the mapping results.

3. The visualization management method for the plastic product production line according to claim 1, wherein, The collection of production line environment data described in Step S2 and the construction of a virtual environment field based on the production line environment data include: Collect production line physical environment parameters; Map the production line physical constraints through the production line physical environment parameters; Construct a production line geometric framework based on the production line physical environment parameters; Perform environmental interaction response analysis on the production line geometric framework according to the production line physical constraints, and construct an environmental response field based on the environmental interaction response data; Perform virtual environment field simulation on the production line geometric framework according to the environmental response field.

4. The visual management method for a plastic product production line according to claim 1, characterized in that The virtual reconstruction of the production line on the raw material topological unit described in Step S2 includes: Perform triangulation on the raw material topological unit and perform coordinate system projection to obtain vertex coordinates; Construct a local surface derivative field according to the vertex coordinates; Perform vertex embedding processing on the local surface derivative field to generate a space-limited envelope; Perform collision detection on the space-limited envelope and filter collision-free data; Perform boundary constraint detection on the space-limited envelope based on the collision-free data, mark the abnormal points and critical points, project and adjust the abnormal points, and perform local geometric optimization on the critical points to generate adjusted vertex coordinates; Model the flow characteristics of plastic products based on the adjustment of vertex coordinates to obtain plastic flow simulation data; Simulate the plastic product simulation entity based on plastic flow simulation data; Conduct temperature-stress coupling analysis on plastic product simulation entities and identify the molding characteristics of plastic products; Predict the deformation error of the molding features of plastic products, and pre-regulate the vertex coordinates based on the predicted deformation error to generate error-preventing vertex coordinates; The error-preventing vertex coordinates are projected onto the virtual environment field to construct a virtual visualization production line.

5. The visualization management method for the plastic product production line according to claim 1, characterized in that Step S3 includes the following steps: Step S31: Obtain historical records of plastic product production lines; perform time series decomposition on the historical records of plastic product production lines and identify periodic temperature fluctuation patterns; Step S32: performing abnormality detection on the periodic temperature fluctuation pattern and eliminating abnormal fluctuations to generate a normal operating temperature feature; Step S33: performing spatial registration of the virtual visualization production line based on the normal operating temperature characteristics to generate a thermal mapping grid; performing radial basis function interpolation on the thermal mapping grid to construct a heat flow vector field; Step S34: extracting the thermal topology skeleton of the heat flow vector field; performing time evolution simulation on the thermal topology skeleton to obtain a dynamic thermal topology; Step S35: Perform stability analysis on the dynamic thermal topology and extract critical heat flow nodes based on the stability data; construct the production line temperature variation field based on the critical heat flow nodes and the thermal mapping grid.

6. The visualization management method for a plastic product production line according to claim 1, characterized in that, The material heating simulation described in step S4 based on the production line temperature change field and the raw material topology unit, and the collection of heating response behaviors corresponding to each heating degree include: Multi-physics field coupling is performed on the production line temperature change field and the raw material topology unit to obtain the heat-material interaction field; Perform material heating simulation on plastic products based on the heat-material interaction field to obtain simulated heating data; Identify the heat distribution of the simulated heat data and mark the corresponding heat degree of the plastic products based on the heat distribution; Analyze the thermal response behavior corresponding to each heating degree according to the corresponding heating degree of the plastic product.

7. The visual management method for a plastic product production line according to claim 1, wherein The long-term impact estimation based on the thermal response behavior of the material described in step S4 includes: The time series of thermal response behaviors corresponding to each heating degree are extrapolated to obtain the predicted cumulative thermal effect; Conduct thermal fatigue mapping based on the predicted cumulative thermal effect to generate thermal fatigue coefficients; Match and link the thermal response behavior based on the thermal fatigue coefficient, and calculate the material life estimation data under various heating levels based on the matching link; The heat impact is quantified based on the material life estimation data to obtain a heat impact coefficient of the production line.

8. The visualization management method for a plastic product production line according to claim 1, wherein, Step S5 includes the following steps: Step S51: reconstructing the tensor field of the heat impact coefficient of the production line to obtain a heat impact spatial distribution matrix; Step S52: Perform spectral density conversion on the production line temperature field to obtain temperature frequency domain characteristics; Step S53: quantifying the thermal stress effect on the heat impact spatial distribution matrix and the temperature frequency domain characteristics to obtain the comprehensive thermal stress effect; Step S54: identifying material vulnerable areas of the raw material topological unit; Step S55: Conduct Monte Carlo simulation iteration N = 10,000 times on the vulnerable area of the material based on the comprehensive effect of thermal stress, calculate the defect state transition probability P(T|T-1), and generate the defect evolution path data matrix based on the state transition matrix S(t, t+Δt) to generate the defect evolution path, where T is the iteration step of the Monte Carlo simulation, that is, the current state number reached by the simulation, t is the physical time parameter, and Δt is the increment of physical time, that is, the time step in each simulation calculation during the defect state evolution process; Step S56: Extrapolate the defect evolution path along the time axis, set the prediction step T = 50 time units, and calculate the defect growth rate G(t) and the cumulative damage degree D(t) at the future time T to obtain the defect development trend; Step S57: Build a visualization interface based on the defect development trend and give an early warning based on the visualization interface.

9. A visual management system for a plastic product production line, characterized in that, A plastic product production line visualization management system for executing the plastic product production line visualization management method as described in claim 1, the plastic product production line visualization management system comprising: A raw material spectrum analysis module for scanning the raw material spectrum characteristics of plastic products; performing multi-dimensional feature analysis based on the raw material spectrum characteristics to generate multi-dimensional characteristics of the raw materials; performing thermal stability evaluation on the multi-dimensional characteristics of the raw materials, and constructing a raw material topology unit based on the evaluation results; A production line virtual reconstruction module for collecting production line environment data and constructing a virtual environment field based on the production line environment data; performing production line virtual reconstruction on the raw material topology unit according to the virtual environment field to generate a virtual visualization production line; A thermal topology tracking module for obtaining historical plastic product production line records; performing thermal topology tracking on the virtual visualization production line according to the historical plastic product production line records, and constructing a production line temperature change field based on the thermal topology; A material heat reception simulation module for performing material heat reception simulation based on the production line temperature change field and the raw material topology unit, and collecting the heat reception response behaviors corresponding to each heat reception degree; performing long-term influence calculation based on the material heat reception response behaviors to obtain the production line heat influence coefficient; A defect prediction and early warning module for predicting the defect development of the raw material topology unit according to the production line heat influence coefficient and the production line temperature field, and giving an early visualization warning based on the defect development trend.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program, when executed, implements the plastic product production line visualization management method as described in any one of claims 1 to 8.

Images