A packaging parameter monitoring method and system for packaging equipment

Abstract

The present application relates to the field of data processing technology, and discloses a method and system for monitoring packaging parameters for packaging equipment. The method includes: collecting and filtering parameters through a sensor network; extracting features based on real-time data to build a model to obtain quality indicators; using dynamic thresholds to detect deviations and generate warnings; adjusting parameters through adaptive control; generating quality traceability marks based on optimized parameters and actual effects; analyzing parameter trends to identify wear indexes and formulate maintenance plans. The present application realizes all-round monitoring, adaptive adjustment, quality traceability, and predictive maintenance of the packaging process, improving packaging quality stability and equipment utilization.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a packaging parameter monitoring method and system for packaging equipment. Background Art

[0002] In the operation of traditional packaging equipment, parameter monitoring usually uses a single sensor to collect data at a specific location, and the operator judges the equipment status and adjusts the parameters based on experience. Most existing packaging equipment monitoring systems use a threshold alarm mechanism. When a parameter exceeds the preset range, an alarm is triggered to remind the operator to check and handle it. This monitoring method can meet basic production needs when the equipment operating parameters are stable and the product specifications are single. Some advanced packaging equipment has introduced automatic control technology to achieve automatic adjustment of parameters through closed-loop control to reduce manual intervention. At the same time, some equipment manufacturers have also begun to use data acquisition and monitoring control systems (SCADA) to record equipment operation data to provide basic data support for quality control and equipment maintenance.

[0003] However, the existing technology has obvious shortcomings. First, traditional single parameter monitoring is difficult to detect the mutual influence and correlation anomalies between parameters, which makes it difficult to identify complex faults in a timely manner; second, the alarm mechanism with fixed thresholds cannot adapt to the needs of different product specifications and changes in working conditions, resulting in false alarms and missed alarms; third, the lack of intelligent analysis means makes it impossible to extract valuable information from massive monitoring data, and equipment adjustment and maintenance mainly rely on the experience of operators; fourth, traditional quality control usually conducts random inspections after production is completed, and it is impossible to achieve full quality traceability; finally, equipment maintenance mostly adopts regular maintenance or fault repair mode, which makes it difficult to optimize maintenance strategies according to the actual status of the equipment, resulting in waste of resources or equipment damage caused by untimely maintenance. Summary of the invention

[0004] The present application provides a packaging parameter monitoring method and system for packaging equipment, which is used to solve the technical problem that multi-parameter collaborative analysis, dynamic threshold adjustment, intelligent abnormality identification, full quality traceability and predictive maintenance cannot be achieved in the existing packaging equipment parameter monitoring, and realize all-round monitoring, adaptive adjustment, quality traceability and predictive maintenance of the packaging process, thereby improving the stability of packaging quality and equipment utilization.

[0005] In a first aspect, the present application provides a packaging parameter monitoring method for a packaging device, the packaging parameter monitoring method for a packaging device comprising: collecting pressure, temperature, displacement, speed and tension parameters of the packaging device through a sensor network, filtering the parameters, and obtaining real-time data of the packaging process; based on the real-time data of the packaging process, extracting parameter features through data mining, building a parameter model and a standard library, and obtaining a packaging quality evaluation index; according to the real-time data of the packaging process and the packaging quality evaluation index, detecting parameter deviations through a dynamic threshold to obtain packaging abnormality warning information;

[0006] Based on the abnormal packaging warning information, the optimal parameter combination is calculated through adaptive control, and the packaging machinery execution unit is adjusted to obtain optimized packaging parameters; according to the optimized packaging parameters and the actual packaging effect, the packaging quality traceability mark is obtained through strength detection and data encryption; based on the packaging quality traceability mark, the equipment wear index is identified through parameter trend analysis, the component replacement cycle is generated, and the equipment maintenance plan is obtained.

[0007] In a second aspect, the present application provides a packaging parameter monitoring system for a packaging device, the packaging parameter monitoring system for a packaging device comprising:

[0008] The filtering module is used to collect the pressure, temperature, displacement, speed and tension parameters of the packaging equipment through the sensor network, filter the parameters and obtain the real-time data of the packaging process;

[0009] A construction module is used to extract parameter features through data mining based on the real-time data of the packaging process, build a parameter model and a standard library, and obtain a packaging quality evaluation index;

[0010] A detection module, used to detect parameter deviations through dynamic thresholds according to the real-time data of the packaging process and the packaging quality evaluation index, and obtain packaging abnormality warning information;

[0011] A calculation module, used to calculate the best parameter combination through adaptive control based on the abnormal packaging warning information, adjust the packaging machine execution unit, and obtain optimized packaging parameters;

[0012] An encryption module, used to obtain a packaging quality traceability mark through strength detection and data encryption according to the optimized packaging parameters and actual packaging effect;

[0013] The identification module is used to identify the equipment wear index through parameter trend analysis based on the packaging quality traceability mark, generate a component replacement cycle, and obtain an equipment maintenance plan.

[0014] The third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the computer device executes the above-mentioned packaging parameter monitoring method for a packaging device.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned packaging parameter monitoring method for a packaging device.

[0016] In the technical solution provided by the present application, the multi-dimensional parameters of the packaging equipment are collected through the sensor network and real-time filtering is performed, which overcomes the limitations of traditional single parameter monitoring, realizes comprehensive monitoring of the operating status of the packaging equipment, and greatly improves the accuracy and integrity of data collection. Data mining is carried out based on real-time data of the packaging process, parameter features are extracted, parameter models and standard libraries are constructed, and the complex correlation between parameters is modeled by artificial intelligence algorithms, so that the system can automatically identify parameter features under different working conditions, establish an objective packaging quality evaluation system, and get rid of the subjectivity and uncertainty of traditional reliance on manual experience judgment. Through the dynamic threshold detection parameter deviation, the system can automatically adjust the monitoring threshold according to different packaging conditions and packaging characteristics, avoid the false alarm and missed alarm problems caused by fixed thresholds, improve the accuracy and sensitivity of abnormal detection, and greatly reduce the missed detection rate of quality problems. Based on the adaptive control system of abnormal warning information, combined with the machine learning algorithm to calculate the optimal parameter combination in real time, realize the intelligent automatic adjustment of parameters, reduce manual intervention, and improve packaging efficiency and consistency. The application of strength detection and data encryption technology constructs an unalterable packaging quality traceability mark, ensures the reliability and integrity of quality data, and provides a solid foundation for full-process quality traceability. Parameter trend analysis based on quality traceability identification, identification of equipment wear patterns through deep learning algorithms, prediction of the remaining life of components, and transformation of traditional planned maintenance into predictive maintenance not only avoids the waste of resources caused by premature replacement of components, but also prevents production interruptions caused by sudden equipment failures, significantly improving equipment reliability and production efficiency. By building a complete packaging parameter monitoring method, the artificial intelligence algorithm is deeply integrated with the packaging equipment control, forming a closed-loop control in the links of data collection, feature extraction, anomaly detection, parameter optimization, quality traceability and predictive maintenance, and realizing intelligent monitoring and optimization of the packaging process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 A schematic diagram of an embodiment of a method for monitoring packaging parameters of a packaging device in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of an embodiment of a packaging parameter monitoring system for packaging equipment in an embodiment of the present application;

[0020] Figure 3 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present application embodiment provides a method and system for monitoring the packaging parameters of packaging equipment. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of a packaging parameter monitoring method for a packaging device in an embodiment of the present application includes:

[0023] Step S101, collecting the pressure, temperature, displacement, speed and tension parameters of the packaging equipment through the sensor network, filtering the parameters, and obtaining real-time data of the packaging process;

[0024] Step S102: Based on the real-time data of the packaging process, extract parameter features through data mining, build a parameter model and a standard library, and obtain packaging quality evaluation indicators;

[0025] Step S103: According to the real-time data of the packaging process and the packaging quality evaluation index, the parameter deviation is detected by the dynamic threshold to obtain the packaging abnormality warning information;

[0026] Step S104: Based on the abnormal packaging warning information, the optimal parameter combination is calculated through adaptive control, and the packaging machine execution unit is adjusted to obtain the optimized packaging parameters;

[0027] Step S105: According to the optimized packaging parameters and the actual packaging effect, a packaging quality traceability mark is obtained through strength detection and data encryption;

[0028] Step S106: Based on the packaging quality traceability mark, the equipment wear index is identified through parameter trend analysis, the component replacement cycle is generated, and the equipment maintenance plan is obtained.

[0029] It is understandable that the execution subject of the present application may be a packaging parameter monitoring system for packaging equipment, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0030] Specifically, the key parameters of the packaging equipment are collected through the sensor network. Various sensors, including pressure sensors, temperature sensors, displacement sensors, speed sensors and tension sensors, are installed at key positions of the packaging equipment to form a data acquisition network. These sensors continuously collect analog signals during the operation of the equipment, and then convert the analog signals into digital signals through a high-precision analog-to-digital converter. In practical applications, the monitoring accuracy required for packaging equipment in different working stages is different, so the sampling frequency will be dynamically adjusted. For example, in the initial tightening stage of the strapping belt, the sampling frequency of the tension sensor can be increased to 200Hz, and reduced to 50Hz in the stable operation stage to balance the data volume and monitoring accuracy. The collected raw data has various interferences and noises, which need to be processed by filtering algorithms. A bandpass filter is applied to remove high-frequency noise and retain the effective signal frequency band. Then, the median filtering method is used to process abnormal peaks and smooth data fluctuations. The processed parameter data is transmitted to the central processing unit in real time through industrial Ethernet to form real-time data of the packaging process. Based on the acquired real-time data of the packaging process, parameter features are extracted through data mining technology. Time domain analysis and frequency domain conversion are performed to extract parameter fluctuation feature points and change trend lines. Time domain analysis identifies the peak, valley and fluctuation frequency of the parameters, while frequency domain conversion converts time domain data into frequency domain data through Fourier transform and analyzes the proportion of different frequency components. Cluster analysis technology is used to classify the parameter performance under different packaging conditions to form a packaging parameter feature vector. Cluster analysis groups similar parameter performances into a group and generates a feature vector. A correlation matrix is ​​established between the parameters to determine the key influencing parameters and secondary influencing parameters. The correlation matrix is ​​constructed by calculating the correlation coefficient between the parameters, and parameters with a correlation coefficient higher than 0.8 are identified as strongly correlated parameters. The key influencing parameters are compared with historical packaging data to construct a multidimensional parameter standard reference interval, and differentiated parameter standard libraries are set for different types of packaging materials to form a packaging process parameter mapping table. Finally, the parameter standard library and the packaging process parameter mapping table are integrated through the weight distribution method to generate packaging quality evaluation indicators.

[0031] Parameter deviation is detected by dynamic threshold to obtain abnormal packing warning information. A multi-threaded parallel computing architecture is used to divert and process the real-time data of the packing process to form an independent parameter monitoring channel. For pressure parameters, the dynamic pressure fluctuation range is calculated based on the sliding window. The sliding window width is automatically adjusted according to the packing speed. A narrow window is used for fast packing and a wide window is used for slow packing. For temperature, displacement, speed and tension parameters, the parameter deviation detection threshold is set in combination with the packing quality evaluation index to calculate the parameter deviation value. The parameter deviation value is compared with the standard interval in the packing quality evaluation index to calculate the parameter deviation degree and form a parameter abnormality degree matrix. Based on the parameter abnormality degree matrix, the abnormal parameters are weighted and scored to distinguish single parameter abnormalities from multi-parameter combination abnormalities to obtain the packing abnormality type mark. According to the packing abnormality type mark and parameter deviation degree, the abnormal priority and processing strategy are determined to generate packing abnormality warning information. Based on the packing abnormality warning information, the optimal parameter combination is calculated by adaptive control technology. The packing abnormality warning information is parsed, the abnormal parameter type and deviation degree are extracted, and the parameter adjustment priority table is established. A mathematical model of the packing process is constructed, the packing quality is taken as the objective function, and the parameter adjustment range is limited by constraint conditions. For small parameter fluctuations, the proportional integral differential control is used to calculate the adjustment increment. For medium deviations, the model is used to predict and calculate the future parameter change trend, and an advance adjustment strategy is generated. The parameter fine-tuning scheme and the advance adjustment strategy are integrated into a parameter adjustment instruction set, which is transmitted to the packaging machinery execution unit through the industrial control network to implement smooth transition adjustment and form optimized packaging parameters.

[0032] According to the optimized packaging parameters and the actual packaging effect, the packaging quality traceability mark is generated through strength detection and data encryption. The physical performance test of the packaged materials completed with the optimized packaging parameters is carried out to obtain the sealing strength, compression resistance and shear resistance data to form the packaging strength index set. The packaging strength index set is compared and analyzed with the packaging quality evaluation index, the packaging quality score is calculated, and the packaging quality evaluation result is generated. The optimized packaging parameters, the real-time data of the packaging process and the packaging quality evaluation results are integrated in time series to form a packaging process data chain. The packaging process data chain is encrypted in blocks through the hash algorithm to generate an unalterable packaging quality data block. A unique digital signature is generated for each packaging quality data block, the corresponding packaging material is associated, and a data block index table is established. Based on the data block index table, a unique identification code is assigned to the packaging material to form a packaging quality traceability mark. Based on the packaging quality traceability mark, the equipment wear index is identified through parameter trend analysis, and the component replacement cycle is generated. The historical parameter data and abnormal records are extracted from the packaging quality traceability mark, and the equipment operation status time series diagram is constructed. The parameter attenuation slope and fluctuation frequency are identified through trend analysis to obtain the health status index of each key component of the equipment. Based on the correlation analysis between the health status index and historical maintenance records, a component wear degradation model is established to calculate the equipment wear index. Based on the equipment wear index and production plan data, the remaining service life of each component is predicted and the component replacement cycle table is determined. The component replacement cycle table is integrated with the maintenance resource data to generate the equipment maintenance time window and spare parts demand list. The maintenance time window is sorted and adjusted through the optimization algorithm to balance the maintenance cost and equipment reliability and form an equipment maintenance plan.

[0033] For example, after a food packaging company used a packaging device that had been running continuously for 12 days, it was found through the sensor network that the fluctuation amplitude of the strapping tension increased. The tension sensor data showed that the normal value should be 400N±20N, but the actual value frequently fluctuated between 350N and 460N. The system immediately conducted data mining and analysis and found that the tension fluctuation was 87% correlated with the temperature increase of the main drive motor. At the same time, the displacement sensor data showed that the packaging arm was not accurately returned, and the deviation value reached 18% of the standard value. The dynamic threshold detection system generated an abnormal warning information, judging that it was a multi-parameter combination abnormality, and the priority was marked as "high". The adaptive control system calculated the best adjustment plan: reduce the main motor operating frequency by 5Hz, increase the feedback sensitivity of the tension control system, and adjust the calibration parameters of the packaging arm position sensor. After the adjustment was performed, the tension parameters stabilized within the range of 395N±15N, and the packaging effect was significantly improved. The system conducted strength tests on 800 packaging samples before and after the adjustment and found that the compressive strength increased by 12%. The quality traceability mark was generated and recorded for storage. Parameter trend analysis showed that the wear index of the main drive motor bearing had reached the warning level. The system generated maintenance recommendations and adjusted the bearing replacement cycle from the originally planned 30 days to 18 days to ensure reliable operation of the equipment.

[0034] In the embodiment of the present application, the multi-dimensional parameters of the packaging equipment are collected through the sensor network and real-time filtering is performed, which overcomes the limitations of traditional single parameter monitoring, realizes comprehensive monitoring of the operating status of the packaging equipment, and greatly improves the accuracy and integrity of data collection. Data mining is performed based on real-time data of the packaging process, parameter features are extracted, and parameter models and standard libraries are constructed. The complex correlation between parameters is modeled by applying artificial intelligence algorithms, so that the system can automatically identify parameter features under different working conditions, establish an objective packaging quality evaluation system, and get rid of the subjectivity and uncertainty of traditional reliance on manual experience judgment. Through the dynamic threshold detection parameter deviation, the system can automatically adjust the monitoring threshold according to different packaging conditions and packaging characteristics, avoid the false alarm and missed alarm problems caused by fixed thresholds, improve the accuracy and sensitivity of abnormal detection, and greatly reduce the missed detection rate of quality problems. Based on the adaptive control system of abnormal warning information, combined with the machine learning algorithm to calculate the optimal parameter combination in real time, realize the intelligent automatic adjustment of parameters, reduce manual intervention, and improve packaging efficiency and consistency. The application of strength detection and data encryption technology constructs an unalterable packaging quality traceability mark, ensures the reliability and integrity of quality data, and provides a solid foundation for full-process quality traceability. Parameter trend analysis based on quality traceability identification, identification of equipment wear patterns through deep learning algorithms, prediction of the remaining life of components, and transformation of traditional planned maintenance into predictive maintenance not only avoids the waste of resources caused by premature replacement of components, but also prevents production interruptions caused by sudden equipment failures, significantly improving equipment reliability and production efficiency. By building a complete packaging parameter monitoring method, the artificial intelligence algorithm is deeply integrated with the packaging equipment control, forming a closed-loop control in the links of data collection, feature extraction, anomaly detection, parameter optimization, quality traceability and predictive maintenance, and realizing intelligent monitoring and optimization of the packaging process.

[0035] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0036] (1) Install pressure sensors, temperature sensors, displacement sensors, speed sensors and tension sensors at key locations of packaging equipment to form a multi-dimensional sensor data acquisition network;

[0037] (2) Convert the analog signals collected by the multi-dimensional sensor data acquisition network into digital signals through a high-precision analog-to-digital converter to obtain the original packaged parameter data;

[0038] (3) Dynamically adjust the sampling frequency of the original packaging parameter data according to the different stages of the packaging process to ensure the timeliness of data collection;

[0039] (4) Applying a filtering algorithm to the original packaged parameter data to remove high-frequency noise and obtain preliminary filtered data;

[0040] (5) Smoothing the abnormal peaks in the preliminary filtered data by the median filtering method to obtain smoothing parameter data;

[0041] (6) The smoothing parameter data is transmitted to the central processing unit in real time via industrial Ethernet to form real-time data of the packaging process.

[0042] Specifically, identify the key monitoring points of the packaging equipment, including the core parts such as the main drive device of the packaging machine, the packaging belt tensioning device, the sealing and heat sealing device, the packaging material conveying device and the robotic arm action mechanism. Different types of sensors are installed at these locations. Pressure sensors are mainly installed at the strapping tensioning device and the sealing and pressing part to monitor the pressure changes during the strapping process. Common pressure sensor types include piezoresistive and capacitive, with a sensitivity range of 0-1000N. Temperature sensors are mainly installed at the heat sealing device and the main drive motor to monitor the heat sealing temperature and the operating temperature of the equipment. Thermocouples or infrared temperature measuring elements are used, and the measurement range is generally 0-300℃. Displacement sensors are installed at the mechanical arm and strapping guide device to monitor the position changes of moving parts. Common displacement sensors are inductive and photoelectric, with an accuracy of ±0.1mm. Speed ​​sensors are installed at the conveyor belt and strapping drive device to monitor the running speed of the transmission parts, including Hall sensors or photoelectric encoders, with a measurement range of 0-2000rpm. Tension sensors are installed at the tensioning wheel and guide wheel of the strapping belt to monitor the tension changes of the strapping belt during operation. Commonly used strain tension sensors have a measurement range of 0-500N. These sensors together constitute a multi-dimensional sensor data acquisition network to form a full range of parameter monitoring of the strapping equipment. The analog signals collected by the above sensors are converted into digital signals through high-precision analog-to-digital converters. The analog-to-digital converter is a bridge connecting the analog world and the digital processing system. The analog-to-digital converter used in packaging equipment monitoring has a resolution of no less than 16 bits, a sampling rate of up to 100kHz, and a signal-to-noise ratio greater than 80dB to ensure the accuracy of signal conversion. The analog-to-digital conversion process preamplifies the analog signal, then locks the instantaneous voltage value through a sample-and-hold circuit, and then converts it into a binary digital signal through quantization encoding. The selection of the analog-to-digital converter also needs to consider the anti-interference of the equipment working environment, and adopts differential input and optoelectronic isolation technology to reduce electromagnetic interference in industrial environments. The original packaging parameter data obtained after the analog-to-digital conversion is stored in the form of data packets, each of which contains the sensor ID, timestamp, measurement value, and status flag.

[0043] According to the different stages of the packaging process, the sampling frequency of the original packaging parameter data needs to be dynamically adjusted. The packaging process is usually divided into the stages of startup preparation, material entry, strapping conveying, tensioning and fixing, heat sealing and cutting, and output unloading. The key parameters and change rates of each stage are different. In the startup preparation stage, each sensor uses low-frequency sampling, such as 10Hz, to monitor the initial state of the equipment; in the material entry and strapping conveying stage, the sampling frequency of the speed sensor and displacement sensor is increased to 50Hz to monitor the conveying state in real time; in the tensioning and fixing stage, the sampling frequency of the pressure sensor and tension sensor is increased to 100Hz to capture the rapidly changing mechanical parameters; in the heat sealing and cutting stage, the sampling frequency of the temperature sensor is increased to 80Hz to closely monitor the heat sealing temperature fluctuation. The dynamic adjustment of the sampling frequency is based on the importance of the current working stage and the parameter change rate. The stage recognition algorithm automatically determines the current working stage, and then calls the preset sampling frequency configuration table to send the sampling frequency adjustment instruction to each sensor node. The filtering algorithm needs to be applied to the collected original packaging parameter data to remove high-frequency noise. In industrial environments, packaging equipment is affected by various interference sources, such as motor electromagnetic interference, mechanical vibration, and power grid fluctuations. These interferences appear as high-frequency noise in the signal. The filtering algorithm needs to determine the effective frequency band of the signal. For the monitoring parameters of packaging equipment, the effective signal frequency is generally in the range of 0-100Hz. Therefore, a bandpass filter is designed with a cutoff frequency set at 10Hz and 100Hz, and an attenuation slope of 24dB / octave. The implementation of the digital filter adopts a finite impulse response filter structure. By designing a suitable time domain window function, noise is suppressed to the greatest extent while retaining the effective signal. The filtered data is called preliminary filtered data. At this time, most of the continuous high-frequency noise has been removed, but there may still be abnormal peaks caused by random burst interference.

[0044] The median filtering method is used to smooth the abnormal peaks in the preliminary filtered data. Median filtering is a nonlinear filtering technology that is particularly suitable for removing pulse interference and spike noise in signals. The basic principle of median filtering is to sort the data points within the time window and then select the middle value as the filter output. In the monitoring of packaging equipment, the median filter window size is dynamically adjusted according to the signal characteristics. The pressure and tension signals use a 5-point window, the temperature signal uses a 3-point window, and the displacement and speed signals use a 7-point window. The data processed by the median filter retains the basic form and trend of the signal, while effectively suppressing the abnormal peaks to form smooth parameter data. For particularly critical parameters, the median filter results are also combined with weighted average filtering to perform secondary smoothing. The smoothed parameter data is transmitted to the central processing unit in real time via industrial Ethernet. Industrial Ethernet is a communication network designed for industrial control environments with high reliability, deterministic delay and anti-interference capabilities. In the packaging equipment monitoring system, 100Mbps industrial Ethernet is used to support PROFINET or EtherNet / IP protocols to ensure the real-time and reliability of data transmission. The data transmission uses the TCP / IP protocol stack, and the QoS function is enabled to assign a higher transmission priority to key parameter data. The data is encapsulated in a standard frame format, including the source address, destination address, frame type, data payload, and checksum fields, and forwarded to the central processing unit through the switch. After receiving the data, the central processing unit performs a CRC check to confirm the data integrity, then parses the data frame, extracts the parameter values ​​of each sensor, and integrates them in time series to form a real-time data stream of the packaging process.

[0045] For example, the automatic packaging production line of a food packaging company uses the above method to realize parameter monitoring when packaging cartons. The tension sensor installed on the strapping belt tensioning wheel monitors the strapping belt tension in real time. The original data sampling rate is 200Hz, and the signal contains 50Hz power supply interference and mechanical vibration noise. When the packaging enters the tensioning and fixing stage, the system automatically increases the tension sensor sampling rate to 500Hz to capture the rapidly changing tension parameters. The original tension data is processed by bandpass filtering, and a 50-60Hz notch filter is designed to eliminate power supply interference while retaining the effective signal of 0-40Hz. There are still random spikes in the preliminary filtered data. The median filter of the 7-point window is applied to smooth the spikes. The processed tension data shows a steady trend change, reflecting the real strapping belt tension state. When the tension data is detected to show a trend of continuous increase and exceeds the preset warning value, the alarm information is immediately transmitted to the central processing unit through the industrial Ethernet. The central processing unit analyzes the cause of the abnormal tension and finds that it is related to the increase of strapping belt friction. The packaging mechanical execution unit is automatically adjusted to reduce the strapping belt tension, avoid the risk of strapping belt breakage, and ensure the packaging quality.

[0046] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0047] (1) Perform time domain analysis and frequency domain conversion on the real-time data of the packaging process to extract parameter fluctuation feature points and change trend lines;

[0048] (2) Based on the parameter fluctuation characteristic points, cluster analysis technology is used to classify the parameter performance under different packaging conditions to form a packaging parameter feature vector;

[0049] (3) According to the packaged parameter feature vector, establish the correlation matrix between the parameters and determine the key influencing parameters and secondary influencing parameters;

[0050] (4) Compare the key influencing parameters and secondary influencing parameters with historical packaged data to construct multidimensional parameter standard reference intervals;

[0051] (5) For packaging materials of different types and specifications, set up differentiated parameter standard libraries to form a packaging process parameter mapping table;

[0052] (6) The parameter standard library and packaging process parameter mapping table are integrated through the weight distribution method to generate packaging quality evaluation indicators.

[0053] Specifically, time domain analysis and frequency domain conversion of real-time data of the packaging process are the basic steps to extract parameter features. Time domain analysis directly examines signal characteristics in the time dimension, including calculating statistical features such as the maximum value, minimum value, average value, standard deviation and rate of change of parameters. For the pressure parameter data of the packaging equipment, time domain analysis calculates the pressure fluctuation range and identifies the peak and valley points of the pressure change. These characteristic points constitute the fluctuation characteristics of the parameters. By connecting the parameter values ​​of adjacent time points, the parameter change curve over time is drawn, and then the trend line is extracted by using methods such as polynomial fitting or wavelet decomposition. At the same time, frequency domain conversion is performed to convert the time domain signal to the frequency domain through Fourier transform to analyze the frequency components of the parameters. In the parameter monitoring of packaging equipment, frequency domain analysis is implemented by fast Fourier transform (FFT) to extract the spectral characteristics of the signal. By analyzing the main frequency components and energy distribution in the spectrum, the periodic characteristics and abnormal vibration components in the packaging process are identified. Based on the extracted parameter fluctuation characteristic points, cluster analysis technology is used to classify the parameter performance under different packaging conditions. Cluster analysis is an unsupervised learning method that classifies data points with similar characteristics into the same category. In the monitoring of packaging equipment parameters, commonly used clustering algorithms include K-means clustering, hierarchical clustering, and density clustering. K-means clustering divides data into K clusters by iteratively calculating the center points of each category and assigning data points. Hierarchical clustering gradually merges or splits clusters by calculating the distance between data points. Density clustering forms clusters by identifying high-density areas and is suitable for processing irregularly shaped data distributions. In the monitoring of packaging equipment, the parameter fluctuation feature points collected under different working conditions are grouped according to similarity to form working condition feature clusters. Each cluster represents a specific type of packaging working condition, such as normal operation, mild anomalies, and severe anomalies. Representative features are extracted from each cluster to form a packaging parameter feature vector, which contains the feature values ​​of multiple parameters such as pressure, temperature, displacement, speed, and tension.

[0054] According to the packaged parameter feature vector, the correlation matrix between the parameters is established. The correlation matrix reflects the degree of mutual influence between different parameters. The formula for constructing the correlation matrix is:

[0055]

[0056] in, represents the correlation coefficient between parameter i and parameter j, and Respectively represent the values ​​of parameter i and parameter j at time t, and Represent the average values ​​of parameter i and parameter j respectively, and N represents the total number of sampling time points. The correlation coefficients between all parameter pairs are calculated to form a correlation matrix. When the absolute value of the correlation coefficient exceeds the preset threshold (such as 0.7), it is judged as a strongly correlated parameter; otherwise it is a weakly correlated parameter. According to the analysis results of the correlation matrix, the key influencing parameters and secondary influencing parameters are determined. Key influencing parameters refer to parameters that have a strong correlation with multiple other parameters and have a direct impact on the packaging quality; secondary influencing parameters are parameters that have a weak correlation or only affect a few parameters. Compare the key influencing parameters and secondary influencing parameters with historical packaging data to construct a multidimensional parameter standard reference interval. Filter out packaging records that meet quality standards from the historical database, and extract the statistical distribution characteristics of each parameter in these records. For each parameter, calculate its mean. and standard deviation , and then according to the normal distribution characteristics, the standard reference interval is set as . Considering the mutual influence between parameters, it is also necessary to build a multidimensional joint distribution model to more accurately describe the normal fluctuation range of parameters. The multidimensional parameter standard reference interval not only considers the fluctuation range of a single parameter, but also includes the joint distribution characteristics of the parameter combination. This multidimensional reference interval can more comprehensively reflect the normal state of the packaging process and provide a basis for abnormal detection. For packaging materials of different types and specifications, a differentiated parameter standard library is set to form a packaging process parameter mapping table. The packaging materials are classified based on the characteristics of the material (such as paper, plastic, metal), shape (such as square, round, irregular) and size (such as large, medium, small). Then, for each type of packaging material, the optimal packaging parameter combination is extracted from the historical data, including the optimal pressure range, the optimal temperature range, the optimal speed range, etc. These parameter combinations constitute a differentiated parameter standard library. The parameter standard library is associated with the packaging material characteristics through an index mechanism to form a packaging process parameter mapping table. When a new packaging task comes, the system queries the optimal parameter combination from the mapping table according to the packaging material characteristics to achieve parameter preset.

[0057] The parameter standard library and the packaging process parameter mapping table are integrated through the weight distribution method to generate the packaging quality evaluation index. The weight distribution is based on the degree of influence of the parameters on the packaging quality, and the hierarchical analysis method or expert scoring method is used to determine the weight coefficient of each parameter. The calculation formula of the packaging quality evaluation index is:

[0058]

[0059] in, It represents the evaluation index of packaging quality. represents the weight coefficient of the i-th parameter, represents the standardized score of the ith parameter, and m represents the total number of parameters. The standardized score of the parameter is calculated based on the deviation between the actual value and the standard reference interval. When it fully meets the standard, the score is 1, and the farther it deviates from the standard, the lower the score. The packaging quality evaluation index comprehensively reflects the overall quality level of the packaging process and is used to guide the optimization and adjustment of packaging parameters.

[0060] For example, this method is used to monitor parameters in the carton packaging process of a beverage production line. The data collected during the packaging process include parameters such as sealing heat sealing temperature, strapping tension, packaging pressure, conveying speed and displacement accuracy. The data are analyzed in the time domain to calculate the maximum, minimum and average values ​​of the heat sealing temperature in a packaging cycle, and the key turning points in the heating and cooling process are identified. At the same time, the spectrum characteristics of the heat sealing temperature are analyzed by FFT transformation, and the main frequency component of 0.5Hz is identified, which corresponds to the working cycle of the packaging equipment. The K-means clustering algorithm is used to perform cluster analysis on the collected parameter data, setting k=3, and the packaging conditions are divided into three categories: normal, slightly abnormal and severely abnormal. Feature vectors are extracted from each category. For example, the feature vector of the normal condition contains feature values ​​such as heat sealing temperature 175℃±3℃, strapping tension 420N±15N, and packaging pressure 280N±10N. By calculating the correlation matrix between parameters, it was found that the correlation coefficient between heat sealing temperature and sealing strength was 0.85, and the correlation coefficient between strapping tension and packaging pressure was 0.78, which determined that heat sealing temperature and strapping tension were key influencing parameters. By analyzing historical data, a multi-dimensional parameter standard reference range was established, such as the standard range of heat sealing temperature was [170℃, 180℃], and the standard range of strapping tension was [400N, 440N]. According to the different sizes of packaging cartons (250ml, 500ml, 1000ml), differentiated parameter standards were set to form a process parameter mapping table. Finally, through weight distribution, the weight of heat sealing temperature was 0.4, the weight of strapping tension was 0.3, the weight of packaging pressure was 0.2, and the weights of conveying speed and displacement accuracy were 0.05 respectively, and the packaging quality evaluation index was generated. When the system detected that the heat sealing temperature fluctuated to 182℃, which exceeded the standard range, the quality evaluation index decreased, triggering parameter adjustment, avoiding the quality problem of loose sealing.

[0061] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0062] (1) A multi-threaded parallel computing architecture is used to divert and process real-time data of the packaging process to form a parameter monitoring channel;

[0063] (2) For the pressure parameters in the parameter monitoring channel, the dynamic pressure fluctuation range is calculated based on the sliding window to generate the pressure parameter monitoring results;

[0064] (3) For the temperature, displacement, speed and tension parameters in the parameter monitoring channel, the parameter deviation detection threshold is set in combination with the corresponding packaging quality evaluation index to obtain the parameter deviation value;

[0065] (4) Compare the parameter deviation value with the standard interval in the packaging quality evaluation index, calculate the parameter deviation, and form a parameter abnormality degree matrix;

[0066] (5) Based on the parameter anomaly degree matrix, the abnormal parameters are weighted and scored to distinguish single parameter anomalies from multi-parameter combination anomalies, and a packaged anomaly type label is obtained;

[0067] (6) Determine the exception priority and processing strategy based on the packaging exception type mark and parameter deviation, and generate packaging exception warning information.

[0068] Specifically, the multi-threaded parallel computing architecture refers to assigning data processing tasks to multiple computing threads for simultaneous execution, making full use of the computing power of modern multi-core processors. In the parameter monitoring of packaging equipment, the real-time data flow is large and diverse, and traditional single-thread serial processing is difficult to meet the real-time requirements. Therefore, a multi-threaded architecture is used to divert data according to parameter type, and each type of parameter is processed by a dedicated thread. The specific implementation method is to create an independent thread pool, and the number of threads is set according to the number of processor cores and parameter types, usually the number of processor cores plus the number of parameter types. Data diversion uses a parameter identification mapping table to route different types of parameter data to the corresponding processing thread. Each processing thread constitutes an independent parameter monitoring channel, which is responsible for the processing of a type of parameter, including data caching, feature extraction and status monitoring. The parameter monitoring channel maintains a circular buffer internally to store parameter data for the most recent period of time. The buffer size is dynamically adjusted according to parameter characteristics and monitoring requirements. For the pressure parameters in the parameter monitoring channel, the dynamic pressure fluctuation range is calculated based on the sliding window to generate the pressure parameter monitoring results. The sliding window is a data processing technology that sets a fixed-length window on the time series, slides over time, and calculates and analyzes the data in the window. In the pressure parameter monitoring of packaging equipment, the length of the sliding window is determined according to the packaging cycle, usually a part of a complete packaging cycle. The sliding step size is set to an integer multiple of the sampling period to balance the amount of calculation and monitoring accuracy. For each sliding window position, the maximum, minimum, average and standard deviation of the pressure parameter in the window are calculated to obtain the dynamic pressure fluctuation range. The dynamic pressure fluctuation range is continuously updated as the window slides, reflecting the real-time changes of the pressure parameters. At the same time, the pressure change rate is calculated, that is, the pressure difference between adjacent sampling points divided by the time interval, to monitor the sudden change of pressure. The calculated pressure fluctuation range and change rate are used as the monitoring results of the pressure parameters for subsequent anomaly detection.

[0069] For the temperature, displacement, speed and tension parameters in the parameter monitoring channel, the parameter deviation detection threshold is set in combination with the corresponding packaging quality evaluation index to obtain the parameter deviation value. The packaging quality evaluation index already contains the standard reference intervals of each parameter, and the deviation detection threshold is set based on these intervals. The deviation detection threshold is divided into multiple levels, including the warning threshold and the alarm threshold. The warning threshold is usually set to a certain proportion of the standard interval boundary, and the alarm threshold is set to the standard interval boundary. For the temperature parameter, considering its physical properties, the upper and lower asymmetric thresholds are set, and the upper threshold is more stringent; for the displacement parameter, the absolute deviation threshold is set according to the packaging accuracy requirements; for the speed parameter, the relative deviation threshold is set in combination with the dynamic characteristics of the equipment; for the tension parameter, the safety threshold is set based on the bearing capacity of the packaging material. The parameter deviation value is calculated by standardizing the difference between the real-time parameter value and the standard reference value to obtain a dimensionless deviation index. The standardization process uses the Z-score method or the Min-Max scaling method to enable parameters of different dimensions to be uniformly compared. The parameter deviation value is compared with the standard interval in the packaging quality evaluation index, the parameter deviation is calculated, and the parameter abnormality degree matrix is ​​formed. Parameter deviation is a quantitative indicator to measure the degree of deviation of parameters from the standard interval. The calculation method is to divide the parameter deviation value by the allowable deviation range to obtain the relative deviation. When the parameter value is within the standard interval, the deviation is 0; when the parameter value exceeds the standard interval, the deviation is greater than 0, and the greater the deviation, the greater the deviation. In order to comprehensively reflect the abnormal conditions of each parameter, the deviations of all parameters are organized into a parameter abnormality matrix. The rows of the matrix represent different parameters, the columns represent different time points, and the matrix element values ​​represent the deviations of the corresponding parameters at the corresponding time points. The parameter abnormality matrix is ​​a dynamically updated data structure that is continuously updated as new data is generated, providing basic data for anomaly detection.

[0070] Based on the parameter anomaly degree matrix, the abnormal parameters are weighted and scored to distinguish single parameter anomalies from multi-parameter combination anomalies, and the packaged anomaly type mark is obtained. The calculation formula for the weight score is:

[0071]

[0072] in, represents the overall weight score of the abnormal parameters, Indicates the number of parameter groups, represents the number of parameters in each group, represents the weight coefficient of the hth parameter of the gth group, Indicates the deviation value of the corresponding parameter in the parameter abnormality matrix, The influence function that represents the degree of deviation is usually in the form of an exponential or piecewise function, which is used to amplify the impact of severe deviation. According to the degree of influence of the parameters on the packaging quality, the weight coefficient of the key parameters is larger, and the weight coefficient of the secondary parameters is smaller. For single-parameter abnormalities, only one parameter's deviation exceeds the threshold, and the other parameters are normal; for multi-parameter combination abnormalities, multiple parameters are abnormal at the same time, and there is a correlation between the parameters. By setting the discrimination rules, the abnormal situations are classified and marked into different types, such as minor abnormalities, moderate abnormalities, and severe abnormalities. Each abnormality type has a unique identification code for subsequent abnormality processing. According to the packaging abnormality type mark and parameter deviation, the abnormality priority and processing strategy are determined, and the packaging abnormality warning information is generated. The abnormality priority reflects the urgency and severity of the abnormal situation, and is usually divided into three levels: low, medium, and high. The determination of priority is based on the comprehensive evaluation of the abnormality type mark and parameter deviation. Multi-parameter combination abnormalities usually have a higher priority than single parameter abnormalities. The abnormality of key parameters has a higher priority than the abnormality of secondary parameters. The greater the deviation, the higher the priority. For abnormalities of different priorities, corresponding processing strategies are formulated. For low-priority abnormalities, monitoring strategies are adopted to continuously observe parameter changes; for medium-priority abnormalities, adjustment strategies are adopted to fine-tune parameters through adaptive control; for high-priority abnormalities, intervention strategies are adopted, and the machine is shut down for maintenance when necessary. The packaged abnormality warning information includes abnormal parameter type, occurrence time, duration, deviation degree, impact range, and processing suggestions, etc. It is generated in a standardized structured data format to facilitate system processing and human-computer interaction.

[0073] For example, the packaging equipment on the electronic product packaging line uses the above method for anomaly detection. The data collected in real time during the packaging process is divided into multiple monitoring channels through a multi-threaded parallel computing architecture to monitor the packaging pressure, heat sealing temperature, conveying speed, robot arm displacement and strapping tension respectively. In the pressure monitoring channel, a sliding window of appropriate width is used to calculate the pressure fluctuation range in real time. When it is detected that the pressure value fluctuation exceeds the normal range in a certain packaging operation, the system generates a preliminary judgment of abnormal pressure fluctuation. At the same time, the temperature monitoring channel detects that the heat sealing temperature exceeds the standard operating temperature range, and the system calculates the deviation of the temperature parameter. The pressure anomaly and temperature anomaly are filled into the parameter anomaly degree matrix respectively, and the weight scoring formula is used to calculate the comprehensive anomaly score. The system determines the type of anomaly and determines the priority, and generates an early warning message containing specific adjustment suggestions. The operator makes corresponding adjustments based on the early warning information to restore the equipment to normal working state and avoid the occurrence of poor product packaging.

[0074] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0075] (1) Analyze the abnormal warning information of packaging, extract the abnormal parameter type and deviation degree, and establish the parameter adjustment priority table;

[0076] (2) Based on the parameter adjustment priority table, a mathematical model of the packaging process is constructed, and the packaging quality is used as the objective function to obtain the parameter adjustment constraints;

[0077] (3) For small parameter fluctuations in the parameter adjustment constraints, the adjustment increment is calculated through proportional integral differential control to form a parameter fine-tuning plan;

[0078] (4) For moderate deviations in parameter adjustment constraints, the model is used to predict and calculate future parameter change trends and generate early adjustment strategies;

[0079] (5) Integrate the parameter fine-tuning scheme and advance adjustment strategy into a parameter adjustment instruction set, and transmit it to the packaging machinery execution unit through the industrial control network;

[0080] (6) Implement smooth transition adjustment in the packaging machinery execution unit to avoid parameter mutations and form optimized packaging parameters.

[0081] Specifically, parsing the packaged abnormal warning information is the starting point of adaptive control, and it is necessary to extract effective content from the structured warning information. Warning information parsing adopts information hierarchical processing method to identify basic information such as timestamp, device ID and abnormal level contained in the information header, and then separates the abnormal parameter description paragraph to extract the parameter type and deviation degree. Parameter type refers to the specific parameter where the abnormality occurs, such as pressure, temperature, displacement, speed or tension; the deviation degree indicates the degree to which the parameter deviates from the standard range, usually expressed as a percentage or absolute value. The parsing process is implemented through regular expressions or string segmentation technology, and the extracted information is structured to form a standard data format. Based on the extracted abnormal parameter type and deviation degree, a parameter adjustment priority table is established. The table is a two-dimensional data structure, where the row represents the parameter type, the column represents the deviation degree, and the element value in the table represents the adjustment priority. Priority setting follows the following principles: the priority of key parameters is higher than that of secondary parameters; the greater the deviation degree, the higher the priority; the priority of multi-parameter collaborative abnormalities is higher than that of single parameter abnormalities. The parameter adjustment priority table is not static, and will be updated in real time according to the changes in the abnormal situation, providing a decision basis for subsequent parameter adjustments. Based on the parameter adjustment priority table, the construction of the mathematical model of the packaging process needs to comprehensively consider the physical characteristics and process requirements of the packaging equipment. The construction of the mathematical model adopts the gray box model method, combining theoretical analysis and data-driven to describe the relationship between parameters in the packaging process and their impact on packaging quality. The core of the model is to take packaging quality as the objective function, which is a function of each parameter and is expressed as the mapping relationship between packaging quality and each parameter. The construction of the objective function is based on historical data analysis and expert knowledge, and is obtained through multivariate regression or neural network fitting. To ensure the accuracy of the model, the cross-validation method is used to verify the prediction ability of the model. Under the constraints of the objective function, the constraint conditions for parameter adjustment are determined, which include the physical limitations of the parameters, the equipment capacity limitations, and the process requirement limitations. Physical limitations refer to the parameters that cannot exceed their physically possible range, such as the temperature cannot be negative; equipment capacity limitations refer to the parameter adjustment range that the equipment can achieve; and process requirement limitations refer to the parameter range set based on product quality requirements. The parameter adjustment constraints form a multidimensional constraint space, in which the optimal parameter combination is found to optimize the objective function.

[0082] For small parameter fluctuations in parameter adjustment constraints, proportional integral differential control is used to calculate the adjustment increment to form a parameter fine-tuning scheme. Proportional integral differential control (PID control) is a classic feedback control algorithm that is suitable for situations where parameter deviations are small and system characteristics are relatively stable. The basic principle of PID control is to calculate the control quantity based on the size, accumulation and change rate of the deviation, where the proportional term responds to the current deviation, the integral term eliminates the steady-state error, and the differential term predicts the change trend of the deviation. In the parameter control of packaging equipment, the parameters of the PID controller are adjusted according to the parameter characteristics and control requirements. For example, pressure parameter control may require a larger proportional coefficient and a smaller integral coefficient, while temperature parameters require a smaller proportional coefficient and a larger integral coefficient. The adjustment increment calculated by PID control is directly used for parameter fine-tuning, such as adjusting the pressure value, temperature set point or speed command. To avoid over-adjustment, an upper limit is set for the adjustment increment, which usually does not exceed a certain proportion of the current parameter value. The parameter fine-tuning scheme is a set of parameter adjustment instructions, including the name of the adjusted parameter, the adjustment direction, the adjustment amount and the execution time, which are stored in the form of structured data. For moderate deviations in parameter adjustment constraints, the model is used to predict and calculate the future parameter change trend, and generate an advance adjustment strategy. Moderate deviation refers to the situation where the parameter deviates significantly from the standard range but has not yet caused serious quality problems. At this time, it is necessary to use the model prediction method to evaluate the parameter change trend and the potential impact on quality, so as to take measures in advance. Model predictive control (MPC) is an advanced control method based on system model prediction and optimization, which is suitable for multi-variable and constrained control problems. In the parameter control of packaging equipment, MPC predicts the parameter change trend in the future through the mathematical model constructed above, and evaluates the parameter trajectory and packaging quality under different adjustment strategies. The prediction adopts the rolling time domain method, that is, each control decision only executes the control amount at the current moment, and re-predicts and optimizes at the next moment. The advance adjustment strategy is a set of parameter adjustment sequences with time marks, which guides the parameter adjustment behavior at multiple moments in the future. The advance adjustment usually adopts the segmented adjustment method, that is, the large-scale adjustment is decomposed into multiple small steps, which are implemented step by step to avoid drastic changes in parameters.

[0083] The parameter fine-tuning scheme and advance adjustment strategy are integrated into a parameter adjustment instruction set to uniformly manage and execute parameter adjustment behaviors. The parameter adjustment instruction set is a time series database that stores various parameter adjustment instructions in order of execution time. The integration process includes time coordination, conflict detection, and priority processing. Time coordination ensures that the execution time of each adjustment instruction is reasonably distributed to avoid overly intensive adjustments; conflict detection identifies and resolves conflicts between different adjustment instructions for the same parameter; priority processing determines which adjustment instructions to retain in the case of conflicts based on the priority table established above. After integrity and consistency checks, the parameter adjustment instruction set is transmitted to the packaging machinery execution unit through the industrial control network. The industrial control network is a communication network designed for industrial control environments with high reliability, low latency, and anti-interference capabilities. Common protocols include PROFINET, EtherNet / IP, and Modbus TCP. Data transmission uses standard industrial communication protocols to ensure safe and reliable data transmission. At the same time, encryption and verification technologies are used for key control instructions to prevent data tampering and misoperation. Implementing smooth transition adjustment in the packaging machinery execution unit is the last step in optimizing packaging parameters. The packaging machinery execution unit is a hardware unit that directly controls the movement and process parameters of the packaging equipment, including various actuators, drivers and controllers. Smooth transition adjustment refers to avoiding sudden changes and drastic fluctuations in the process of parameter changes from the current value to the target value, so that the change process is smooth and controllable. The technologies to achieve smooth transition include ramp control and S-curve control. Ramp control limits parameter changes to a fixed rate of change, which is suitable for parameters that do not require smoothness; S-curve control uses a low rate of change at the beginning and end of the change, and a high rate of change in the middle stage to form a smooth S-shaped trajectory, which is suitable for parameters that require high stability. Smooth transition adjustment also considers the mutual influence between parameters and adopts a coordinated control strategy, that is, when multiple parameters need to be adjusted at the same time, the adjustment order and rate are determined according to the influence relationship between the parameters to avoid instability caused by mutual interference between the parameters. The parameter changes achieved through smooth transition adjustment not only meet the control requirements, but also avoid mechanical shock and process fluctuations of the equipment, forming an optimized packaging parameter combination.

[0084] For example, a biscuit automatic packaging line faces packaging quality problems and detects abnormal conditions of high heat sealing temperature and unstable pressure. After analyzing the warning information, the abnormal type and degree of deviation of the high heat sealing temperature are extracted, and the pressure fluctuation abnormality is identified. According to the nature and degree of these two types of abnormalities, a parameter adjustment priority table is established to determine that the heat sealing temperature adjustment takes precedence over the pressure adjustment. Based on historical data analysis, it is found that the heat sealing temperature and packaging quality have a nonlinear relationship. A mathematical model is constructed to describe this relationship and determine the constraint range of temperature adjustment. For small deviations of high temperature, PID control is used to calculate the specific value of temperature reduction. At the same time, based on the frequency characteristics of pressure fluctuations, an adjustment plan for the parameters of the pressure control system is generated. For moderate pressure deviations, model predictive control is used to analyze the pressure change trend and generate an adjustment plan containing multiple time points. The temperature fine-tuning plan and pressure adjustment strategy are integrated into a unified instruction set and transmitted to the execution unit via industrial Ethernet. The execution unit uses ramp control to achieve a smooth decline in the heat sealing temperature and a phased adjustment strategy for the pressure control parameters to avoid packaging instability caused by sudden changes. Through this series of coordinated control, the heat sealing temperature of the packaging line returns to normal, the pressure fluctuation is significantly reduced, and the packaging quality problem is solved.

[0085] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0086] (1) Conduct physical property tests on the packaging materials completed with optimized packaging parameters to obtain data on sealing strength, compression resistance and shear resistance, and form a set of packaging material strength indicators;

[0087] (2) Compare and analyze the packaging material strength index set with the packaging quality evaluation index, calculate the packaging quality score, and generate the packaging quality evaluation result;

[0088] (3) Integrate the optimized packaging parameters, real-time data of the packaging process, and packaging quality assessment results in time series to form a packaging process data chain;

[0089] (4) The packaging process data chain is encrypted in blocks using a hash algorithm to generate packaging quality data blocks that cannot be tampered with;

[0090] (5) Generate a unique digital signature for each packaging quality data block, associate it with the corresponding packaging material, and establish a data block index table;

[0091] (6) Based on the data block index table, a unique identification code is assigned to the packaging to form a packaging quality traceability mark.

[0092] Specifically, the various physical performance indicators of the packaging are quantitatively evaluated. The test items include sealing strength, pressure resistance and shear resistance. The sealing strength test uses an air tightness test or a water tightness test method to apply a certain pressure to the sealing seam of the packaging, measure the pressure value when the seam starts to leak, or measure the leakage rate under a fixed pressure. The pressure resistance test uses a compression test device to apply uniform pressure to the packaging and measure the pressure value when the packaging is significantly deformed or damaged. The shear resistance test measures the ability of the packaging material to resist shear damage by applying shear force to the packaging. The test process is strictly carried out in accordance with industry standards to ensure the reliability and comparability of the test results. The test data is recorded in real time by a dedicated data acquisition device, and a test report in a standard format is formed after preliminary processing. The various index values ​​in the report constitute a packaging strength index set, which is an objective basis for evaluating the packaging quality. Comparative analysis of the packaging strength index set with the packaging quality evaluation index requires a reasonable data processing method. The packaging quality evaluation index is a set of quality standards obtained through parameter modeling and data analysis in the previous steps, which includes the standard range and target value of each performance index. Comparative analysis requires standardization of strength indicators so that indicators of different dimensions can be compared uniformly. The standardization method usually uses normalization to convert the values ​​of each indicator to the range of 0-1. Then calculate the conformity of each indicator with the standard value. The conformity calculation considers whether the indicator value falls within the standard range and how close it is to the target value. According to the importance of each indicator, a weight coefficient is assigned, and the weighted sum is obtained to obtain a comprehensive packaging quality score. The quality score is usually set as a percentage, reflecting the overall quality level of the packaging. According to the score range, the quality grade is divided into four levels: excellent, good, qualified and unqualified. The quality assessment results are stored in the form of structured data, including the overall score, the score of each sub-item and the quality grade.

[0093] The data chain is constructed by integrating the optimized packaging parameters, real-time data of the packaging process and the packaging quality assessment results in time series. Time series integration refers to the unified organization of data from different sources and formats in chronological order to form a coherent data stream. The integration process determines the time reference, usually taking the start time of the packaging process as the reference point to establish a unified time coordinate system. Then the packaging parameter data, process monitoring data and quality assessment data are mapped to this time coordinate system to form a time-aligned data set. Data integration adopts a multi-layer structure, with the bottom layer storing raw data, the middle layer storing processed feature data, and the top layer storing analysis results and evaluation indicators. The association between data is achieved through an index mechanism to ensure that data of different levels and types can be queried in association with each other. The integrated data constitutes a packaging process data chain, which reflects the entire process from parameter setting, process monitoring to quality assessment.

[0094] Block encryption of the packaging process data chain through hash algorithm is an important technology to ensure data security and non-tamperability. Hash algorithm is an algorithm that compresses messages of any length into fixed-length outputs, and is unidirectional and collision-resistant. In packaging quality traceability, commonly used hash algorithms include SHA-256, SHA-3 and other secure hash algorithms. Block encryption refers to dividing the data chain into multiple data blocks according to certain rules, and each data block contains complete data within a period of time. The block size is determined according to the amount of data and processing requirements, usually the data of a complete packaging cycle. The hash value is calculated for each data block, and the hash value is used as the unique identifier of the data block. At the same time, the content of the data block is encrypted, and the encryption adopts a combination of symmetric encryption and asymmetric encryption to ensure data security. The encrypted data block is stored together with its hash value to form a tamper-proof packaging quality data block. The data blocks are associated through a chain structure, and each data block contains a hash value reference of the previous data block, forming a data structure similar to the blockchain. Any tampering with historical data will cause the chain structure to break and be detected. Generating a unique digital signature for each packaging quality data block is a key technology to ensure the reliability and integrity of the data source. Digital signature is a method of signing data using asymmetric key technology, which can verify the source and integrity of the data. The digital signature generation process includes encrypting the hash value of the data block using a private key to form signature data. The signature is performed using the private key of the device to ensure that only the person holding the corresponding device can generate a valid signature. The signature data is stored together with the original data block and the hash value to form a signature data packet. At the same time, an association relationship between the data block and the physical packaging is established, that is, the identification information of the data block is associated with the physical identification of the packaging. The association method includes printing a QR code or barcode on the packaging, which contains the query index of the data block. The index information of all data blocks is aggregated to form a data block index table, which contains information such as the data block ID, generation time, associated packaging ID and storage location, providing an entry for quick query and access.

[0095] Based on the data block index table, assigning a unique identification code to the packaging is the last step to achieve packaging quality traceability. The unique identification code is a globally unique code used to identify and track specific packaging throughout the supply chain. The generation of the identification code adopts a hierarchical coding method, which includes multiple parts such as manufacturer code, production line code, time code and serial number to ensure global uniqueness. The identification code is associated with the entry in the data block index table. The corresponding data block can be queried through the identification code, and then the packaging process data and quality assessment information can be obtained. The identification code is attached to the packaging in an easy-to-identify and read form, such as a barcode, QR code or RFID tag. At the same time, an identification code parsing system is established to support scanning the identification code through mobile devices to obtain packaging quality traceability information. The display of traceability information adopts a multi-level structure, from basic quality information to detailed parameter data, to meet the query needs of different users.

[0096] On a certain beverage automatic packaging line, quality traceability marks are generated for paper beverage boxes produced with optimized packaging parameters. Physical performance tests are performed on random samples of packaged beverage boxes. The test equipment applies pressure to the sealed seam until leakage occurs, and the sealing strength value is recorded; then the beverage box is placed on a compression tester, and the pressure is gradually increased until the package is deformed, and the pressure resistance value is recorded; finally, the shear strength of the packaging material is tested. These test data are compared with the pre-set quality standards to calculate the quality compliance score. At the same time, the heat sealing temperature curve, pressure parameters, and equipment operation status data recorded during the production process are integrated in chronological order to form a data chain. The SHA-256 algorithm is used to calculate the hash value of the integrated data, and the data is stored in blocks, each of which contains complete information about a batch of products. RSA digital signatures are generated for the data blocks to ensure that the data cannot be tampered with. A QR code is printed on the packaged product, which contains index information pointing to the data block. When quality problems occur in the market, the production parameters and quality assessment data can be traced back by scanning the QR code, and the cause of the problem can be quickly located and targeted measures can be taken.

[0097] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0098] (1) Extract historical parameter data and abnormal records from the packaging quality traceability mark and construct a time series diagram of the equipment operation status;

[0099] (2) Perform trend analysis on the equipment operation status time series diagram, identify the parameter attenuation slope and fluctuation frequency, and obtain the health status index of each key component of the equipment;

[0100] (3) Based on the correlation analysis between the health status index and historical maintenance records, a component wear degradation model is established to calculate the equipment wear index;

[0101] (4) Based on the equipment wear index and production plan data, predict the remaining service life of each component and determine the component replacement cycle schedule;

[0102] (5) Integrate the parts replacement cycle table with maintenance resource data to generate equipment maintenance time windows and spare parts requirement lists;

[0103] (6) The maintenance time windows are sorted and adjusted through optimization algorithms to balance maintenance costs and equipment reliability and form an equipment maintenance plan.

[0104] Specifically, by parsing the identification database, the parameter values ​​such as pressure, temperature, displacement, speed and tension recorded during the operation of the equipment, as well as the marked abnormal events, are extracted. The extracted data are arranged in time series to form a time series diagram of the equipment operation status, which intuitively reflects the equipment operation trend and the distribution of abnormal points. The time series diagram adopts a multi-curve display method, and different parameters are distinguished by different colors, which is convenient for identifying the relationship between parameters and abnormal patterns. Trend analysis is performed on the time series diagram of the equipment operation status, and regression analysis and spectrum analysis methods are used to identify key features. Regression analysis calculates the slope of the parameter change over time, reflecting the parameter attenuation rate; spectrum analysis identifies the frequency characteristics of parameter fluctuations through Fourier transform, reflecting the vibration characteristics of the equipment. Through these analyses, the health status index of each key component of the equipment is obtained. The index quantitatively represents the health of the component, and the higher the value, the better the status.

[0105] Based on the correlation analysis between the health status index and historical maintenance records, a component wear degradation model is established. This model describes the change of component health status with usage time and workload, and uses the cumulative damage theory to weight the impact of multiple factors. Based on this model, the equipment wear index is calculated to comprehensively reflect the overall wear degree of the equipment. Based on the equipment wear index and production plan data, the life prediction algorithm is used to predict the remaining service life of each component. The prediction takes into account the current wear degree and future workload to derive the time range in which the component may fail. Based on the prediction results, a component replacement cycle table is formulated to determine the optimal replacement time point for each component.

[0106] Integrate the parts replacement cycle table with maintenance resource data, consider the impact of maintenance personnel, spare parts inventory and equipment downtime, and generate equipment maintenance time windows and spare parts demand lists. The maintenance time window refers to the time period suitable for maintenance, taking into account the production plan and the availability of maintenance resources. The maintenance time window is sorted and adjusted through a multi-objective optimization algorithm to balance maintenance costs and equipment reliability. The optimization goals include minimizing maintenance costs, maximizing equipment reliability and minimizing production impact. The resulting equipment maintenance plan contains a detailed maintenance plan, spare parts list and execution steps to guide the actual maintenance work.

[0107] On a beverage packaging production line, three months of historical data extracted from the traceability mark revealed that the temperature control accuracy of the heat sealing device gradually decreased and the fluctuation frequency increased. Trend analysis showed that the temperature sensor signal was significantly attenuated and the health status index decreased. Correlating historical maintenance records found that similar symptoms were usually caused by sensor aging, and a degradation model was established to predict that the remaining life of the sensor was less than two weeks. Considering the production plan and spare parts situation, the sensor was replaced during the planned downtime next week, and the relevant control circuits were repaired at the same time, avoiding the possibility of large-scale packaging defects.

[0108] The above describes the packaging parameter monitoring method for the packaging device in the embodiment of the present application. The following describes the packaging parameter monitoring system for the packaging device in the embodiment of the present application. Figure 2 In one embodiment of the present application, a packaging parameter monitoring system for packaging equipment includes:

[0109] The filtering module is used to collect the pressure, temperature, displacement, speed and tension parameters of the packaging equipment through the sensor network, filter the parameters and obtain the real-time data of the packaging process;

[0110] A construction module is used to extract parameter features through data mining based on the real-time data of the packaging process, build a parameter model and a standard library, and obtain a packaging quality evaluation index;

[0111] A detection module, used to detect parameter deviations through dynamic thresholds according to the real-time data of the packaging process and the packaging quality evaluation index, and obtain packaging abnormality warning information;

[0112] A calculation module, used to calculate the best parameter combination through adaptive control based on the abnormal packaging warning information, adjust the packaging machine execution unit, and obtain optimized packaging parameters;

[0113] An encryption module, used to obtain a packaging quality traceability mark through strength detection and data encryption according to the optimized packaging parameters and actual packaging effect;

[0114] The identification module is used to identify the equipment wear index through parameter trend analysis based on the packaging quality traceability mark, generate a component replacement cycle, and obtain an equipment maintenance plan.

[0115] Through the collaboration of the above components, the multi-dimensional parameters of the packaging equipment are collected through the sensor network and filtered in real time, which overcomes the limitations of traditional single parameter monitoring, realizes comprehensive monitoring of the operating status of the packaging equipment, and greatly improves the accuracy and integrity of data collection. Based on the real-time data of the packaging process, data mining is carried out to extract parameter features and build parameter models and standard libraries. The complex correlation between parameters is modeled by artificial intelligence algorithms, so that the system can automatically identify parameter features under different working conditions and establish an objective packaging quality evaluation system, getting rid of the subjectivity and uncertainty of traditional reliance on manual experience judgment. Through the dynamic threshold detection parameter deviation, the system can automatically adjust the monitoring threshold according to different packaging conditions and packaging characteristics, avoiding the false alarm and missed alarm problems caused by fixed thresholds, improving the accuracy and sensitivity of anomaly detection, and greatly reducing the missed detection rate of quality problems. Based on the adaptive control system of abnormal warning information, combined with the machine learning algorithm to calculate the optimal parameter combination in real time, it realizes the intelligent automatic adjustment of parameters, reduces manual intervention, and improves packaging efficiency and consistency. The application of strength detection and data encryption technology builds an unalterable packaging quality traceability mark, ensures the reliability and integrity of quality data, and provides a solid foundation for full-process quality traceability. Parameter trend analysis based on quality traceability identification, identification of equipment wear patterns through deep learning algorithms, prediction of the remaining life of components, and transformation of traditional planned maintenance into predictive maintenance not only avoids the waste of resources caused by premature replacement of components, but also prevents production interruptions caused by sudden equipment failures, significantly improving equipment reliability and production efficiency. By building a complete packaging parameter monitoring method, the artificial intelligence algorithm is deeply integrated with the packaging equipment control, forming a closed-loop control in the links of data collection, feature extraction, anomaly detection, parameter optimization, quality traceability and predictive maintenance, and realizing intelligent monitoring and optimization of the packaging process.

[0116] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0117] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0118] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0119] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0120] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0121] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0122] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A packaging parameter monitoring method for packaging equipment, characterized in that: It includes collecting the pressure, temperature, displacement, speed and tension parameters of the packaging equipment through the sensor network, filtering the parameters and obtaining the real-time data of the packaging process; Based on the real-time data of the packaging process, parameter features are extracted through data mining, parameter models and standard libraries are constructed, and packaging quality evaluation indicators are obtained; According to the real-time data of the packaging process and the packaging quality evaluation indicators, the parameter deviation is detected through the dynamic threshold to obtain the packaging abnormality warning information; Based on the abnormal packaging warning information, the optimal parameter combination is calculated through adaptive control, and the packaging machinery execution unit is adjusted to obtain the optimized packaging parameters; According to the optimized packaging parameters and the actual packaging effect, the packaging quality traceability mark is obtained through strength detection and data encryption, including: physical performance testing of the packaging completed with the optimized packaging parameters, obtaining the sealing strength, compression resistance and shear resistance data, and forming a packaging strength index set; comparing and analyzing the packaging strength index set with the packaging quality evaluation index, calculating the packaging quality score, and generating the packaging quality evaluation result; integrating the optimized packaging parameters, the real-time data of the packaging process and the packaging quality evaluation result in time series to form a packaging process data chain; performing block encryption processing on the packaging process data chain through the hash algorithm to generate a tamper-proof packaging quality data block; generating a unique digital signature for each packaging quality data block, and associating it with the corresponding packaging to establish a data block index table; assigning a unique identification code to the packaging based on the data block index table to form a packaging quality traceability mark Based on the packaging quality traceability mark, the equipment wear index is identified through parameter trend analysis, the component replacement cycle is generated, and the equipment maintenance plan is obtained, including: extracting historical parameter data and abnormal records from the packaging quality traceability mark, and constructing a time series diagram of the equipment operation status; performing trend analysis on the equipment operation status time series diagram, identifying the parameter attenuation slope and fluctuation frequency, and obtaining the health status index of each key component of the equipment; establishing a component wear degradation model based on the correlation analysis between the health status index and historical maintenance records, and calculating the equipment wear index; based on the equipment wear index and production plan data, predicting the remaining service life of each component and determining the component replacement cycle table; integrating the component replacement cycle table with the maintenance resource data to generate the equipment maintenance time window and spare parts demand list; sorting and adjusting the maintenance time window through the optimization algorithm, balancing the maintenance cost and equipment reliability, and forming an equipment maintenance plan.

2. The packaging parameter monitoring method for packaging equipment according to claim 1, characterized in that: The pressure, temperature, displacement, speed and tension parameters of the packaging equipment are collected through a sensor network, and the parameters are filtered to obtain real-time data of the packaging process, including: installing pressure sensors, temperature sensors, displacement sensors, speed sensors and tension sensors at key positions of the packaging equipment to form a multi-dimensional sensor data collection network; converting analog signals collected by the multi-dimensional sensor data collection network into digital signals through a high-precision analog-to-digital converter to obtain original packaging parameter data; dynamically adjusting the sampling frequency of the original packaging parameter data according to different stages of the packaging process to ensure the timeliness of data collection; applying a filtering algorithm to the original packaging parameter data to remove high-frequency noise to obtain preliminary filtered data; smoothing abnormal peaks in the preliminary filtered data through a median filtering method to obtain smoothed parameter data; transmitting the smoothed parameter data to a central processing unit in real time through industrial Ethernet to form real-time data of the packaging process.

3. The packaging parameter monitoring method for packaging equipment according to claim 1, characterized in that: Based on the real-time data of the packaging process, parameter features are extracted through data mining, and parameter models and standard libraries are constructed to obtain packaging quality evaluation indicators, including: time domain analysis and frequency domain conversion of the real-time data of the packaging process to extract parameter fluctuation characteristic points and change trend lines; based on the parameter fluctuation characteristic points, cluster analysis technology is used to classify the parameter performance under different packaging conditions to form a packaging parameter characteristic vector; according to the packaging parameter characteristic vector, a correlation matrix between the parameters is established to determine the key influencing parameters and secondary influencing parameters; the key influencing parameters and secondary influencing parameters are compared with historical packaging data to construct a multi-dimensional parameter standard reference interval; for different types and specifications of packaging materials, differentiated parameter standard libraries are set to form a packaging process parameter mapping table; the parameter standard library and the packaging process parameter mapping table are integrated through the weight distribution method to generate packaging quality evaluation indicators.

4. The packaging parameter monitoring method for packaging equipment according to claim 1, characterized in that: According to the real-time data of the packaging process and the packaging quality evaluation index, the parameter deviation is detected through the dynamic threshold to obtain the packaging abnormality warning information, including: using a multi-threaded parallel computing architecture to shunt the real-time data of the packaging process to form a parameter monitoring channel; for the pressure parameters in the parameter monitoring channel, the dynamic pressure fluctuation range is calculated based on the sliding window to generate the pressure parameter monitoring result; for the temperature, displacement, speed and tension parameters in the parameter monitoring channel, the parameter deviation detection threshold is set in combination with the corresponding packaging quality evaluation index to obtain the parameter deviation value; the parameter deviation value is compared with the standard interval in the packaging quality evaluation index, the parameter deviation degree is calculated, and the parameter abnormality degree matrix is ​​formed; based on the parameter abnormality degree matrix, the abnormal parameters are weighted and scored to distinguish between single parameter abnormalities and multi-parameter combination abnormalities to obtain the packaging abnormality type mark; according to the packaging abnormality type mark and the parameter deviation degree, the abnormal priority and processing strategy are determined to generate the packaging abnormality warning information.

5. The packaging parameter monitoring method for packaging equipment according to claim 1, characterized in that: Based on the abnormal packaging warning information, the optimal parameter combination is calculated through adaptive control, and the packaging machinery execution unit is adjusted to obtain the optimized packaging parameters, including: parsing the abnormal packaging warning information, extracting the abnormal parameter type and the degree of deviation, and establishing a parameter adjustment priority table; based on the parameter adjustment priority table, a mathematical model of the packaging process is constructed, and the packaging quality is used as the objective function to obtain the parameter adjustment constraints; for small parameter fluctuations in the parameter adjustment constraints, the adjustment increment is calculated through proportional integral differential control to form a parameter fine-tuning plan; for medium-degree deviations in the parameter adjustment constraints, the future parameter change trend is calculated through model prediction to generate an advance adjustment strategy; the parameter fine-tuning plan and the advance adjustment strategy are integrated into a parameter adjustment instruction set, which is transmitted to the packaging machinery execution unit through the industrial control network; Implement smooth transition adjustment in the execution unit of the packaging machinery to avoid sudden changes in parameters and form optimized packaging parameters.

6. A packaging parameter monitoring system for packaging equipment, used to implement the packaging parameter monitoring method for packaging equipment according to any one of claims 1 to 5, characterized in that: The packaging parameter monitoring system for packaging equipment includes: a filtering module for collecting pressure, temperature, displacement, speed and tension parameters of the packaging equipment through a sensor network, filtering the parameters, and obtaining real-time data of the packaging process; The construction module is used to extract parameter features through data mining based on real-time data of the packaging process, build parameter models and standard libraries, and obtain packaging quality evaluation indicators; The detection module is used to detect parameter deviations through dynamic thresholds based on real-time data of the packaging process and packaging quality evaluation indicators, and obtain packaging abnormality warning information; A calculation module is used to calculate the best parameter combination through adaptive control based on the abnormal warning information of packaging, adjust the execution unit of the packaging machinery, and obtain the optimized packaging parameters; The encryption module is used to obtain the packaging quality traceability mark through strength detection and data encryption according to the optimized packaging parameters and actual packaging effect; The identification module is used to identify the equipment wear index based on the packaging quality traceability mark through parameter trend analysis, generate the component replacement cycle, and obtain the equipment maintenance plan.

7. A computer device, characterized in that: It comprises a memory and a processor, the memory stores a computer program that can be run on the processor, and is characterized in that when the processor executes the computer program, the packaging parameter monitoring method for packaging equipment described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is enabled to execute the packaging parameter monitoring method for a packaging device according to any one of claims 1 to 5.

Images