Adaptive control method of fully automatic packaging equipment based on dynamic data acquisition

Abstract

The present invention discloses an adaptive control method for fully automatic packaging equipment based on dynamic data acquisition. The method comprises configuring sensors in the packaging equipment to collect dynamic data in real time, and using multi-dimensional data analysis technology to clean, detect anomalies and predict trends of the collected data, thereby generating a characteristic parameter set related to the current packaging working condition; dynamically adjusting the operating parameters of the equipment according to the characteristic parameter set, including speed adjustment parameters, temperature control parameters and tension control parameters, so as to achieve adaptive control of different packaging materials and specifications; in addition, based on the results of multi-dimensional data analysis, the system can identify potential equipment anomalies or process problems in the packaging process, and generate optimization suggestions for operators to refer to or automatically adjust parameters, thereby improving the stability of equipment operation and the consistency of packaging quality; the present invention is suitable for fully automatic packaging equipment with multiple specifications and multiple categories of packaging, and significantly improves production efficiency and product qualification rate.

Description

Technical Field

[0001] The present invention relates to the field of control technology, and in particular to a self-adaptive control method of full-automatic packaging equipment based on dynamic data acquisition. Background Art

[0002] In the existing technology, fully automatic packaging equipment is widely used in production packaging in the fields of food, medicine and industrial products. It mainly relies on fixed preset parameters and basic control algorithms to realize the automation of the packaging process. Some advanced equipment is equipped with sensors and simple parameter adjustment functions to monitor basic parameters such as packaging speed, temperature or sealing to meet production needs within a certain range. However, the parameter adjustment of these devices is usually static or manually triggered, and cannot effectively cope with complex and changeable working conditions.

[0003] However, existing equipment often suffers from insufficient adaptability when faced with multi-specification and multi-material packaging requirements. Specifically, traditional equipment lacks the ability to fully collect and analyze dynamic data and is unable to detect and adjust key operating parameters in real time. In addition, when dealing with packaging anomalies or process problems, equipment usually relies on manual intervention, which not only prolongs the time to solve the problem, but may also lead to reduced production efficiency and unstable packaging quality.

[0004] In order to solve the above problems, the present invention proposes a self-adaptive control method for full-automatic packaging equipment based on dynamic data acquisition. Summary of the invention

[0005] The present application provides a fully automatic packaging equipment adaptive control method based on dynamic data acquisition to improve the production efficiency and product qualification rate of the packaging equipment.

[0006] The present application provides a method for adaptively controlling a fully automatic packaging device based on dynamic data collection, comprising:

[0007] The dynamic data in the packaging process are collected by sensors installed in the fully automatic packaging equipment, and the dynamic data include packaging speed, temperature, material stretchability and sealing property;

[0008] Perform multi-dimensional analysis on the collected dynamic data to generate a set of characteristic parameters related to the current packaging working conditions, wherein the multi-dimensional analysis includes data cleaning, anomaly detection and trend prediction;

[0009] Using a data-driven adaptive control algorithm, dynamically adjusting equipment operating parameters according to a set of characteristic parameters, the equipment operating parameters including speed adjustment parameters, temperature control parameters and tension control parameters;

[0010] Based on the analysis results of multidimensional data analysis, possible equipment anomalies or process problems in the packaging process are identified, and optimization suggestions are generated for operators' reference or equipment parameters are automatically adjusted.

[0011] Furthermore, the dynamic data collected during the packaging process by the sensor arranged in the fully automatic packaging equipment includes:

[0012] During the operation of the equipment, the packaging speed is monitored in real time by a speed sensor configured on the conveyor belt or the driving device, and its signal is converted into a digital signal and transmitted to the data processing unit;

[0013] Arrange thermocouples or infrared temperature sensors near the heating device to collect temperature data of the sealing area in real time and transmit it to the central control system synchronously;

[0014] Install a strain gauge sensor in the tension control device to dynamically monitor the stretching state of the packaging material;

[0015] Pressure or vacuum sensors are installed in the sealing area to monitor the stability of the seal and generate a corresponding data stream.

[0016] Furthermore, the method of collecting dynamic data during the packaging process by using sensors disposed in the fully automatic packaging equipment also includes:

[0017] After the sensor collects dynamic data, the original data is cleaned through the built-in noise filtering module of the device to remove mutation points and invalid values;

[0018] Use data smoothing algorithm to process continuous data to ensure the stability of data flow;

[0019] Before the data is transmitted to the central control system, the collected packaging speed, temperature, stretchability and sealing data are verified for validity according to the preset threshold range, and the verified data is cached in the dynamic data processing module.

[0020] Furthermore, the collected dynamic data is analyzed in multiple dimensions to generate a set of characteristic parameters related to the current packaging working conditions, including:

[0021] Perform preliminary filtering on the raw data collected by the sensor to remove invalid data caused by sensor fluctuations or environmental interference;

[0022] The continuously collected data is smoothed by the time window sliding method to reduce noise interference;

[0023] Different types of dynamic data are classified and stored.

[0024] Furthermore, the collected dynamic data is analyzed in multiple dimensions to generate a set of characteristic parameters related to the current packaging working conditions, including:

[0025] By comparing real-time dynamic data with historical operation data, abnormal values ​​that deviate from the preset range can be identified;

[0026] For mutation data points, interpolation algorithms are used to generate correction values ​​to ensure data continuity;

[0027] The data stream is checked item by item according to the preset anomaly rules, and the detected anomaly marks are stored for system alarm or further analysis.

[0028] Furthermore, the collected dynamic data is analyzed in multiple dimensions to generate a set of characteristic parameters related to the current packaging working conditions, including:

[0029] Based on the time series model of historical collected data, the packaging speed, temperature change trend and material stretchability change are predicted;

[0030] Combined with current real-time data, the weight of trend forecasts is dynamically adjusted to enhance sensitivity to short-term changes;

[0031] Generate a set of characteristic parameters containing future change trends of key parameters for subsequent optimization and adjustment of equipment operating parameters.

[0032] Furthermore, the data-driven adaptive control algorithm dynamically adjusts the equipment operating parameters according to the characteristic parameter set, and the equipment operating parameters include speed adjustment parameters, temperature control parameters and tension control parameters, including:

[0033] The collected characteristic parameter sets are classified according to data types, including real-time dynamic parameters and predictive parameters, and different weight factors are assigned; wherein the real-time dynamic parameters include packaging speed, temperature, and tension; the predictive parameters include sealing trend and material fatigue curve;

[0034] Based on the classified characteristic parameters, the interaction model between parameters is constructed through multivariate regression analysis, including the coupling model of packaging speed and tension, and the correlation model of temperature and material stretchability;

[0035] According to the current packaging task requirements and material characteristics, the weight distribution of each parameter in the interactive model is adjusted in real time;

[0036] Combining the interaction model and dynamic weights, the optimal values ​​of the equipment operating parameters are calculated in stages. Specifically, in the first stage, the current equipment operating parameters are preliminarily optimized to ensure that the packaging quality meets the basic requirements; in the second stage, based on real-time feedback data, the equipment operating parameters are iterated to improve packaging efficiency; in the third stage: through predictive parameter analysis, the equipment operating parameters are adjusted in advance to avoid potential abnormalities;

[0037] The optimized operating parameters are input into the equipment control unit to adjust the speed, temperature control and tension of the equipment in real time, and the adjustment effect is verified through the online monitoring system.

[0038] Furthermore, the interaction model between the parameters is constructed based on the classified characteristic parameters through multivariate regression analysis, including:

[0039] The coupling model of packaging speed and tension is expressed by the following formula 1:

[0040] ;

[0041] in, Indicates tension; Indicates packaging speed; Represents the material fatigue curve characteristics; is the frequency factor; , is the regression coefficient; is the material characteristic constant; is the basic tension offset;

[0042] The correlation model between temperature and material stretchability is expressed using the following formula 2:

[0043] ;

[0044] in, is the temperature control parameter; Represents the material fatigue curve characteristics; is the regulating factor; is the material response frequency; Indicates the sealing tendency characteristics; , and is the regression coefficient; is the temperature control basic offset.

[0045] Furthermore, the weight distribution of each parameter in the interactive model is adjusted in real time according to the current packaging task requirements and material characteristics, including:

[0046] According to the following formula 3, the regression coefficient Adjust and obtain new regression coefficients :

[0047] ;

[0048] in, Indicates the current speed; Indicates the target speed;

[0049] According to the following formula 4, the regression coefficient Adjust and obtain new regression coefficients :

[0050] ;

[0051] in, Represents the fatigue curve characteristics of the reference material; is the material characteristic constant; Represents the material fatigue curve characteristics;

[0052] According to the following formula 5, the regression coefficient Adjust and obtain new regression coefficients :

[0053] ;

[0054] in, is the frequency factor; Indicates the current speed.

[0055] Furthermore, the weight distribution of each parameter in the interactive model is adjusted in real time according to the current packaging task requirements and material characteristics, including:

[0056] According to the following formula 6, the regression coefficient Adjust and obtain new regression coefficients :

[0057] ;

[0058] in, Represents the material fatigue curve characteristics; is the regulating factor; is the sensitivity coefficient; Represents the fatigue curve characteristics of the reference material;

[0059] According to the following formula 7, the regression coefficient Adjust and obtain new regression coefficients :

[0060]

[0061] in, is the sealing trend sensitivity coefficient; Indicates the reference tightness trend; Indicates the sealing tendency characteristics;

[0062] According to the following formula 8, the regression coefficient Adjust and obtain new regression coefficients :

[0063] ;

[0064] in, is the material response frequency; Represents the material fatigue curve characteristics; Represents the fatigue curve characteristics of the reference material.

[0065] The beneficial effects of the technical solution provided by this application include:

[0066] (1) By collecting key dynamic data such as packaging speed, temperature, material stretchability and sealing in real time, and using multi-dimensional data analysis to generate a set of characteristic parameters, it is possible to accurately adjust the equipment operating parameters under different packaging conditions, thereby effectively improving the consistency of packaging quality and reducing the defective rate. (2) Using data-driven adaptive control algorithms, the speed adjustment, temperature control and tension control parameters of the equipment are dynamically optimized to ensure that the equipment can still operate smoothly under complex working conditions, thereby significantly improving the overall stability of the equipment. (3) Based on the real-time and predictive nature of multi-dimensional data analysis, it is possible to quickly identify equipment anomalies or process problems in the packaging process, generate optimization suggestions or automatically adjust equipment parameters, thereby reducing downtime or quality problems caused by delayed problem handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is a flow chart of a method for adaptively controlling a fully automatic packaging device based on dynamic data collection provided in the first embodiment of the present application. DETAILED DESCRIPTION

[0068] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0069] The first embodiment of the present application provides a method for adaptive control of fully automatic packaging equipment based on dynamic data collection. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 The first embodiment of the present application provides a method for adaptively controlling a fully automatic packaging equipment based on dynamic data collection, which is described in detail.

[0070] Step S101: Dynamic data in the packaging process is collected by sensors installed in the fully automatic packaging equipment. The dynamic data includes packaging speed, temperature, material stretchability and sealing.

[0071] In fully automatic packaging equipment, a variety of sensor modules are installed to collect dynamic data related to the packaging process in real time. The type and location of the sensor are configured according to the specific packaging requirements and equipment structure design. For example, the sensor used to collect packaging speed can be a rotary encoder or photoelectric sensor installed on the drive device, the sensor used to collect temperature can be a thermocouple or infrared temperature sensor distributed in the heating area, the sensor used to collect material stretchability can be a strain gauge sensor integrated in the tension control device, and the sensor used to collect sealing can be a pressure sensor or vacuum detection device installed on the sealing device.

[0072] The specific process of data collection includes the following steps. First, the sensor detects the changes in the target parameters in real time and converts the signal into a digital signal, which is transmitted to the central processing unit (CPU) of the device through the built-in signal acquisition module. To ensure the accuracy and real-time performance of the data, the signal acquisition module needs to set a suitable sampling frequency to capture rapidly changing dynamic data, and at the same time remove noise and invalid data through filtering algorithms to ensure the reliability of data collection.

[0073] Secondly, the collected data will be initially sorted and grouped according to the classification labels, for example, speed data, temperature data, tension data and sealing data are classified into different data sets. The collected values ​​of each type of data will be accompanied by a timestamp to ensure that it can be accurately matched with the packaging process.

[0074] Finally, to meet the data processing requirements of subsequent steps, the collected data will be stored in the dynamic data cache of the device, and the data preprocessing module will be triggered to prepare for multi-dimensional analysis. It should be pointed out that in order to ensure the continuity and reliability of data collection, the sensor needs to be calibrated and maintained regularly, and the hardware and software of the acquisition module need to have anti-interference capabilities to meet the needs of complex industrial environments.

[0075] Through the above implementation, step S101 ensures real-time, efficient and accurate acquisition of dynamic data required in the packaging process, laying a foundation for subsequent data analysis and parameter optimization.

[0076] Furthermore, the dynamic data collected during the packaging process by the sensor arranged in the fully automatic packaging equipment includes:

[0077] During the operation of the equipment, the packaging speed is monitored in real time by a speed sensor configured on the conveyor belt or the driving device, and its signal is converted into a digital signal and transmitted to the data processing unit;

[0078] Arrange thermocouples or infrared temperature sensors near the heating device to collect temperature data of the sealing area in real time and transmit it to the central control system synchronously;

[0079] Install a strain gauge sensor in the tension control device to dynamically monitor the stretching state of the packaging material;

[0080] Pressure or vacuum sensors are installed in the sealing area to monitor the stability of the seal and generate a corresponding data stream.

[0081] During the operation of the equipment, in order to accurately collect dynamic data and provide a basis for subsequent control, various types of sensors are used to collect key parameters in the packaging process, and these data are transmitted to the central processing unit in real time for processing and analysis.

[0082] First, the packaging speed is monitored in real time by a speed sensor configured on the conveyor belt or drive device. The speed sensor can be a rotary encoder or a linear speed sensor, and its working principle is to collect displacement or speed data according to the motion state of the conveyor belt or drive device. After being processed by the signal conversion module, these data are converted into digital signals and transmitted to the data processing unit in real time to describe the dynamic changes of the packaging speed.

[0083] Secondly, thermocouples or infrared temperature sensors are arranged near the heating device of the equipment to monitor the temperature of the sealing area in real time. Thermocouples convert temperature changes into voltage signals through the thermoelectric effect of the material, while infrared temperature sensors perform non-contact temperature measurement by detecting thermal radiation. These temperature data are synchronously transmitted to the central control system to ensure that the temperature of the sealing area remains within the predetermined range, thereby avoiding sealing quality problems caused by temperature fluctuations.

[0084] Strain gauge sensors are installed in the tension control device to dynamically monitor the stretching state of the packaging material. The strain gauge sensor can sense the stress changes generated by the packaging material during the stretching process and convert this change into an electrical signal. By analyzing these signals in real time, it is possible to accurately determine whether the tension of the packaging material is within a reasonable range, thereby avoiding looseness caused by insufficient tension or material breakage caused by excessive tension.

[0085] At the same time, pressure or vacuum sensors are installed in the sealing area to monitor the stability of the seal in real time. The pressure sensor is used to sense the pressure distribution in the sealing area to ensure that the pressure is uniform and meets the sealing process requirements; while the vacuum sensor is used to detect the vacuum degree in the sealing cavity to ensure that it meets special packaging requirements. These data generate a stable data stream through the signal transmission module for the central control system to analyze and judge the sealing quality.

[0086] Through the above method, the layout and signal acquisition of various sensors can cover the key parameters in the packaging process and provide accurate dynamic data support for subsequent analysis and control, thereby ensuring the efficiency of equipment operation and the consistency of packaging quality.

[0087] Furthermore, the method of collecting dynamic data during the packaging process by using sensors disposed in the fully automatic packaging equipment also includes:

[0088] After the sensor collects dynamic data, the original data is cleaned through the built-in noise filtering module of the device to remove mutation points and invalid values;

[0089] Use data smoothing algorithm to process continuous data to ensure the stability of data flow;

[0090] Before the data is transmitted to the central control system, the collected packaging speed, temperature, stretchability and sealing data are verified for validity according to the preset threshold range, and the verified data is cached in the dynamic data processing module.

[0091] To ensure the accuracy and reliability of dynamic data collection, after the sensor collects the data, the system further cleans, processes and verifies the raw data to form a high-quality data stream that can be analyzed.

[0092] First, the system uses the built-in noise filtering module to perform preliminary data cleaning on the raw data collected by the sensor. This process is designed to eliminate mutation points and invalid values ​​caused by environmental interference or fluctuations in the sensor itself. For example, in the packaging speed data, if a sampling value has an abnormal large jump and is not related to the previous and subsequent data, it will be marked as a mutation point by the filtering module and eliminated. Similarly, there may be abnormal values ​​in the temperature or tension data caused by transient hardware failures, which will also be automatically identified and cleared by the noise filtering module.

[0093] Next, the system uses a data smoothing algorithm to further process the cleaned continuous data to ensure the stability of the data stream. Smoothing reduces high-frequency fluctuations and random noise in the data, making the changing trends of key parameters clearer. For example, processing packaging speed and temperature data through sliding average or weighted smoothing methods can eliminate instantaneous random disturbances and generate a more stable input data stream. This stability is crucial for subsequent dynamic parameter adjustments and can significantly improve control accuracy.

[0094] After data cleaning and smoothing, the system verifies the validity of the processed packaging speed, temperature, stretchability and sealing data according to the preset threshold range. The verification process is set according to the physical meaning and process requirements of different parameters. For example, the packaging speed should be between the minimum and maximum values ​​allowed by the equipment, the temperature data should meet the expected range of the sealing process, and the tension and sealing data should meet the requirements of material properties. If a data exceeds the preset threshold range, the data will be marked as abnormal and will not be further analyzed or applied; the verified data will be cached in the dynamic data processing module for subsequent multi-dimensional analysis and parameter optimization.

[0095] Through the above steps, the accuracy and reliability of the data collected by the sensor are ensured, providing high-quality data support for the adaptive control of the fully automatic packaging equipment, thereby improving the stability and efficiency of the packaging process.

[0096] Step S102: Perform multi-dimensional analysis on the collected dynamic data, including data cleaning, anomaly detection and trend prediction, to generate a set of characteristic parameters related to the current packaging working conditions.

[0097] The specific implementation process of step S102 includes performing a systematic multi-dimensional analysis on the collected dynamic data to generate a set of characteristic parameters related to the current packaging working conditions, thereby providing a reliable basis for subsequent control.

[0098] First, the dynamic data collected in step S101 needs to be cleaned. Since there may be noise interference, outliers or missing data in the actual collection process, a series of data processing methods need to be applied to ensure the accuracy and integrity of the data. For example, for noise signals, a noise reduction algorithm based on moving average or Kalman filtering can be used for smoothing; for outlier detection, a statistical outlier discrimination rule can be set to eliminate abnormal data that exceeds the normal range; for missing data, an interpolation algorithm or a historical data completion method can be used to fill in the missing data to ensure data continuity.

[0099] After data cleaning, the system performs anomaly detection on the data. Anomaly detection is mainly achieved by establishing a normal working model based on historical data and analyzing the deviation between the current collected data and the model. If certain features of the data deviate significantly from the historical normal working range, such as a sudden and significant increase in temperature or fluctuations in tension values, the system will mark it as a potential anomaly for further verification in subsequent analysis. This step is achieved by combining machine learning models (such as algorithms based on clustering or classification) or rule-based judgment methods to improve the sensitivity and accuracy of detection.

[0100] Next, the system performs trend prediction on the cleaned data. Trend prediction is based on time series analysis algorithms, such as the autoregressive moving average (ARIMA) model, long short-term memory (LSTM) network, etc., combined with the current real-time data and historical data collected to predict the future trend of key parameters. For example, by analyzing the historical data of the current packaging speed and material tension, the tension changes at several points in the future can be predicted. This prediction can provide advance adjustment for the dynamic adjustment of equipment operating parameters to avoid the impact of sudden abnormalities on the equipment.

[0101] Finally, the system generates a characteristic parameter set based on the cleaned data, detected abnormal information and predicted trend data. The characteristic parameter set includes key parameters and trend information that can reflect the current packaging conditions, such as the current optimal packaging speed, the current temperature control range, the dynamic adjustment limit of tension, etc. These characteristic parameters are transmitted to the subsequent adaptive control algorithm module through a unified data interface, laying a data foundation for dynamically adjusting the equipment operation parameters.

[0102] Through the above implementation, step S102 can extract valuable information from the collected dynamic data in an efficient and reliable manner, ensuring that the packaging equipment can perform precise adaptive control according to actual working conditions and achieve high-quality and high-stability packaging operations.

[0103] Furthermore, the collected dynamic data is analyzed in multiple dimensions to generate a set of characteristic parameters related to the current packaging working conditions, including:

[0104] Perform preliminary filtering on the raw data collected by the sensor to remove invalid data caused by sensor fluctuations or environmental interference;

[0105] The continuously collected data is smoothed by the time window sliding method to reduce noise interference;

[0106] Different types of dynamic data are classified and stored.

[0107] First, the system will perform preliminary filtering on the raw data collected by the sensor. This process mainly targets invalid data introduced by sensor fluctuations or external environmental interference, such as sudden speed anomalies, transient noise signals in temperature control data, and erroneous output values ​​of tension sensors due to overload. The system identifies these abnormal data through preset rules and threshold ranges and removes them to ensure the basic reliability of subsequent data processing.

[0108] Subsequently, the dynamic data that has been initially filtered will be smoothed using the time window sliding method. Specifically, the sliding window calculates the continuously collected data according to a set time length (such as several sampling cycles) and reduces the impact of noise interference by taking an average or weighted value. For example, there may be small fluctuations in the packaging speed data due to mechanical vibration. Smoothing can eliminate these irregular high-frequency components, making the data more stable and suitable for analysis. Similarly, after sliding window smoothing, the temperature and tension data can better reflect the real trend of the system operation status rather than short-term interference.

[0109] After smoothing, the system stores different types of dynamic data in categories. The classification process groups the data according to its physical characteristics and purpose. For example, packaging speed, tension, and temperature control data are stored in corresponding cache areas, while sealing-related data are stored separately for more targeted analysis and optimization. The purpose of classified storage is to ensure that the data flow is clearly organized, facilitate quick access during multi-dimensional analysis, and avoid confusion or processing delays between different data.

[0110] Through these steps, the collected dynamic data can be analyzed and processed in an efficient and reliable manner, laying a solid data foundation for the subsequent extraction of characteristic parameters and optimization of equipment operating parameters. This multi-dimensional analysis method significantly improves the accuracy of data and the adaptability of system operation, enabling the equipment to cope with complex and changing packaging conditions.

[0111] Furthermore, the collected dynamic data is analyzed in multiple dimensions to generate a set of characteristic parameters related to the current packaging working conditions, including:

[0112] By comparing real-time dynamic data with historical operation data, abnormal values ​​that deviate from the preset range can be identified;

[0113] For mutation data points, interpolation algorithms are used to generate correction values ​​to ensure data continuity;

[0114] The data stream is checked item by item according to the preset anomaly rules, and the detected anomaly marks are stored for system alarm or further analysis.

[0115] By performing multi-dimensional analysis on the collected dynamic data, the system can generate a set of characteristic parameters that are closely related to the current packaging conditions. First, the system collects dynamic data such as the packaging speed, temperature control parameters, tension, and sealing of the equipment in real time, and compares it with the stored historical operation data. The comparison process is based on a preset reference range and is used to identify abnormal values ​​that exceed a reasonable range. For example, during the packaging process, the tension data collected in real time may fluctuate significantly in a short period of time. By comparing the normal range in the historical data, the system can quickly determine that the fluctuation is an abnormal phenomenon.

[0116] After identifying the outliers, the system further processes these mutation data points to ensure the continuity of the data stream. Specifically, for abnormal data points that are obviously deviated, the system calculates reasonable correction values ​​through an interpolation algorithm. The interpolation algorithm is based on the normal data before and after the mutation point, and uses linear interpolation or polynomial interpolation methods to generate alternative values. For example, in the temperature control parameters, if the temperature data at a certain moment shows an abnormal jump due to sensor interference, the interpolation algorithm will calculate the correction data closest to the true value based on the temperature change trend before and after the moment, thereby replacing the abnormal point. The corrected data stream avoids the interference of abnormal data on subsequent analysis and parameter optimization, ensuring the continuity and reliability of the system processing data.

[0117] After completing the interpolation correction, the system checks the data stream item by item according to the preset exception rules. These exception rules may include whether the packaging speed exceeds the safety range, whether the temperature control parameters are stable in the target range, whether the tension meets the mechanical properties requirements of the material, etc. For each data item, the system verifies whether it complies with the rules item by item. For example, when the temperature control parameter is detected to deviate from the target temperature range for several consecutive times, the system will immediately mark the parameter as abnormal and store it in the exception log. At the same time, for persistent or serious abnormal phenomena, such as tension data frequently approaching the material breakage threshold, the system will trigger an alarm signal to remind the operator to check the equipment or process settings.

[0118] Finally, through these analysis steps, the system not only generates a high-quality feature parameter set, but also records all detected abnormal data and their processing records, providing an important basis for subsequent system optimization and troubleshooting. Through the combined effects of comparison, correction and abnormal marking, this method ensures the accuracy and continuity of data analysis and improves the adaptive ability and operational stability of fully automatic packaging equipment.

[0119] Furthermore, the collected dynamic data is analyzed in multiple dimensions to generate a set of characteristic parameters related to the current packaging working conditions, including:

[0120] Based on the time series model of historical collected data, the packaging speed, temperature change trend and material stretchability change are predicted;

[0121] Combined with current real-time data, the weight of trend forecasts is dynamically adjusted to enhance sensitivity to short-term changes;

[0122] Generate a set of characteristic parameters containing future change trends of key parameters for subsequent optimization and adjustment of equipment operating parameters.

[0123] Based on the combination of historical data and real-time dynamic data, multi-dimensional analysis is performed to generate a set of characteristic parameters related to the current packaging conditions. This process includes three closely related steps, which involve time series model prediction, dynamic weight adjustment, and the generation of characteristic parameter sets.

[0124] First, the system uses historical collected data to build a time series model to predict the changing trends of key parameters such as packaging speed, temperature, and material stretchability. The time series model extracts the periodic characteristics and laws of parameter changes by analyzing long-term accumulated data. For example, the time series model of packaging speed can reveal the speed change pattern of the equipment under different load conditions, the temperature model can reflect the response characteristics of the temperature control device in the sealing area to different operating environments, and the material stretchability model reveals the time evolution of material performance degradation through historical fatigue data. These models are established through regression analysis, smoothing filtering or dynamic curve fitting techniques to ensure that the changing trends of various parameters in the future can be accurately predicted.

[0125] Based on the prediction of the time series model, combined with the current real-time collected data, the system dynamically adjusts the weight of the prediction results to enhance the model's sensitivity to short-term changes. Real-time data provides the latest information on the current packaging conditions, such as instantaneous changes in packaging speed or real-time fluctuations in temperature control data. When real-time data deviates from historical trends, the system will reduce the weight of historical data in the time series model and give priority to real-time data to generate prediction results that are more in line with the current state. Specifically, if real-time monitoring shows that the packaging speed fluctuates rapidly due to load changes, the system will reallocate the prediction weights so that short-term predictions more accurately reflect this dynamic change. In addition, in the temperature control system, when real-time temperature data fluctuates abnormally, the system will give priority to real-time collected trends rather than long-term historical patterns to ensure accurate control of sealing temperature.

[0126] Finally, combining the analysis of the time series model prediction results and dynamic weight adjustment, the system generates a set of characteristic parameters that contain future trends in key parameters. This set includes not only the data on current packaging speed, temperature, and material stretchability, but also the range of future parameter changes based on the forecast. For example, if the forecast shows that the packaging speed may gradually decrease due to the increase in equipment load, the system will generate recommendations for optimizing operating parameters based on this to prevent speed fluctuations from affecting packaging quality. Similarly, if the predicted trend of material stretchability shows a decrease, the system will optimize operating conditions in advance by adjusting the tension control parameters to avoid material breakage or loose packaging. In addition, the predicted trend of temperature control parameters may prompt operators to prevent sealing defects caused by temperature out of range.

[0127] Through the above steps, this embodiment achieves accurate prediction of future change trends of key parameters and generates a set of characteristic parameters for real-time optimization. This process effectively combines the long-term regularity of historical data with the dynamic sensitivity of real-time data, enabling the equipment to more intelligently adapt to complex and changing packaging conditions, thereby improving the stability and efficiency of the packaging process.

[0128] Step S103: using a data-driven adaptive control algorithm to dynamically adjust the equipment operating parameters according to the characteristic parameter set, the equipment operating parameters including speed adjustment parameters, temperature control parameters and tension control parameters.

[0129] In step S103, by utilizing a data-driven adaptive control algorithm, the operating parameters of the equipment, including speed regulation parameters, temperature control parameters, and tension control parameters, are dynamically adjusted according to the characteristic parameter set generated in step S102.

[0130] First, the adaptive control algorithm receives a set of characteristic parameters and prioritizes these parameters. Depending on the requirements of the packaging task, such as high-speed packaging, specific material processing, or high sealing requirements, the system assigns different weights to key operating parameters such as speed, temperature control, and tension. The weight allocation can be based on pre-established parameter priority rules or can be dynamically adjusted based on changes in real-time operating conditions. For example, in a high-speed packaging scenario, the system will prioritize speed regulation parameters, while when processing heat-sensitive materials, it will prioritize temperature control parameters.

[0131] Next, the system enters the operating parameter optimization stage. In this stage, the adaptive control algorithm uses control logic based on multivariate regression analysis or machine learning models to establish a mapping relationship between key parameters in the characteristic parameter set and operating parameters. For example, there may be a linear or nonlinear relationship between packaging speed and material stretchability, while temperature control parameters may need to consider the combined effects of ambient temperature and material sealing. The algorithm calculates the optimal operating parameter combination under the current working conditions through iterative optimization. This process may use methods such as gradient descent or heuristic optimization algorithms to quickly converge to the optimal solution under complex multivariate conditions.

[0132] Subsequently, the optimized operating parameters are transmitted to the corresponding actuators through the control interface of the equipment. For example, the speed adjustment parameters are transmitted to the drive motor control unit to achieve precise adjustment of the packaging speed; the temperature control parameters are transmitted to the control unit of the heating device to accurately adjust the temperature of the heating area; the tension control parameters are transmitted to the tension adjustment device to dynamically adjust the tension of the packaging material to ensure the smoothness of the packaging process and the integrity of the material.

[0133] After the operating parameters are actually applied, the system dynamically tracks the operating status of the equipment through the real-time monitoring module and feeds the new data back to the adaptive control algorithm. If it is detected that the adjustment effect of the operating parameters does not meet expectations, the algorithm will adjust the parameters according to the real-time feedback data, thus forming a closed-loop control system. This closed-loop control can respond quickly when the packaging conditions change and continuously optimize the operating performance of the equipment.

[0134] Through the above detailed operation steps, step S103 realizes the dynamic adjustment of the equipment operating parameters, so that the equipment can be optimized in real time according to the current working conditions and packaging requirements, ensuring the dual improvement of packaging efficiency and quality.

[0135] Furthermore, the data-driven adaptive control algorithm dynamically adjusts the equipment operating parameters according to the characteristic parameter set, and the equipment operating parameters include speed adjustment parameters, temperature control parameters and tension control parameters, including:

[0136] The collected characteristic parameter sets are classified by data type, including real-time dynamic parameters and predictive parameters, and different weight factors are assigned; wherein the real-time dynamic parameters include packaging speed, temperature, tension; the predictive parameters include sealing trend and material fatigue curve;

[0137] Based on the classified characteristic parameters, the interaction model between parameters is constructed, including the coupling model between packaging speed and tension, and the correlation model between temperature and material stretchability. The optimal coupling relationship between parameters is obtained through multivariate regression analysis.

[0138] According to the current packaging task requirements and material characteristics, the weight distribution of each parameter in the interactive model is adjusted in real time;

[0139] Combining the interaction model and dynamic weights, the optimal values ​​of the equipment operating parameters are calculated in stages. Specifically, in the first stage, the current characteristic parameters are preliminarily optimized to ensure that the packaging quality meets the basic requirements; in the second stage, based on real-time feedback data, the parameters are iteratively optimized to improve packaging efficiency; in the third stage: through predictive parameter analysis, key operating parameters are adjusted in advance to avoid potential abnormalities;

[0140] The optimized operating parameters are input into the equipment control unit to adjust the speed, temperature control and tension of the equipment in real time, and the adjustment effect is verified through the online monitoring system.

[0141] First, the collected characteristic parameter sets are classified into real-time dynamic parameters and predictive parameters according to the data type, and different weight factors are assigned to each. Real-time dynamic parameters include key parameters such as packaging speed, temperature and tension that directly affect the current packaging process. These parameters are updated in real time through high-frequency sampling and have significant timeliness. Predictive parameters are generated based on historical data and trend analysis, such as sealing trends and material fatigue curves, which are used to infer possible working conditions in the future.

[0142] Packaging speed refers to the speed at which fully automatic packaging equipment completes packaging actions within a unit of time, usually measured by the number of packages completed per minute or the linear speed of the conveyor belt. It is one of the important indicators of packaging efficiency and is directly related to the speed of production. The packaging speed needs to be dynamically adjusted according to the characteristics of the packaging material and the specific working conditions. For example, for flexible materials, if the speed is too high, it may cause uneven tension or material damage, while if the speed is too low, it will affect production efficiency. Therefore, it is necessary to continuously optimize the speed adjustment parameters during the packaging process to achieve efficient and stable operation.

[0143] Temperature is the actual working temperature of the heating area of ​​the fully automatic packaging equipment during the sealing or processing process. For packaging materials that need to be heat-sealed, temperature directly affects the sealing effect. Too low a temperature may result in a loose seal, air or liquid leakage, while too high a temperature may damage the packaging material or even cause it to lose functionality. The temperature control parameters need to be adjusted in real time according to the thermal sensitivity of the material and the sealing process requirements. For example, the temperature of the heat-sealing area must be precisely controlled to ensure that the packaging material achieves the desired bonding effect when sealed without damaging its physical structure.

[0144] Tension refers to the pulling force on the packaging material during the stretching process, which is usually dynamically adjusted through a tension control device. Tension is an important parameter to ensure packaging quality, especially in the film rolling, winding and sealing processes. If the tension is too low, the packaging material may become wrinkled or loose, affecting the packaging appearance and sealing; if the tension is too high, it may cause the material to break or affect the reliability of its subsequent processing. Tension control needs to be adjusted in real time according to the changes in the tensile properties of the material and the packaging speed to ensure that the material is always in the optimal tension state in each process.

[0145] Sealing trend refers to the predictive description of the change of packaging sealing effect over time or operating conditions. It analyzes dynamic data related to sealing performance (such as temperature, pressure, material properties, etc.), combines historical data and real-time monitoring results, and predicts whether the sealing status of the package will gradually deteriorate or remain stable.

[0146] Sealing trends are used to detect problems that could lead to sealing defects in advance. For example, insufficient temperature may result in an incomplete seal, while excessive temperature may damage the packaging material. By analyzing the sealing trend, operating parameters can be adjusted (such as increasing sealing pressure or adjusting temperature control parameters) before the sealing performance deteriorates, avoiding the production of defective products.

[0147] The material fatigue curve describes the law of how the performance of packaging materials gradually changes over time or the number of uses after being subjected to multiple mechanical stresses or environmental influences (such as stretching, compression, temperature changes). This is a key parameter for evaluating the durability and adaptability of packaging materials.

[0148] Material fatigue curves are used to evaluate whether packaging materials can maintain their original performance under long-term use or extreme conditions. For example, under high tension or frequent operation conditions, the stretchability of some materials may decrease, thus affecting the tightness and integrity of the package. By understanding the fatigue characteristics of the material, appropriate tension or speed parameters can be selected during operation to avoid material breakage or premature failure.

[0149] The system assigns different weight factors to each type of parameter. The weight of real-time dynamic parameters is usually higher to ensure that current production needs are met first, while the weight of predictive parameters will be dynamically adjusted according to the importance of the working conditions.

[0150] Then, based on the classified characteristic parameters, the system constructs interaction models between parameters through multivariate regression analysis. These models reflect the dependencies and interactions between different operating parameters. For example, there is a coupling relationship between packaging speed and tension. An increase in speed may cause tension fluctuations, so the tension compensation value required for speed adjustment needs to be calculated through the model; the correlation model between temperature and material stretchability can help determine the optimal heating temperature range to avoid the impact of overheating or cooling of the material on packaging quality. Through these interaction models, the system can generate the optimal coupling relationship between operating parameters to ensure the accuracy of adjustments.

[0151] Next, the system adjusts the weight distribution of each parameter in the interactive model in real time according to the current packaging task requirements and material characteristics. The dynamic adjustment of weights is based on real-time analysis of task priorities and material characteristics. For example, in high-speed packaging tasks, the weight of packaging speed may be increased, while when processing materials that are sensitive to temperature control, the weight of temperature control parameters will be increased accordingly. Such dynamic adjustments make the control algorithm more adaptable and able to respond flexibly under different working conditions.

[0152] After the weight distribution is completed, the system enters the operation parameter optimization stage. The optimization process is divided into three stages and proceeds step by step. In the first stage, the system preliminarily optimizes the current characteristic parameters and calculates the basic operation parameters to ensure that the packaging quality meets the minimum requirements. In the second stage, iterative optimization is performed based on real-time feedback data, and the operation parameters are further adjusted to improve packaging efficiency. In the third stage, combined with the analysis results of predictive parameters, the system adjusts key operation parameters in advance. For example, when it detects that the sealing trend may decrease, it increases the tension or reduces the packaging speed in advance to avoid potential abnormalities.

[0153] Finally, the optimized operating parameters are transmitted to the control unit of the equipment for real-time adjustment of the equipment's packaging speed, temperature control parameters, and tension control. After the adjustment is completed, the system verifies the adjustment effect through the online monitoring system to ensure that the actual operation results are consistent with expectations. If the adjustment effect is found to be unsatisfactory, the system will automatically trigger a new round of optimization process to form a closed-loop control.

[0154] Furthermore, the interaction model between the parameters is constructed based on the classified characteristic parameters through multivariate regression analysis, including:

[0155] The coupling model of packaging speed and tension is expressed by the following formula 1:

[0156] ;

[0157] in, Indicates tension; Indicates packaging speed; Represents the material fatigue curve characteristics; is the frequency factor; , , and is the regression coefficient; is the material characteristic constant;

[0158] The correlation model between temperature and material stretchability is expressed using the following formula 2:

[0159] ;

[0160] in, is the temperature control parameter; Represents the material fatigue curve characteristics; is the regulating factor; is the material response frequency; Indicates the sealing tendency characteristics; , , and is the regression coefficient.

[0161] In formula 1, Indicates tension, which is measured in real time by a strain gauge sensor installed in the equipment, and its unit is usually Newton (N). Tension is a key parameter in the packaging process to ensure material stability and packaging quality.

[0162] is the packaging speed, which refers to the running speed of the conveyor belt or packaging machinery, usually obtained by a rotary encoder or linear speed sensor, in meters per second ( ). It directly affects the tension change of the material.

[0163] is the regression coefficient of the square term of packaging speed, which indicates the quadratic effect of packaging speed on tension. Through multivariate regression analysis, the experimentally measured tension and speed data are fitted to obtain the specific value of this coefficient.

[0164] It is the regression coefficient of the interaction term between speed and material fatigue properties, describing the contribution of the interaction between packaging speed and material fatigue properties to tension. Its value is calculated by fitting the tension performance of different materials at various speeds recorded in the experiment.

[0165] The fatigue curve characteristic of a material is a function that describes the degradation of the mechanical properties of a material under multiple stretches or loads. In the experiment, the material sample is subjected to multiple stretch tests, its stress-strain relationship is recorded, and a fatigue curve is established.

[0166] It is a material property constant used to describe the material's fatigue resistance. Its value is determined experimentally, for example, statistically obtained from tension decay tests of different packaging materials.

[0167] is the regression coefficient of the speed dynamic oscillation term, which indicates the amplitude strength of the tension fluctuation caused by the packaging speed. This value is obtained by fitting the tension and speed oscillation data after spectrum analysis.

[0168] is the frequency factor, describing the effect of the dynamic frequency of the packaging speed on the tension, usually extracted from the speed sensor data by Fast Fourier Transform (FFT).

[0169] The basic tension offset is the basic value of the tension of the device when there is no dynamic input. This parameter is obtained by statically testing the tension of the device.

[0170] In formula 2, Indicates the temperature control parameter, which is measured in real time by the thermocouple or infrared temperature sensor configured in the device, in degrees Celsius ( ). It reflects the operating temperature of the sealing area.

[0171] It is the regression coefficient of the square term of fatigue characteristics, which is used to describe the influence of material fatigue characteristics on temperature control requirements. This parameter is fitted by measuring the sealing effect of materials under different fatigue states.

[0172] It is the square of the material fatigue characteristics, which amplifies the influence of fatigue characteristics on temperature control parameters. Its value is directly calculated from fatigue curve data.

[0173] It is an adjustment factor used to balance the influence of material fatigue characteristics on temperature control parameters. This value is obtained through experimental adjustment.

[0174] is the regression coefficient of the sealing trend characteristic term, which describes the contribution of the sealing trend to the temperature control parameters. This value is obtained by fitting by monitoring the relationship between the sealing effect and the temperature control parameters.

[0175] It is a sealing trend characteristic, which indicates the trend of sealing effect over time or working conditions. The unit is dimensionless. It is calculated by measuring the sealing effect in real time through pressure sensor or vacuum sensor and combining with historical data.

[0176] It is the regression coefficient of the material response frequency term, which indicates the influence of the material dynamic response frequency on the temperature control demand. This value is obtained by fitting the thermal response performance of the material at different frequencies through dynamic testing.

[0177] It is the material response frequency, which is used to describe the response characteristics of the material under dynamic loading. Its characteristic frequency is usually extracted through experiments.

[0178] The temperature control basic offset is the initial temperature control value of the device under standard conditions and is measured through the device calibration process.

[0179] Furthermore, the weight distribution of each parameter in the interactive model is adjusted in real time according to the current packaging task requirements and material characteristics, including:

[0180] According to the following formula 3, the regression coefficient Adjust and obtain new regression coefficients :

[0181] ;

[0182] in, Indicates the current speed; Indicates the target speed;

[0183] According to the following formula 4, the regression coefficient Adjust and obtain new regression coefficients :

[0184] ;

[0185] in, Represents the fatigue curve characteristics of the reference material; is the material characteristic constant; Represents the material fatigue curve characteristics;

[0186] According to the following formula 5, the regression coefficient Adjust and obtain new regression coefficients :

[0187] ;

[0188] in, is the frequency factor; Indicates the current speed.

[0189] In this embodiment, in order to adapt to the current packaging task requirements and changes in material properties, the system optimizes the operating parameters of the packaging equipment by dynamically adjusting the key regression coefficients in the interaction model. The adjustment method includes real-time calculations based on formulas 3, 4, and 5. The adjustment logic and physical meaning of each parameter are described in detail below.

[0190] First, the regression coefficient is the weight of the square term of packaging speed in the tension model, which indicates the intensity of the nonlinear effect of speed on tension. According to formula 3, The adjustment is based on the current speed and target speed :

[0191] ;

[0192] in, It is the adjustment sensitivity coefficient, which is used to control the sensitivity of speed change to weight adjustment. The target speed is obtained by real-time measurement of the speed sensor on the conveyor belt. It is the ideal running speed value set according to the current packaging task requirements. For example, in a high-speed packaging task, when Close to or exceed When The value of is increased to strengthen the influence of the speed square term in the tension model and ensure that the system prioritizes optimizing speed-related tension control.

[0193] Secondly, the regression coefficient is the weight of the interaction term between packaging speed and material fatigue properties, describing the synergistic effect between speed and material fatigue performance. According to Formula 4, the adjustment logic is based on the current material fatigue properties. and reference fatigue properties Deviation:

[0194] ;

[0195] in, Indicates the fatigue characteristics of the material at the current time point, which is obtained by measuring the material fatigue curve calibrated by the experiment; It is a reference characteristic that matches the design performance of the material; is a material property constant that reflects the rate at which the material's performance decays under fatigue conditions. For example, when the actual material fatigue property Deviation from reference characteristics When it is larger, the formula will decrease value, thereby weakening the interaction between velocity and fatigue characteristics to adapt to the dynamic changes of current material properties.

[0196] Finally, the regression coefficient is the weight of the speed dynamic oscillation term, which indicates the influence of speed fluctuation on tension change. According to formula 5, the adjustment logic is based on the current speed and frequency factor :

[0197] ;

[0198] in, It is the frequency factor, extracted from the speed data through Fast Fourier Transform (FFT), reflecting the characteristic frequency of the packaging speed fluctuation. The speed sensor measures the speed in real time. The formula directly maps the amplitude of dynamic speed fluctuations to the weight adjustment by introducing a sine function. For example, in a high-frequency speed change scenario, the formula will dynamically amplify value, thereby improving the system's ability to respond to tension changes caused by speed fluctuations.

[0199] After the above adjustment, the regression coefficient , and The dynamic optimization enables the model to reflect the changes in packaging task requirements and material properties in real time. This adjustment mechanism ensures the adaptability of the tension model, thereby improving the operating efficiency and quality consistency of packaging equipment under complex working conditions. All parameters are calculated through real-time combination of sensor data, experimental measurements and task requirements, with good operability and accuracy.

[0200] Furthermore, the weight distribution of each parameter in the interactive model is adjusted in real time according to the current packaging task requirements and material characteristics, including:

[0201] According to the following formula 6, the regression coefficient Adjust and obtain new regression coefficients :

[0202] ;

[0203] in, Represents the material fatigue curve characteristics; is the regulating factor; is the sensitivity coefficient; Represents the fatigue curve characteristics of the reference material;

[0204] According to the following formula 7, the regression coefficient Adjust and obtain new regression coefficients :

[0205] ;

[0206] in, is the sealing trend sensitivity coefficient; Indicates the reference tightness trend; Indicates the sealing tendency characteristics;

[0207] According to the following formula 8, the regression coefficient Adjust and obtain new regression coefficients :

[0208] ;

[0209] in, is the material response frequency; Represents the material fatigue curve characteristics; Represents the fatigue curve characteristics of the reference material.

[0210] In this embodiment, in order to adapt to the current packaging task requirements and changes in material properties, the system dynamically adjusts the key regression coefficients in the interaction model. , and , and the real-time optimization of the relationship between temperature control parameters and material stretchability and sealing is achieved. The following is a detailed explanation of formulas 6, 7 and 8, explaining the meaning of each parameter and how to obtain it.

[0211] First, the regression coefficient It represents the weight of the square term of material fatigue characteristics in the temperature control model, reflecting the influence of material fatigue characteristics on temperature control requirements. According to formula 6, the adjusted The calculation is as follows:

[0212] ;

[0213] In the formula, It is the current fatigue property of the material. The fatigue curve measured experimentally indicates the degree of performance degradation of the material after multiple stretching, usually obtained by stress-strain testing; It is a reference material fatigue characteristic, reflecting the best performance curve of the material under the design state. It is an adjustment factor that describes the sensitivity of material fatigue properties to temperature control requirements. Its value is obtained by experimental calibration. Used to enlarge or reduce the correlation between fatigue properties and temperature control requirements, obtained by fitting historical data. For example, when the fatigue performance of the material decreases Significantly lower than The adjustment term in the formula will increase , to reflect the strengthening effect of fatigue characteristics on temperature control requirements.

[0214] Secondly, the regression coefficient The weight that describes the influence of the sealing trend characteristics on the temperature control parameters is especially important in tasks with high sealing performance requirements. According to formula 7, the adjusted The calculation is as follows:

[0215] ;

[0216] In the formula, It represents the sealing trend characteristic, which is measured in real time by the pressure sensor or vacuum sensor of the equipment, and the unit is dimensionless.

[0217] It is a reference to the sealing trend, indicating the sealing effect of the equipment under ideal working conditions.

[0218] It is the sealing trend sensitivity coefficient, which is used to adjust the influence of sealing trend changes on temperature control parameters. It is usually determined through statistical analysis of historical sealing data.

[0219] For example, when the current sealing trend Significantly higher than the reference value When value to ensure that the temperature control parameters can meet higher sealing performance requirements.

[0220] Finally, the regression coefficient Reflects the weight of the influence of the material dynamic response frequency on the temperature control parameters, especially when the material stretchability is affected by high-frequency dynamics. According to formula 8, the adjusted The calculation is as follows:

[0221] ;

[0222] In the formula, It is the material response frequency, which represents the characteristic frequency of the material under dynamic loading conditions, and is usually obtained by analyzing tensile test data using Fast Fourier Transform (FFT).

[0223] and are the fatigue properties of the current material and the reference fatigue properties respectively. Inverse tangent function Limit the influence of dynamic frequency to a reasonable range to ensure the numerical stability of the model. For example, when the current fatigue characteristics of the material With reference value When the frequency response is significantly enhanced, the formula will dynamically adjust The value of reflects the strengthening effect of frequency on temperature control requirements.

[0224] Through , and The system can optimize the temperature control model in real time to adapt to different task requirements and changes in material properties. This adjustment mechanism provides reliable data support through real-time sensor measurement, experimental calibration and model fitting, ensuring the operating stability and efficiency of packaging equipment under complex working conditions.

[0225] Step S104: Based on the analysis results of the multidimensional data analysis, identify possible equipment anomalies or process problems in the packaging process, and generate optimization suggestions for operators to refer to or automatically adjust equipment parameters.

[0226] In step S104, through the analysis results based on multi-dimensional data analysis, possible equipment anomalies or process problems in the packaging process are identified, and optimization suggestions are generated for operators to refer to or directly used to automatically adjust equipment parameters. The implementation of this process involves in-depth analysis of data, problem judgment and feedback processing to ensure that packaging equipment can maintain efficient operation under various complex working conditions.

[0227] First, the system receives the optimized operating parameters from the previous step and the real-time collected packaging dynamic data, and performs in-depth comparison and analysis in combination with historical operating data. The system detects abnormal signals in the equipment operation status through built-in abnormal recognition models, such as static models based on rule matching or dynamic models based on machine learning. The judgment criteria for abnormal signals include abnormal fluctuations in key parameters, parameter values ​​outside the normal range, or inconsistencies with historical data trends, such as a sudden decrease in packaging speed, abnormal fluctuations in temperature in the temperature control area, or drastic changes in tension values.

[0228] After identifying an anomaly or potential problem, the system will make a preliminary classification of the cause of the anomaly, such as attributing it to hardware failure, material property differences, or improper process parameter settings. At this point, the analysis module will call the built-in multi-dimensional problem analysis algorithm, combined with the current working conditions and comprehensive data of equipment operation, to infer the possible cause of the problem. For example, if the material tensile data is significantly low and the tension parameter adjustment fails to stabilize, it may be due to batch differences in packaging materials; if the temperature in the temperature control area is continuously high, it may be due to the performance of the heating device deteriorating or the temperature control parameter settings being unreasonable.

[0229] After completing the abnormality identification and cause analysis, the system generates optimization suggestions. These suggestions can include specific parameter adjustment values ​​or equipment operation adjustment plans, such as reducing the packaging speed to a certain range, recalibrating the temperature control device, or replacing the adapter material. These suggestions are provided to the operator through the interactive interface of the equipment, and the operator can manually adjust the equipment operation parameters according to the suggestions. At the same time, for certain specific types of problems, the system can automatically adjust the relevant parameters directly through the closed-loop control mechanism, such as instantly reducing the temperature to avoid the impact of overheating on the material, or dynamically increasing the tension to compensate for the lack of material stretchability.

[0230] In addition, to ensure that the adjusted operating status meets actual needs, the system will continue to monitor the feedback data of the equipment operation after the adjustment is completed. If the problem is not completely solved or a new problem arises, the system will re-enter the problem identification and optimization process until the equipment operating status is stable. Through this iterative optimization mechanism, the system can achieve continuous adaptive adjustment under changing working conditions.

[0231] Through the above steps, step S104 can efficiently identify equipment anomalies or process problems in the packaging process, provide accurate optimization solutions and implement dynamic adjustments, thereby improving the operating stability of the equipment and the consistency of packaging quality.

[0232] A second embodiment of the present application provides an electronic device, the electronic device comprising:

[0233] processor;

[0234] The memory is used to store a program. When the program is read and executed by the processor, it executes an adaptive control method for fully automatic packaging equipment based on dynamic data acquisition provided in the first embodiment of the present application.

[0235] The third embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, an adaptive control method for fully automatic packaging equipment based on dynamic data acquisition provided in the first embodiment of the present application is executed.

[0236] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. A method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition, characterized in that: include: The dynamic data in the packaging process are collected by sensors installed in the fully automatic packaging equipment, and the dynamic data include packaging speed, temperature, material stretchability and sealing property; Perform multi-dimensional analysis on the collected dynamic data to generate a set of characteristic parameters related to the current packaging working conditions, wherein the multi-dimensional analysis includes data cleaning, anomaly detection and trend prediction; Using a data-driven adaptive control algorithm, dynamically adjusting equipment operating parameters according to a set of characteristic parameters, the equipment operating parameters including speed adjustment parameters, temperature control parameters and tension control parameters; Based on the analysis results of multi-dimensional data analysis, it identifies possible equipment anomalies or process problems in the packaging process, and generates optimization suggestions for operators to refer to or automatically adjust equipment parameters; The data-driven adaptive control algorithm dynamically adjusts the equipment operating parameters according to the characteristic parameter set, and the equipment operating parameters include speed adjustment parameters, temperature control parameters and tension control parameters, including: The collected characteristic parameter sets are classified according to data types, including real-time dynamic parameters and predictive parameters, and different weight factors are assigned; wherein the real-time dynamic parameters include packaging speed, temperature, and tension; the predictive parameters include sealing trend and material fatigue curve; Based on the classified characteristic parameters, the interaction model between parameters is constructed through multivariate regression analysis, including the coupling model of packaging speed and tension, and the correlation model of temperature and material stretchability; According to the current packaging task requirements and material characteristics, the weight distribution of each parameter in the interactive model is adjusted in real time; Combining the interaction model and dynamic weights, the optimal values ​​of the equipment operating parameters are calculated in stages. Specifically, in the first stage, the current equipment operating parameters are preliminarily optimized to ensure that the packaging quality meets the basic requirements; in the second stage, based on real-time feedback data, the equipment operating parameters are iterated to improve packaging efficiency; in the third stage: through predictive parameter analysis, the equipment operating parameters are adjusted in advance to avoid potential abnormalities; The optimized operating parameters are input into the equipment control unit to adjust the speed, temperature control and tension of the equipment in real time, and the adjustment effect is verified through the online monitoring system.

2. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 1 is characterized in that: The sensors provided in the fully automatic packaging equipment are used to collect dynamic data during the packaging process, including: During the operation of the equipment, the packaging speed is monitored in real time by a speed sensor configured on the conveyor belt or the driving device, and its signal is converted into a digital signal and transmitted to the data processing unit; Arrange thermocouples or infrared temperature sensors near the heating device to collect temperature data of the sealing area in real time and transmit it to the central control system synchronously; Install a strain gauge sensor in the tension control device to dynamically monitor the stretching state of the packaging material; Pressure or vacuum sensors are installed in the sealing area to monitor the stability of the seal and generate a corresponding data stream.

3. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 2 is characterized in that: The method of collecting dynamic data during the packaging process by using sensors arranged in the fully automatic packaging equipment also includes: After the sensor collects dynamic data, the original data is cleaned through the built-in noise filtering module of the device to remove mutation points and invalid values; Use data smoothing algorithm to process continuous data to ensure the stability of data flow; Before the data is transmitted to the central control system, the collected packaging speed, temperature, stretchability and sealing data are verified for validity according to the preset threshold range, and the verified data is cached in the dynamic data processing module.

4. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 1 is characterized in that: The multi-dimensional analysis of the collected dynamic data generates a set of characteristic parameters related to the current packaging working conditions, including: Perform preliminary filtering on the raw data collected by the sensor to remove invalid data caused by sensor fluctuations or environmental interference; The continuously collected data is smoothed by the time window sliding method to reduce noise interference; Different types of dynamic data are classified and stored.

5. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 1 is characterized in that: The multi-dimensional analysis of the collected dynamic data generates a set of characteristic parameters related to the current packaging working conditions, including: By comparing real-time dynamic data with historical operation data, abnormal values ​​that deviate from the preset range can be identified; For mutation data points, interpolation algorithms are used to generate correction values ​​to ensure data continuity; The data stream is checked item by item according to the preset anomaly rules, and the detected anomaly marks are stored for system alarm or further analysis.

6. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 1 is characterized in that: The multi-dimensional analysis of the collected dynamic data generates a set of characteristic parameters related to the current packaging working conditions, including: Based on the time series model of historical collected data, the packaging speed, temperature change trend and material stretchability change are predicted; Combined with current real-time data, the weight of trend forecasts is dynamically adjusted to enhance sensitivity to short-term changes; Generate a set of characteristic parameters containing future change trends of key parameters for subsequent optimization and adjustment of equipment operating parameters.

7. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 1 is characterized in that: The interaction model between the parameters is constructed based on the classified characteristic parameters through multivariate regression analysis, including: The coupling model of packaging speed and tension is expressed by the following formula 1: ； in, Indicates tension; Indicates packaging speed; Represents the material fatigue curve characteristics; is the frequency factor; , is the regression coefficient; is the material characteristic constant; is the basic tension offset; The correlation model between temperature and material stretchability is expressed using the following formula 2: ； in, is the temperature control parameter; Represents the material fatigue curve characteristics; is the regulating factor; is the material response frequency; Indicates the sealing tendency characteristics; , and is the regression coefficient; is the temperature control basic offset.

8. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 7 is characterized in that: The weight distribution of each parameter in the interactive model is adjusted in real time according to the current packaging task requirements and material characteristics, including: According to the following formula 3, the regression coefficient Adjust and obtain new regression coefficients : ； in, Indicates the current speed; Indicates the target speed; According to the following formula 4, the regression coefficient Adjust and obtain new regression coefficients : ； in, Represents the fatigue curve characteristics of the reference material; is the material characteristic constant; Represents the material fatigue curve characteristics; According to the following formula 5, the regression coefficient Adjust and obtain new regression coefficients : ； in, is the frequency factor; Indicates the current speed.

9. The method for adaptive control of fully automatic packaging equipment based on dynamic data acquisition according to claim 7 is characterized in that: The weight distribution of each parameter in the interactive model is adjusted in real time according to the current packaging task requirements and material characteristics, including: According to the following formula 6, the regression coefficient Adjust and obtain new regression coefficients : ； in, Represents the material fatigue curve characteristics; is the regulating factor; is the sensitivity coefficient; Represents the fatigue curve characteristics of the reference material; According to the following formula 7, the regression coefficient Adjust and obtain new regression coefficients : ； in, is the sealing trend sensitivity coefficient; Indicates the reference tightness trend; Indicates the sealing tendency characteristics; According to the following formula 8, the regression coefficient Adjust and obtain new regression coefficients : ； in, is the material response frequency; Represents the material fatigue curve characteristics; Represents the fatigue curve characteristics of the reference material.

Images