An intelligent bale quality hub control system based on AI dynamic analysis

Abstract

The embodiment of the application discloses a kind of based on AI dynamic analysis's roll package quality intelligent hub control system.Quality monitoring module is used to set each monitoring time according to preset time interval and the production condition and each roll package parameter in each monitoring time are monitored;Defect identification module is used to determine whether roll package quality is qualified, if roll package quality is unqualified, then with production condition as premise, the inducing root cause of different roll package defect types is analyzed, and according to the use length of each equipment in workshop, the analysis result is compensated, and roll package defect is identified;Hub control module is used to obtain the identification result of roll package defect and obtain the equipment distribution in current workshop from database, according to the equipment distribution in current workshop and roll package defect type, each equipment in workshop is synergistically controlled.Through the technical scheme of the embodiment of the application, the accuracy of roll package defect type and roll package defect type inducing cause identification is guaranteed, and the effectiveness of equipment control and the quality of roll package are also guaranteed.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of roll and package quality monitoring and control technology, and in particular to an intelligent central control system for roll and package quality based on AI dynamic analysis. Background Technology

[0002] The cigarette rolling and packaging process is the core of cigarette production, directly determining the final product's quality stability, brand image, and the company's economic benefits. With the digital transformation and intelligent upgrading of the tobacco industry, cigarette production is showing a trend towards higher speed, greater precision, and greater flexibility. The industry's requirements for controlling the finished product's appearance quality, process stability, and production efficiency are continuously increasing.

[0003] Traditional intelligent central control systems for package quality rely on correlation analysis to determine the causes of different package defect types and identify these defects based on the analysis results and monitoring data. When a package defect occurs, the system adjusts the equipment corresponding to the defect type. Clearly, this type of intelligent central control system for package quality has at least the following shortcomings.

[0004] Firstly, traditional intelligent central control systems for package quality rely on correlation analysis to determine the causes of different package defect types, failing to identify the root causes. While abnormal parameters may frequently occur alongside certain package defect types, this doesn't necessarily mean the abnormality of a single parameter will induce that specific defect type. The root cause of the defect might be triggering the parameter abnormality, and workshop production conditions are constantly changing. The root causes of each package defect differ under different production conditions, and these root causes can shift as equipment usage time increases. Traditional technologies overlook this, thus failing to guarantee the accuracy of identifying package defect types and their causes, consequently compromising package quality.

[0005] Secondly, traditional intelligent central control systems for package quality lack analysis of the impact of package defect types on other equipment. Package defects not only affect their corresponding equipment but also other equipment. This impact has a certain lag, and the affected objects vary depending on the distribution of equipment. Traditional technology only regulates the equipment corresponding to the package defect type that has already occurred, thus failing to guarantee the effectiveness of equipment regulation and consequently, the quality of the package. Summary of the Invention

[0006] To address the aforementioned technical shortcomings, this invention provides an intelligent central control system for package quality based on AI dynamic analysis.

[0007] To address the aforementioned technical problems, embodiments of the present invention provide an intelligent central control system for package quality based on AI dynamic analysis, comprising: a quality monitoring module, a defect identification module, a central control module, and a database; The quality monitoring module is used to set each monitoring time according to a preset time interval, and to monitor the production conditions of the workshop and the parameters of each roll package at each monitoring time. The defect identification module is used to determine whether the quality of the roll package is qualified. If the quality of the roll package is not qualified, the root causes of different types of roll package defects are analyzed based on the production conditions. The analysis results are compensated according to the usage time of each piece of equipment in the workshop, and the roll package defects are identified at the same time. The central control module is used to obtain the identification results of the packaging defects, and to obtain the current equipment distribution in the workshop from the database. Based on the current equipment distribution in the workshop and the type of packaging defects, the module coordinates and controls each piece of equipment in the workshop. The database is used to store information from each historical monitoring session, the distribution of equipment in the workshop, the equipment corresponding to each type of package defect, and the current distribution of equipment in the workshop.

[0008] Optionally, the defect identification module includes: a quality analysis unit, a root cause database, and a defect identification unit; the quality analysis unit is used to obtain the production conditions and package parameter values ​​at each monitoring time to determine whether the package quality meets the standards; the root cause database is used to obtain information from historical monitoring times from the database, analyze the root causes of different package defect types under different production conditions based on the information from historical monitoring times, and compensate for the root causes based on the usage time of each device to construct the root cause database; the defect identification unit is used to match the current workshop production conditions, package parameter values, and device usage time with the root cause database to obtain the current package defect type.

[0009] Optionally, the production conditions and packaging quality inspection reports of the workshop during each historical monitoring session are retrieved from the database, and the packaging quality inspection results are obtained from the packaging quality inspection reports. Historical monitoring sessions where the packaging quality inspection results indicate unqualified packaging quality are recorded as marked historical monitoring sessions. The production conditions of the workshop during all marked historical monitoring sessions are compared, and marked historical monitoring sessions with the same production conditions are grouped into the same historical monitoring group to obtain each historical monitoring group. Within each historical monitoring group, the root causes of each packaging defect type under different production conditions are analyzed. Based on the usage time of each piece of equipment during each marked historical monitoring session, compensation is made for the root causes of each packaging defect type under different production conditions. Based on the analysis results and compensation results, the root causes of different production conditions, equipment usage time, and each packaging defect type are stored one-to-one to form a root cause database.

[0010] Optionally, for all historical monitoring groups, obtain the historical monitoring data corresponding to each package defect type, and obtain the abnormal package parameters from the corresponding package quality inspection report. Based on the abnormal package parameters of the historical monitoring data corresponding to each package defect type, obtain the abnormal package parameter set for each package defect type. For each package defect type, divide the corresponding abnormal package parameter set according to different dimensions to obtain a subset of abnormal package parameters, and remove the abnormal package parameters without causal relationship from the abnormal package parameter set through experiments. The abnormal package parameters retained in the abnormal package parameter set of each package defect type in each historical monitoring group are used as the root causes of each package defect type in each historical monitoring group, and the root causes are used as the root causes of each package defect type under the corresponding production conditions of each historical monitoring group.

[0011] Optionally, a simulation platform is built, and the production conditions and various equipment in the workshop are simulated and 3D modeled in the simulation platform to obtain subsets of abnormal package parameters. In the simulation platform, for all subsets of abnormal package parameters, the values ​​of each abnormal package parameter in the current subset are adjusted, while keeping the values ​​of the remaining abnormal package parameters in the abnormal package parameter set unchanged, and the package defect type is monitored. If a package defect occurs and the package defect type does not change, it is determined that there is no causal relationship between each abnormal parameter in the current subset of abnormal package parameters and the package defect type, and each abnormal package parameter in the current subset of abnormal package parameters is removed from the abnormal package parameter set; otherwise, it is determined that there is a causal relationship.

[0012] Optionally, the abnormal package parameters retained in the abnormal package parameter set for each package defect type in each historical monitoring group are recorded as causal parameters for each package defect type under different production conditions; the equipment corresponding to each package defect type is retrieved from the database; for all historical monitoring groups, the causal parameter values ​​and corresponding equipment usage time for each package defect type at the time of historical monitoring are retrieved from the package quality inspection report at the time of each marked historical monitoring; for each package defect type, the usage time of the corresponding equipment at the time of historical monitoring is compared, and the marked equipment with the same usage time is identified. Historical monitoring data is grouped into the same group to obtain the average value of each causal parameter within each group. This yields the limit values ​​of each causal parameter for each type of package defect under different usage times of the corresponding equipment. The limit values ​​of each causal parameter are then plotted on a curve to obtain the functional relationship between the usage time of the corresponding equipment for each package defect type and its causal parameter limit values. Under different production conditions, compensation is made for the causal parameter limit values ​​of each package defect type based on the functional relationship between the usage time of the corresponding equipment for each package defect type and its causal parameter limit values, as well as the usage time of the corresponding equipment for each package defect type.

[0013] Optionally, the central control module includes: a defect impact analysis unit and a collaborative control unit; the defect impact analysis unit is used to obtain the current package defect type, and predict whether other equipment will be affected by the current package defect type based on the current package defect type and the current equipment distribution in the workshop; the collaborative control unit is used to mark each equipment affected by the current package defect type as a marked equipment, mark the equipment corresponding to the current package defect type as the main equipment, and perform collaborative control on the operation of the main equipment and each marked equipment based on the current equipment distribution in the workshop and the package defect type.

[0014] Optionally, the current package defect type and the usage duration of the main equipment are obtained, and the equipment distribution in the workshop during each historical monitoring is obtained from the database; based on the current package defect type, the equipment distribution in the workshop, and the usage duration of the main equipment, similar historical monitoring data are obtained; each affected device and its corresponding usage duration are obtained from the database, and the affected devices during each similar historical monitoring are compared to obtain a set of affected devices and a usage duration threshold for each device in the set of affected devices; the current usage duration of each device in the set of affected devices is compared with the corresponding usage duration threshold to determine that devices with a current usage duration greater than or equal to the corresponding usage duration threshold are affected by the current package defect type.

[0015] Optionally, the production conditions of the current workshop and the distribution and operation of each piece of equipment are simulated and modeled in three dimensions in the simulation platform, and the abnormal parameters of the current main equipment and each marked equipment are obtained according to the type of package defect. In the simulation platform, several parameter adjustment groups are set according to the abnormal parameters of the current main equipment and each marked equipment, as well as the standard value range of the abnormal parameters of the main equipment and each marked equipment. The parameter adjustment groups are composed of the abnormal parameter values ​​of the main equipment and each marked equipment. Each parameter adjustment group is applied to the three-dimensional model of the simulation platform in sequence for simulation adjustment, and the package parameters are monitored in real time during the simulation adjustment process to obtain the package quality change coefficient under each parameter adjustment group. The package quality change coefficients of each parameter adjustment group are compared, and the abnormal parameter values ​​of the main equipment and each marked equipment in the parameter adjustment group with the largest package quality change coefficient are taken as the target values ​​of the abnormal parameters of the main equipment and each marked equipment. The abnormal parameter values ​​of the current main equipment and each marked equipment are adjusted to the corresponding abnormal parameter target values.

[0016] Optionally, the impact status of each marked device during each similar historical monitoring is obtained from the database, and each similar historical monitoring of each marked device is obtained; each abnormal parameter of each marked device during each similar historical monitoring is obtained from the database, and the obtained abnormal parameters are clustered to obtain each abnormal parameter of each marked device at present; and each abnormal parameter of the current master device is obtained in the same way as the abnormal parameters of each marked device.

[0017] On the other hand, embodiments of the present invention also provide an intelligent central control method for package quality based on AI dynamic analysis, the method comprising: The quality monitoring module sets monitoring times according to preset time intervals, and monitors the production conditions and parameters of each roll package in the workshop at each monitoring time. The defect identification module determines whether the quality of the roll package is qualified. If the quality of the roll package is unqualified, the root causes of different types of roll package defects are analyzed based on the production conditions. The analysis results are compensated according to the usage time of each piece of equipment in the workshop, and the roll package defects are identified at the same time. The identification results of the packaging defects are obtained through the central control module, and the current equipment distribution in the workshop is obtained from the database. Based on the current equipment distribution in the workshop and the type of packaging defects, the equipment in the workshop is coordinated and controlled. The database stores information from each historical monitoring session, the distribution of equipment within the workshop, the equipment corresponding to each type of package defect, and the current distribution of equipment within the workshop.

[0018] On the other hand, embodiments of the present invention also provide an electronic device, the electronic device comprising: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the AI-based dynamic analysis-based intelligent central control method for package quality as provided in any embodiment of the present invention.

[0019] On the other hand, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent central control method for package quality based on AI dynamic analysis as provided in any embodiment of the present invention.

[0020] On the other hand, embodiments of the present invention provide a computer program product, including a computer program that, when executed by a processor, implements the AI-based dynamic analysis-based intelligent central control method for package quality as provided in any embodiment of the present invention.

[0021] The technical solution of this invention achieves real-time data coverage of the entire production process by automatically collecting production conditions and packaging parameters at preset time intervals through a quality monitoring module. This provides a timely and comprehensive data foundation for quality judgment, avoiding the lag and bias of manual sampling. The defect identification module analyzes the root causes of quality defects based on production conditions and compensates for them according to equipment usage time when quality is deemed unqualified. This effectively eliminates the interference of differences in production conditions and equipment aging drift on root cause identification, significantly improving the accuracy and reliability of defect type and cause judgment. The central control module performs coordinated control based on the current equipment distribution and defect type, not only controlling the equipment directly corresponding to the defect but also predicting and controlling other potentially affected equipment. This overcomes the insufficient effectiveness of traditional single control methods and ensures the overall stability of packaging quality. The database categorizes and stores historical monitoring information, equipment distribution, and defect correspondence, providing comprehensive data support for online decision-making by each module.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the structure of an AI-based dynamic analysis-based intelligent central control system for package quality provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of an intelligent central control method for package quality based on AI dynamic analysis provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device that implements the AI-based dynamic analysis-based intelligent central control method for package quality according to embodiments of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] Example 1 Figure 1 This is a schematic diagram of a roll packaging quality intelligent central control system based on AI dynamic analysis, provided in Embodiment 1 of the present invention. Figure 1 As shown, the system includes: a quality monitoring module, a defect identification module, a central control module, and a database.

[0028] The quality monitoring module is used to set monitoring times according to preset time intervals and monitor the production conditions and packaging parameters of the workshop at each monitoring time. The defect identification module is used to determine whether the packaging quality is qualified. If the packaging quality is unqualified, it analyzes the root causes of different packaging defect types based on the production conditions and compensates for the analysis results according to the usage time of each piece of equipment in the workshop, while identifying packaging defects. The central control module is used to obtain the identification results of packaging defects and obtain the current equipment distribution in the workshop from the database. Based on the current equipment distribution in the workshop and the packaging defect type, it coordinates and controls each piece of equipment in the workshop. The database is used to store information from each historical monitoring session, the equipment distribution in the workshop, the equipment corresponding to each packaging defect type, and the current equipment distribution in the workshop.

[0029] In this embodiment, the quality monitoring module refers to a functional unit responsible for automatically collecting data on the workshop production environment and equipment operating status according to preset rules. The time interval refers to the length of time between two adjacent monitoring sessions. The time interval can be preset by staff based on the daily production status of the workshop. The monitoring time refers to a specific point in time divided according to the time interval. Data collection is performed at this point in time. The workshop refers to the physical location where rolling and packaging production takes place. The workshop contains various production equipment and operators. Production conditions refer to the general term for various environmental and material factors that affect the stability of the rolling and packaging production process and product quality. For example, production conditions may include the workshop environment (temperature, humidity, dust concentration, etc.), equipment status within the workshop (vibration amplitude, equipment temperature, pressure, etc.), and material properties (tobacco moisture content, tobacco filling value, etc.). Rolling and packaging parameters refer to quantifiable indicators that reflect the equipment operating status or product quality characteristics during the rolling and packaging production process. For example, rolling and packaging parameters may include tobacco moisture content, tobacco gun temperature, tobacco flow rate, instantaneous tobacco delivery flow rate, and nozzle adhesive flow rate, etc.

[0030] In this embodiment, the defect identification module can refer to a functional unit that analyzes monitoring data, determines whether the product quality is qualified, and identifies the defect type and its cause when it is unqualified. Packaging quality can refer to the comprehensive condition of whether the finished cigarettes meet preset standards in terms of appearance, physical indicators, and draw resistance. Unqualified packaging quality can refer to a situation where the monitored value of a certain packaging parameter deviates from its standard value range, thus determining that the quality is unqualified. Using production conditions as a premise can refer to using the current specific production environment conditions as the basis for analyzing the causes of defects. This is because the causes of the same defect may differ under different production conditions. Packaging defect type can refer to the specific manifestation of the finished cigarettes not meeting quality standards. For example, packaging defect types may include empty cigarette ends, excessive cigarette circumference, cigarette weight deviation, cigarette bursting, and poor sealing. The root cause can refer to the direct causal factor that truly leads to the packaging defect, rather than merely related factors accompanying the defect. Various equipment in the workshop can refer to various mechanical equipment involved in packaging production in the workshop. For example, equipment may include conveying mechanisms, hopper mechanisms, paper winding mechanisms, and pressure mechanisms. Usage time can refer to the cumulative operating time of the equipment from its introduction into the production line to the present moment. Analysis results can refer to the preliminary conclusions regarding the root causes obtained by the defect identification module during the analysis process.

[0031] In this embodiment, the central control module can refer to a functional unit that formulates and executes equipment control schemes based on defect identification results. The identification result corresponds to the current packaging defect type, i.e., the final determined defect type. The current equipment distribution within the workshop can refer to the spatial layout of each piece of equipment within the workshop at the current moment, as well as their material flow and process connection relationships. Coordinated control can refer to the simultaneous joint adjustment of the operating parameters of multiple related equipment, rather than independent control of only a single piece of equipment. The database can refer to a storage unit used for persistently storing various historical data and configuration information. Historical monitoring records can refer to the collection records of all past monitoring moments. Information can refer to the data recorded during each monitoring session. Information may include production conditions, packaging quality inspection reports, etc. The equipment distribution within the workshop during historical monitoring sessions can refer to the spatial layout of equipment recorded at each past monitoring moment. The equipment corresponding to the packaging defect type can refer to equipment that has a direct causal relationship with the specific defect type, i.e., the abnormal operation of this equipment is the direct cause of the defect.

[0032] Specifically, the quality monitoring module automatically collects data at various monitoring times using various sensors and instruments. For example, it monitors the workshop environment using temperature sensors, humidity sensors, high-precision absolute pressure sensors, and online laser dust detectors; it monitors equipment status using vibration sensors, temperature transmitters, and pressure transmitters; it obtains material properties from material suppliers; it monitors the moisture content of tobacco using a near-infrared moisture meter; it monitors the temperature of the tobacco gun using a platinum resistance temperature sensor; it monitors the flow rate of tobacco using an electronic belt scale; and it monitors the flow rate of the nozzle adhesive using a micro-flow ultrasonic flow meter. The preset time intervals are determined by staff based on production rhythm and control requirements.

[0033] Specifically, the defect identification module determines whether the package quality is acceptable by retrieving the standard value range of each package parameter under the current production conditions from the database. It then compares the monitored parameter values ​​with the corresponding standard ranges one by one. If any value exceeds the standard range, the quality is deemed unacceptable. When unacceptable, the defect identification module analyzes the root cause of each defect type under the current production conditions, and incorporates equipment usage time to correct for the root cause. This is because equipment aging can lead to performance drift, and the critical value for the same parameter to trigger defects differs at different usage times. After compensation, the module outputs the accurate defect type.

[0034] Specifically, after obtaining the identification results, the central control module reads the current device distribution data (i.e., the spatial location and connection relationship of each device) from the database. Since device distribution directly affects the propagation path of defects, the module predicts other devices that may be affected based on the distribution and defect type, and then performs joint parameter adjustments on all predicted affected devices. The database stores all of the above data in a structured manner, providing data support for each stage of monitoring, identification, and control.

[0035] The technical solution of this invention achieves real-time data coverage of the entire production process by automatically collecting production conditions and packaging parameters at preset time intervals through a quality monitoring module. This provides a timely and comprehensive data foundation for quality judgment, avoiding the lag and bias of manual sampling. The defect identification module analyzes the root causes of quality defects based on production conditions and compensates for them according to equipment usage time when quality is deemed unqualified. This effectively eliminates the interference of differences in production conditions and equipment aging drift on root cause identification, significantly improving the accuracy and reliability of defect type and cause judgment. The central control module performs coordinated control based on the current equipment distribution and defect type, not only controlling the equipment directly corresponding to the defect but also predicting and controlling other potentially affected equipment. This overcomes the insufficient effectiveness of traditional single control methods and ensures the overall stability of packaging quality. The database categorizes and stores historical monitoring information, equipment distribution, and defect correspondence, providing comprehensive data support for online decision-making by each module.

[0036] As an optional implementation of this disclosure, the defect identification module includes: a quality analysis unit, a root cause database, and a defect identification unit. The defect identification module is used to determine whether the package quality is qualified. If the package quality is unqualified, it analyzes the root causes of different package defect types based on production conditions and compensates the analysis results according to the usage time of each piece of equipment in the workshop. Simultaneously, it identifies package defects. Specifically, the module includes: the quality analysis unit acquiring production conditions and package parameter values ​​at each monitoring time to determine whether the package quality meets standards; the root cause database acquiring information from historical monitoring times, analyzing the root causes of different package defect types under different production conditions based on the historical monitoring information, and compensating the root causes based on the usage time of each piece of equipment to construct the root cause database; and the defect identification unit matching the current workshop production conditions, package parameter values, and equipment usage time with the root cause database to obtain the current package defect type.

[0037] In this embodiment, the quality analysis unit can refer to a functional submodule responsible for performing quality conformity determination. The values ​​of each package parameter can refer to the specific values ​​of the package parameters collected at each monitoring time. The root cause database can refer to a pre-built knowledge base that stores the true causes (i.e., causal parameters) of each defect type under different production conditions and different equipment usage durations. The defect identification unit can refer to a submodule responsible for matching the current online monitoring data with the root cause database to determine the current defect type. The current workshop production conditions are the same as the aforementioned production conditions, but specifically refer to the conditions at the current moment. The current package defect type is equivalent to the identification result.

[0038] Specifically, the quality analysis unit obtains the production conditions and packaging parameter values ​​at each monitoring time from the quality monitoring module, and reads the standard value range of each parameter under the current production conditions from the database. After comparing them one by one, it outputs a judgment result of whether the standard is met or not. When the result is not met, the subsequent analysis process is triggered.

[0039] Specifically, the construction of the root cause database is an offline data mining process, as follows: All historical monitoring data is extracted from the database. For each non-conforming record, the causal parameters that truly led to the defect under the given production conditions are analyzed. Then, the critical values ​​of the causal parameters are adjusted based on the usage duration of each piece of equipment at that time. Finally, the mapping relationship between production conditions, usage duration, defect type, and causal parameters is stored as a knowledge base. Once completed, this database is used for online identification and matching.

[0040] Specifically, after the quality analysis unit determines that a defect is non-conforming, the defect identification unit uses the current production conditions, the parameter values ​​of each package, and the usage time of each piece of equipment as query conditions to search for matching entries in the root cause database. For example, it first filters out the corresponding causal parameter limits for each defect type based on the current production conditions and usage time, then compares the current parameter values ​​of each package with these limits to identify parameters that exceed the limits, and then matches these parameters with the causal parameters of each defect type in the database to determine the current defect type.

[0041] This optional implementation decouples quality judgment, knowledge accumulation, and defect identification by refining the defect identification module into three independent but collaborative units, allowing each step to be optimized independently and improving the maintainability and scalability of the system.

[0042] As an optional implementation of this disclosure, the process of constructing the root cause database is as follows: The production conditions and packaging quality inspection reports of the workshop during each historical monitoring session are retrieved from the database, and the packaging quality inspection results are obtained from the packaging quality inspection reports. Historical monitoring sessions with packaging quality inspection results indicating unqualified packaging quality are recorded as marked historical monitoring sessions. The production conditions of the workshop during all marked historical monitoring sessions are compared, and marked historical monitoring sessions with the same production conditions are grouped into the same historical monitoring group to obtain each historical monitoring group. The root causes of each packaging defect type under different production conditions are analyzed within each historical monitoring group. Based on the usage time of each piece of equipment during each marked historical monitoring session, compensation is performed for the root causes of each packaging defect type under different production conditions. Based on the analysis results and compensation results, the root causes of different production conditions, equipment usage time, and each packaging defect type are stored one-to-one to form the root cause database.

[0043] In this embodiment of the disclosure, the production conditions of the workshop at each historical monitoring time can refer to the production environment data recorded at each past monitoring moment. The package quality inspection report can refer to the record file generated after quality inspection of the packaged finished product. The package quality inspection report can include information such as inspection results (pass / fail), defect type (if any), abnormal parameters, operating time of each piece of equipment, and affected status. The package quality inspection result can refer to the pass / fail judgment conclusion in the report. Marked historical monitoring can refer to those historical monitoring records with unqualified quality inspection results. The historical monitoring group can refer to the set of all marked historical monitoring records with the same production conditions. The compensation result can refer to the precipitating root cause parameter limit after correction for equipment usage time.

[0044] Specifically, the construction process first extracts all historical monitoring records and their corresponding package quality inspection reports from the database, filtering out all non-conforming records as marked historical monitoring data. These marked records are then clustered according to production conditions. Since changes in production conditions can alter defect causes, records with similar conditions are grouped together. Within each historical monitoring group, causal analysis methods (described later) are used to determine the true root cause (rather than related factors) of each defect type under specific production conditions. Next, because different equipment usage durations lead to varying degrees of equipment wear, thus affecting the critical values ​​of causal parameters, the analyzed root cause parameters are corrected and compensated based on the equipment usage duration recorded during each marked monitoring session. Finally, the combination of production conditions and usage duration is used as a composite key, mapped one-to-one with the corresponding defect type and causal parameters, forming a root cause database that can be quickly queried online.

[0045] This construction method distinguishes between two dimensions: production conditions and usage duration, enabling the knowledge base to have refined matching capabilities and laying a data foundation for subsequent accurate identification.

[0046] As an optional implementation of this disclosure, the specific process for analyzing the root causes of various package defect types under different production conditions is as follows: For all historical monitoring groups, obtain the historical monitoring data corresponding to each package defect type, and obtain the abnormal package parameters from the corresponding package quality inspection report. Based on the abnormal package parameters of the historical monitoring data corresponding to each package defect type, obtain the abnormal package parameter set for each package defect type. For each package defect type, divide the corresponding abnormal package parameter set according to different dimensions to obtain a subset of abnormal package parameters, and remove the abnormal package parameters without causal relationship from the abnormal package parameter set through experiments. The abnormal package parameters retained in the abnormal package parameter set of each package defect type in each historical monitoring group are used as the root causes of each package defect type in each historical monitoring group, and the root causes are used as the root causes of each package defect type under the corresponding production conditions of each historical monitoring group.

[0047] In this embodiment of the disclosure, an abnormal package parameter can refer to a package parameter whose value deviates from its standard value range. An abnormal package parameter set can refer to a set of all abnormal package parameters that have appeared in historical monitoring for a specific defect type (which may include accompanying parameters that are not causally related to the defect). An abnormal package parameter subset can refer to a combination of parameters obtained by dividing the set according to dimensions (one-dimensional, two-dimensional, three-dimensional, etc.), with the highest dimension determined by the total number of parameters in the set. An experiment can refer to a testing process that verifies whether there is a causal relationship between the parameter and the defect through simulation. A causal association can refer to a direct causal relationship between the abnormal parameter and the occurrence of the defect, that is, the abnormal parameter is the direct cause of the defect.

[0048] Specifically, for each historical monitoring group, for each defect type, all marked records of the defect that have occurred are collected. All abnormal parameters are extracted from the detection reports of these records and compiled into an initial set. This set contains both the true cause and accompanying parameters that may only co-occur with the defect. To eliminate accompanying parameters, this set is split into multiple subsets by dimension (e.g., a one-dimensional subset of a single parameter, a two-dimensional subset of a combination of two parameters, etc.). Then, experiments are conducted to verify whether the parameters in each subset have a causal relationship with the defect. That is, in the simulation platform, the parameters of that subset are made abnormal while other parameters are kept normal, and it is observed whether the defect is triggered. If it is triggered and the defect type remains unchanged, it is retained; otherwise, it is removed. The parameters that are ultimately retained are the true root causes.

[0049] By verifying the results experimentally rather than through simple correlation analysis, the causal nature of the root cause identification was ensured, and the misjudgment of accompanying phenomena as the cause was avoided.

[0050] As an optional implementation of this disclosure, the experimental process is as follows: A simulation platform is built, and production conditions and various equipment in the workshop are simulated and modeled in three dimensions within the simulation platform to obtain subsets of abnormal package parameters. Within the simulation platform, for all subsets of abnormal package parameters, the values ​​of each abnormal package parameter in the current subset are adjusted, while keeping the remaining abnormal package parameter values ​​in the abnormal package parameter set unchanged. Package defect types are monitored. If a package defect occurs and the package defect type remains unchanged, it is determined that there is no causal relationship between the abnormal parameters in the current subset and the package defect type, and the abnormal package parameters in the current subset are removed from the abnormal package parameter set. Otherwise, a causal relationship is determined.

[0051] In this embodiment of the disclosure, the simulation platform can refer to a digital platform built using simulation software that can virtually reproduce the workshop production environment and equipment operation. Simulation can refer to constructing a mathematical model and logical rules for the production process within the platform. Three-dimensional modeling can refer to constructing a three-dimensional geometric model of the equipment to realistically reflect its spatial location and interrelationships. The current subset of abnormal packet parameters can refer to the set of abnormal parameters currently being tested. The numerical value of the abnormal packet parameter can refer to the specific deviation value. The remaining abnormal packet parameters can refer to other abnormal parameters in the set besides the currently tested subset. Abnormal parameters are equivalent to abnormal packet parameters, referring to parameters that deviate from the standard range.

[0052] Specifically, firstly, a platform is built using simulation software (selected by staff). This platform establishes production condition parameters, 3D models of each piece of equipment, and their operational logic consistent with the actual workshop. Predefined subsets of abnormal packaging parameters are then input into the platform. Each subset is then tested one by one. Specifically, within the platform, only the parameter values ​​in the current subset are adjusted to be abnormal (deviating from the normal range), while other parameters in the subset remain normal. The simulation is run, and the occurrence and type of packaging defects are observed. If the investigated defect type appears under these conditions (without changing the type), it indicates that the parameters in that subset are sufficient conditions for inducing the defect, i.e., a causal relationship exists, and it is retained. If no defect appears, or other types of defects appear, it indicates that the subset is not a sufficient condition, and no causal relationship exists, and it is discarded. By traversing all subsets, the final parameter combination retained is the true root cause.

[0053] This experimental method uses a simulation platform to replace the real production line for destructive testing, which avoids affecting actual production, allows for precise control of variables, and effectively verifies causal relationships.

[0054] As an optional implementation of this disclosure, the specific process for compensating for the root causes of each type of package defect under different production conditions is as follows: Each abnormal package parameter retained in the abnormal package parameter set of each package defect type in each historical monitoring group is recorded as a causal parameter for each package defect type under different production conditions; the equipment corresponding to each package defect type is retrieved from the database; for all historical monitoring groups, the causal parameter values ​​and corresponding equipment usage time for each package defect type at the marked historical monitoring time are retrieved from the package quality inspection report at each marked historical monitoring time; for each package defect type, the corresponding equipment at each marked historical monitoring time... By comparing the usage time, historical monitoring data of devices with the same usage time are grouped together to obtain each group. The average value of each causal parameter within each group is calculated to obtain the limit value of each causal parameter for each package defect type under different usage times of its corresponding device. The limit values ​​of each causal parameter are plotted on a curve to obtain the functional relationship between the usage time of the device corresponding to each package defect type and its limit values ​​of each causal parameter. Under different production conditions, based on the functional relationship between the usage time of the device corresponding to each package defect type and its limit values ​​of each causal parameter, and the usage time of the device corresponding to each package defect type, compensation is made for the limit values ​​of each causal parameter for each package defect type.

[0055] In this embodiment, the causal parameter can refer to an abnormal packet parameter that has been experimentally verified to have a causal relationship with the defect type. Each group can refer to a marked historical monitoring set divided according to the same device usage duration. The causal parameter limit can refer to the critical value at which a defect is triggered when the causal parameter is exceeded for a specific defect type under a specific usage duration of its corresponding device. The graph can refer to a function graph plotted with usage duration as the horizontal axis and the parameter limit as the vertical axis. The functional relationship can refer to the mathematical correspondence between usage duration and the parameter limit. Exemplarily, the functional relationship can be obtained by fitting a fully connected neural network.

[0056] Specifically, the compensation process aims to establish a quantitative relationship between equipment usage time and causal parameter limits, enabling dynamic adjustment of limits under different usage periods. First, the retained causal parameters are formally named. Then, the specific equipment corresponding to each defect type (e.g., the conveyor mechanism corresponding to excessive cigarette circumference) is retrieved from the database. For all historical monitoring groups, the actual values ​​of the causal parameters and the corresponding actual usage time of the equipment are extracted from the inspection reports of each marked record. For each defect type, all records with the same usage time are grouped together, and the average value of each causal parameter within each group is calculated as the limit for that usage time. These (duration, limit) points are plotted on a graph, and a fully connected neural network is used to train and fit the functional relationship between duration and limit. For example, the fully connected neural network process involves: standardizing the graph and inputting it into the network; after training, it outputs a function expression. Once this function is obtained, under different production conditions, the actual usage time of the current equipment is substituted into the function to calculate the compensated limit, thereby eliminating the impact of equipment aging on root cause analysis.

[0057] This compensation method quantifies the impact of equipment aging on defect causes through a data-driven approach, enabling the root cause database to have time-adaptive capabilities and improving the accuracy of the identification model throughout its entire lifecycle.

[0058] As an optional implementation of this disclosure, the central control module includes: a defect impact analysis unit and a collaborative control unit; the central control module is used to obtain the identification result of the packaging defect, and obtain the current equipment distribution in the workshop from the database, and to perform collaborative control on each piece of equipment in the workshop according to the current equipment distribution in the workshop and the packaging defect type, which may specifically include: the defect impact analysis unit is used to obtain the current packaging defect type, and predict whether other equipment will be affected by the current packaging defect type according to the current packaging defect type and the current equipment distribution in the workshop; the collaborative control unit is used to mark each piece of equipment affected by the current packaging defect type as each marked piece of equipment, mark the equipment corresponding to the current packaging defect type as the main piece of equipment, and perform collaborative control on the operation of the main piece of equipment and each marked piece of equipment according to the current equipment distribution in the workshop and the packaging defect type.

[0059] In this embodiment, the defect impact analysis unit can refer to a functional submodule responsible for assessing the potential scope of a defect. Other equipment can refer to all equipment in the workshop other than the main equipment. The collaborative control unit can refer to a functional submodule responsible for formulating and implementing a joint control plan. The marked equipment can refer to equipment predicted to be affected by the defect. The main equipment can refer to the equipment directly corresponding to the defect, i.e., the malfunction of this equipment is the direct cause of the defect.

[0060] Specifically, after obtaining the current defect type, the defect impact analysis unit, in conjunction with the spatial layout of various equipment and material flow paths within the workshop, analyzes how the defect might affect other equipment through material transfer, process linkages, etc. Based on the analysis results, the collaborative control unit marks the affected equipment and designates the equipment corresponding to the defect as the primary equipment. Then, a joint control plan covering the primary equipment and all marked equipment is formulated, simultaneously adjusting the operating parameters of these devices to eliminate the defect and prevent its spread.

[0061] This optional implementation expands the control scope from a single device to multiple devices by introducing impact analysis and coordinated control, thus making up for the shortcomings of traditional solutions that ignore the chain reaction of defects.

[0062] As an optional implementation of this disclosure, the specific process for predicting whether other devices will be affected by the current package defect type is as follows: Obtain the current package defect type and the usage duration of the main device, and obtain the device distribution within the workshop during each historical monitoring session from the database; based on the current package defect type, the device distribution within the workshop, and the usage duration of the main device, obtain each similar historical monitoring session; obtain each affected device and its corresponding usage duration from the database, and compare each affected device during each similar historical monitoring session to obtain the affected device set and the usage duration threshold for each device in the affected device set; compare the current usage duration of each device in the affected device set with the corresponding usage duration threshold, and determine that devices with a current usage duration greater than or equal to the corresponding usage duration threshold are affected by the current package defect type.

[0063] In this embodiment of the disclosure, similar historical monitoring can refer to monitoring records in historical records that are identical to the current situation in terms of defect type, production conditions, equipment distribution, and main equipment usage time. The affected equipment set can refer to the aggregated set of all affected equipment that has appeared in similar historical monitoring. The usage time threshold can refer to the minimum usage time threshold at which equipment begins to be affected by defects (taking the minimum duration from similar historical monitoring). The current usage time of the equipment is equivalent to the current usage duration, which can refer to the cumulative operating time of the equipment at the current moment.

[0064] Specifically, the process begins by obtaining the current defect type and the main device's usage duration, and then retrieving historical device distribution data from the database. Next, using the current defect type, current device distribution, and main device usage duration as combined criteria, historical monitoring records that perfectly match are selected as similar historical monitoring data. From these records, all devices previously recorded as affected and their corresponding usage durations are extracted and aggregated into a set of affected devices. For each device in the set, the minimum usage duration across all similar records is taken as a threshold, indicating that when a device reaches this usage duration, it is likely to be affected under similar conditions. Finally, the current actual usage duration of each device is compared with the threshold; if it reaches or exceeds the threshold, it is determined that the device is currently affected by the defect.

[0065] By using analogical reasoning based on similar historical cases, the scope of defect impact can be effectively predicted without the need to build complex physical models, making it simple and reliable.

[0066] As an optional implementation of this disclosure, the specific process of coordinating and controlling the operation of the main equipment and each marking device is as follows: Simulate and create a 3D model of the current workshop production conditions and the distribution and operation of each device in a simulation platform, and obtain the abnormal parameters of the current main equipment and each marking device according to the type of package defect; In the simulation platform, based on the abnormal parameters of the current main equipment and each marking device, and the standard value range of the abnormal parameters of the main equipment and each marking device, set several parameter adjustment groups, each parameter adjustment group consisting of the abnormal parameter values ​​of the main equipment and each marking device; Apply each parameter adjustment group sequentially to the 3D model of the simulation platform for simulation adjustment, and monitor each package parameter in real time during the simulation adjustment process to obtain the package quality change coefficient under each parameter adjustment group; Compare the package quality change coefficients of each parameter adjustment group, and take the abnormal parameter values ​​of the main equipment and each marking device in the parameter adjustment group with the largest package quality change coefficient as the target values ​​of the abnormal parameters of the main equipment and each marking device; Adjust the abnormal parameter values ​​of the current main equipment and each marking device to the corresponding target values ​​of the abnormal parameters.

[0067] In this embodiment, the current master equipment can refer to the equipment directly corresponding to the defect at the current moment. The standard value range of abnormal parameters can refer to the allowable normal fluctuation range of each operating parameter of the equipment. The standard value range of abnormal parameters can be preset by personnel according to equipment specifications and production requirements. A parameter adjustment group can refer to a combination scheme that includes specific values ​​of abnormal parameters for the master equipment and all marked equipment. The package quality change coefficient can refer to a quantitative indicator that measures the degree of improvement in package quality relative to the initial state under a certain parameter adjustment scheme. A larger value indicates a more significant improvement. The target value of abnormal parameters can refer to the specific value that each abnormal parameter of the equipment should be adjusted to after optimization.

[0068] Specifically, the current workshop production conditions, equipment distribution, and operating status are first modeled in the simulation platform. Then, abnormal parameters (i.e., parameters deviating from the standard and their deviation values) of the main equipment and each marked equipment are retrieved from the database. For example, different equipment has different parameter types; for instance, conveyor parameters include conveyor line speed, belt speed, and belt tension; hopper parameters include material flow rate and feeding frequency. Next, using the standard value range of these abnormal parameters as constraints (adjusted values ​​must be within the standard range), several different adjustment schemes (parameter adjustment groups) are generated. Each scheme contains a specific combination of abnormal parameters for each equipment. These schemes are then applied sequentially to the simulation platform for simulation operation. During the simulation, changes in each package parameter are monitored, and the package quality variation coefficient under each scheme is calculated based on the monitoring results. For example, the package quality variation coefficient is calculated by normalizing the difference between the monitored values ​​of each package parameter and the median of its standard range, summing the results, and taking the reciprocal of the sum; if the sum is zero, the coefficient is infinite. Compare the coefficients of each group, select the group with the largest coefficient (i.e., the most significant quality improvement), and use the parameter values ​​in this group as the target values. Finally, adjust the abnormal parameters of each piece of equipment to the target values ​​on the actual production line.

[0069] This control method uses a simulation platform to pre-evaluate the effects of multiple schemes and selects the optimal scheme for execution, avoiding blind trial and error on the real production line and improving control efficiency and safety.

[0070] As an optional implementation of this disclosure, the specific process for obtaining the abnormal parameters of the current master device and each tagging device is as follows: obtain the impact status of each tagging device during each similar historical monitoring from the database, and obtain each similar historical monitoring of each tagging device; obtain each abnormal parameter of each tagging device during each similar historical monitoring from the database, and cluster the obtained abnormal parameters to obtain the abnormal parameters of each current tagging device; obtain the abnormal parameters of the current master device in the same way as obtaining the abnormal parameters of each tagging device.

[0071] In this embodiment of the disclosure, the impact status can refer to the status indicator of whether the device is affected by the defect at a specific monitoring time. The impact status can include both "affected" and "unaffected". The currently marked devices can refer to the devices that are currently predicted to be affected by the defect. The same method as obtaining the abnormal parameters of each marked device can refer to using the same clustering processing method to extract the abnormal parameters of the master device from historical data.

[0072] Specifically, the process begins by retrieving the impact status of each device from the database based on historical monitoring data. Records indicating an affected status are then selected, and the devices corresponding to these records are identified as the historical similar monitoring data for the marked devices. Next, a list of abnormal parameters for each marked device at that time is extracted from these records. All historical abnormal parameters are clustered (categorized and summarized by parameter type) to obtain the most likely abnormal parameters for each marked device currently in operation. Abnormal parameters for the main device are obtained using the same clustering method: extracting the main device's abnormal parameters from similar historical monitoring data and then clustering them. Finally, a list of abnormal parameters for all devices requiring current control is obtained.

[0073] This acquisition method utilizes cluster statistics of historical cases to quickly infer abnormal parameters of the current equipment without relying on real-time detection, providing input for the formulation of subsequent control plans.

[0074] The AI-based dynamic analysis-based intelligent central control system for package quality provided in this invention can execute the AI-based dynamic analysis-based intelligent central control method for package quality provided in any embodiment of this invention, and has the corresponding functional modules and beneficial effects of executing the AI-based dynamic analysis-based intelligent central control method for package quality.

[0075] It is worth noting that in the above embodiments of intelligent central control for roll package quality based on AI dynamic analysis, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0076] The above are embodiments of the intelligent central control system for package quality based on AI dynamic analysis provided by the present invention. This system and the intelligent central control method for package quality based on AI dynamic analysis in the following embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the intelligent central control method for package quality based on AI dynamic analysis, please refer to the above embodiments of the intelligent central control system for package quality based on AI dynamic analysis.

[0077] Example 2 Figure 2 This invention provides a flowchart of an AI-based dynamic analysis-based intelligent central control method for package quality in a second embodiment. This embodiment is applicable to situations where real-time quality monitoring, accurate identification of defect root causes, and coordinated control of related equipment are performed during package production. This method can be executed by an AI-based dynamic analysis-based intelligent central control system for package quality. This system can be implemented in hardware and / or software and can be configured in electronic devices. Figure 2 As shown, the method includes: S210. The quality monitoring module sets each monitoring time according to a preset time interval, and monitors the production conditions of the workshop and the parameters of each roll package at each monitoring time.

[0078] This step enables real-time, periodic automatic collection of workshop production environment and equipment operating parameters, ensuring the comprehensiveness and timeliness of monitoring data, avoiding omissions and delays in manual sampling, and providing a complete and reliable data foundation for subsequent quality assessment.

[0079] S220. Determine whether the quality of the roll package is qualified through the defect identification module. If the quality of the roll package is not qualified, analyze the root causes of different types of roll package defects based on the production conditions, and compensate for the analysis results according to the usage time of each piece of equipment in the workshop. At the same time, identify the roll package defects.

[0080] This step, after determining non-compliance, involves conducting root cause analysis under the current production conditions and incorporating equipment usage time to compensate and correct root cause parameters. This effectively eliminates the interference of differences in production conditions and equipment aging and drift on the accuracy of identification, significantly improves the accuracy of defect type and its causes, and ensures the correct direction of subsequent control.

[0081] S230. Obtain the identification results of the packaging defects through the central control module, and obtain the current equipment distribution in the workshop from the database. Based on the current equipment distribution in the workshop and the type of packaging defects, coordinate and control each piece of equipment in the workshop.

[0082] This step predicts other equipment that may be affected by defects based on the current equipment distribution, and formulates a coordinated control plan covering the main equipment and all marked equipment. This overcomes the limitations of traditional single-equipment control, ensures the comprehensiveness and effectiveness of control, prevents the spread of defects, and guarantees the stability of the overall package quality.

[0083] S240. Store information from each historical monitoring session, the distribution of equipment in the workshop, the equipment corresponding to each type of package defect, and the current distribution of equipment in the workshop through a database.

[0084] This step involves structured storage of historical monitoring data, changes in equipment distribution, and corresponding defect relationships, providing data support for all aspects of quality monitoring, defect identification, and collaborative control. It also provides a data foundation for the offline construction and online matching of the root cause database, enabling the accumulation of system knowledge and continuous optimization.

[0085] The technical solution of this invention achieves real-time automatic collection of production data through a quality monitoring module, ensuring the timeliness and comprehensiveness of monitoring; it improves the accuracy and reliability of defect identification by performing condition-limited root cause analysis and introducing usage time compensation through a defect identification module; it ensures the effectiveness of control and overall quality stability through coordinated control based on equipment distribution through a central control module; and it provides data support for system learning and continuous improvement by persistently storing historical data through a database. The entire solution forms a closed-loop intelligent control system from data acquisition, quality judgment, root cause identification, defect location, impact prediction to coordinated control, effectively solving the problems of inaccurate root cause identification and insufficient control range in traditional solutions.

[0086] As an optional implementation of this disclosure, the defect identification module includes: a quality analysis unit, a root cause database, and a defect identification unit. The defect identification module determines whether the package quality is up to standard. If the package quality is not up to standard, it analyzes the root causes of different package defect types based on production conditions and compensates for the analysis results based on the usage time of each piece of equipment in the workshop. Specifically, the module identifies package defects by: obtaining production conditions and package parameter values ​​at each monitoring time through the quality analysis unit to determine whether the package quality meets standards; obtaining information from the database of historical monitoring times through the root cause database, analyzing the root causes of different package defect types under different production conditions based on the historical monitoring information, and compensating for the root causes based on the usage time of each piece of equipment to construct the root cause database; and matching the current workshop production conditions, package parameter values, and equipment usage time with the root cause database through the defect identification unit to obtain the current package defect type.

[0087] As an optional implementation of this disclosure, the process of constructing the root cause database is as follows: The production conditions and packaging quality inspection reports of the workshop during each historical monitoring session are retrieved from the database, and the packaging quality inspection results are obtained from the packaging quality inspection reports. Historical monitoring sessions with packaging quality inspection results indicating unqualified packaging quality are recorded as marked historical monitoring sessions. The production conditions of the workshop during all marked historical monitoring sessions are compared, and marked historical monitoring sessions with the same production conditions are grouped into the same historical monitoring group to obtain each historical monitoring group. The root causes of each packaging defect type under different production conditions are analyzed within each historical monitoring group. Based on the usage time of each piece of equipment during each marked historical monitoring session, compensation is performed for the root causes of each packaging defect type under different production conditions. Based on the analysis results and compensation results, the root causes of different production conditions, equipment usage time, and each packaging defect type are stored one-to-one to form the root cause database.

[0088] As an optional implementation of this disclosure, the specific process for analyzing the root causes of various package defect types under different production conditions is as follows: For all historical monitoring groups, obtain the historical monitoring data corresponding to each package defect type, and obtain the abnormal package parameters from the corresponding package quality inspection report. Based on the abnormal package parameters of the historical monitoring data corresponding to each package defect type, obtain the abnormal package parameter set for each package defect type. For each package defect type, divide the corresponding abnormal package parameter set according to different dimensions to obtain a subset of abnormal package parameters, and remove the abnormal package parameters without causal relationship from the abnormal package parameter set through experiments. The abnormal package parameters retained in the abnormal package parameter set of each package defect type in each historical monitoring group are used as the root causes of each package defect type in each historical monitoring group, and the root causes are used as the root causes of each package defect type under the corresponding production conditions of each historical monitoring group.

[0089] As an optional implementation of this disclosure, the experimental process is as follows: A simulation platform is built, and production conditions and various equipment in the workshop are simulated and modeled in three dimensions within the simulation platform to obtain subsets of abnormal package parameters. Within the simulation platform, for all subsets of abnormal package parameters, the values ​​of each abnormal package parameter in the current subset are adjusted, while keeping the remaining abnormal package parameter values ​​in the abnormal package parameter set unchanged. Package defect types are monitored. If a package defect occurs and the package defect type remains unchanged, it is determined that there is no causal relationship between the abnormal parameters in the current subset and the package defect type, and the abnormal package parameters in the current subset are removed from the abnormal package parameter set. Otherwise, a causal relationship is determined.

[0090] As an optional implementation of this disclosure, the specific process for compensating for the root causes of each type of package defect under different production conditions is as follows: Each abnormal package parameter retained in the abnormal package parameter set of each package defect type in each historical monitoring group is recorded as a causal parameter for each package defect type under different production conditions; the equipment corresponding to each package defect type is retrieved from the database; for all historical monitoring groups, the causal parameter values ​​and corresponding equipment usage time for each package defect type at the marked historical monitoring time are retrieved from the package quality inspection report at each marked historical monitoring time; for each package defect type, the corresponding equipment at each marked historical monitoring time... By comparing the usage time, historical monitoring data of devices with the same usage time are grouped together to obtain each group. The average value of each causal parameter within each group is calculated to obtain the limit value of each causal parameter for each package defect type under different usage times of its corresponding device. The limit values ​​of each causal parameter are plotted on a curve to obtain the functional relationship between the usage time of the device corresponding to each package defect type and its limit values ​​of each causal parameter. Under different production conditions, based on the functional relationship between the usage time of the device corresponding to each package defect type and its limit values ​​of each causal parameter, and the usage time of the device corresponding to each package defect type, compensation is made for the limit values ​​of each causal parameter for each package defect type.

[0091] As an optional implementation of this disclosure, the central control module includes: a defect impact analysis unit and a collaborative control unit; the central control module obtains the identification result of the packaging defect and obtains the current equipment distribution in the workshop from the database; and performs collaborative control on each piece of equipment in the workshop according to the current equipment distribution and the packaging defect type. Specifically, this includes: obtaining the current packaging defect type through the defect impact analysis unit; predicting whether other equipment will be affected by the current packaging defect type based on the current packaging defect type and the current equipment distribution in the workshop; marking each piece of equipment affected by the current packaging defect type as a marked piece of equipment through the collaborative control unit; marking the equipment corresponding to the current packaging defect type as the main piece of equipment; and performing collaborative control on the operation of the main piece of equipment and each marked piece of equipment according to the current equipment distribution and the packaging defect type in the workshop.

[0092] As an optional implementation of this disclosure, the specific process for predicting whether other devices will be affected by the current package defect type is as follows: Obtain the current package defect type and the usage duration of the main device, and obtain the device distribution within the workshop during each historical monitoring session from the database; based on the current package defect type, the device distribution within the workshop, and the usage duration of the main device, obtain each similar historical monitoring session; obtain each affected device and its corresponding usage duration from the database, and compare each affected device during each similar historical monitoring session to obtain the affected device set and the usage duration threshold for each device in the affected device set; compare the current usage duration of each device in the affected device set with the corresponding usage duration threshold, and determine that devices with a current usage duration greater than or equal to the corresponding usage duration threshold are affected by the current package defect type.

[0093] As an optional implementation of this disclosure, the specific process of coordinating and controlling the operation of the main equipment and each marking device is as follows: Simulate and create a 3D model of the current workshop production conditions and the distribution and operation of each device in a simulation platform, and obtain the abnormal parameters of the current main equipment and each marking device according to the type of package defect; In the simulation platform, based on the abnormal parameters of the current main equipment and each marking device, and the standard value range of the abnormal parameters of the main equipment and each marking device, set several parameter adjustment groups, each parameter adjustment group consisting of the abnormal parameter values ​​of the main equipment and each marking device; Apply each parameter adjustment group sequentially to the 3D model of the simulation platform for simulation adjustment, and monitor each package parameter in real time during the simulation adjustment process to obtain the package quality change coefficient under each parameter adjustment group; Compare the package quality change coefficients of each parameter adjustment group, and take the abnormal parameter values ​​of the main equipment and each marking device in the parameter adjustment group with the largest package quality change coefficient as the target values ​​of the abnormal parameters of the main equipment and each marking device; Adjust the abnormal parameter values ​​of the current main equipment and each marking device to the corresponding target values ​​of the abnormal parameters.

[0094] As an optional implementation of this disclosure, the specific process for obtaining the abnormal parameters of the current master device and each tagging device is as follows: obtain the impact status of each tagging device during each similar historical monitoring from the database, and obtain each similar historical monitoring of each tagging device; obtain each abnormal parameter of each tagging device during each similar historical monitoring from the database, and cluster the obtained abnormal parameters to obtain the abnormal parameters of each current tagging device; obtain the abnormal parameters of the current master device in the same way as obtaining the abnormal parameters of each tagging device.

[0095] Example 3 Figure 3A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0096] like Figure 3 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0098] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the intelligent central control method for package quality based on AI dynamic analysis.

[0099] In some embodiments, the AI-based dynamic analysis-based intelligent central control method for package quality can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the AI-based dynamic analysis-based intelligent central control method for package quality described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the AI-based dynamic analysis-based intelligent central control method for package quality by any other suitable means (e.g., by means of firmware).

[0100] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0102] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0103] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0104] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0105] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0106] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the AI-based dynamic analysis-based intelligent central control method for package quality as provided in any embodiment of this application.

[0107] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider). This program product belongs to the same inventive concept as the AI-based dynamic analysis-based intelligent central control method for package quality disclosed in the embodiments of this application, and therefore will not be described further here.

[0108] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and no limitation is imposed herein.

[0109] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A roll-packing quality intelligent central control system based on AI dynamic analysis, characterized in that, include: The system comprises a quality monitoring module, a defect identification module, a central control module, and a database. The quality monitoring module is used to set each monitoring time according to a preset time interval, and to monitor the production conditions of the workshop and the parameters of each roll package at each monitoring time. The defect identification module is used to determine whether the quality of the roll package is qualified. If the quality of the roll package is not qualified, the root causes of different types of roll package defects are analyzed based on the production conditions. The analysis results are compensated according to the usage time of each piece of equipment in the workshop, and the roll package defects are identified at the same time. The central control module is used to obtain the identification results of the packaging defects, and to obtain the current equipment distribution in the workshop from the database. Based on the current equipment distribution in the workshop and the type of packaging defects, the module coordinates and controls each piece of equipment in the workshop. The database is used to store information from each historical monitoring session, the distribution of equipment in the workshop, the equipment corresponding to each type of package defect, and the current distribution of equipment in the workshop.

2. The system according to claim 1, characterized in that, The defect identification module includes: a quality analysis unit, a root cause library, and a defect identification unit; The quality analysis unit is used to obtain the production conditions and roll package parameter values ​​at each monitoring time to determine whether the roll package quality meets the standards. The root cause database is used to obtain information from historical monitoring times in the database, analyze the root causes of different packaging defect types under different production conditions based on the information from historical monitoring times, and compensate for the root causes based on the usage time of each device in order to construct the root cause database. The defect identification unit is used to match the current production conditions of the workshop, the parameter values ​​of each roll package, and the usage time of each device with the root cause database to obtain the current roll package defect type.

3. The system according to claim 2, characterized in that, The process of constructing the inducing root cause library is as follows: Retrieve the production conditions and roll packaging quality inspection reports of the workshop during each historical monitoring from the database, and obtain the roll packaging quality inspection results from the roll packaging quality inspection reports. Record each historical monitoring in which the roll packaging quality inspection result is unqualified as a marked historical monitoring. The production conditions of the workshop during all the marked historical monitoring were compared, and the marked historical monitoring with the same production conditions were divided into the same historical monitoring group to obtain each historical monitoring group. The root causes of each type of package defect under different production conditions were analyzed in each historical monitoring group. Based on the usage time of each device during the historical monitoring of each marker, compensation is made for the root causes of each type of package defect under different production conditions; Based on the analysis and compensation results, the root causes of different production conditions, equipment usage time and various package defect types are stored one-to-one to form a root cause database.

4. The system according to claim 3, characterized in that, The specific process of analyzing the root causes of various package defect types under different production conditions is as follows: For all historical monitoring groups, obtain the historical monitoring of each marker corresponding to each package defect type, and obtain the parameters of each abnormal package from the corresponding package quality inspection report. Based on the parameters of each abnormal package in the historical monitoring of each marker corresponding to each package defect type, obtain the set of abnormal package parameters for each package defect type. For each type of volume packet defect, the corresponding abnormal volume packet parameter set is divided according to different dimensions to obtain each abnormal volume packet parameter subset, and the abnormal volume packet parameters without causal relationship are removed from the abnormal volume packet parameter set through experiments. Each abnormal package parameter retained in the abnormal package parameter set of each package defect type in each historical monitoring group is taken as the root cause of each package defect type in each historical monitoring group, and the root cause is taken as the root cause of each package defect type under the corresponding production conditions in each historical monitoring group.

5. The system according to claim 4, characterized in that, The experiment was conducted as follows: A simulation platform was built, and the production conditions and various equipment in the workshop were simulated and 3D modeled in the simulation platform to obtain a subset of abnormal package parameters; In the simulation platform, for all subsets of abnormal volume packets, the values ​​of each abnormal volume packet parameter in the current subset of abnormal volume packets are adjusted, while keeping the values ​​of the remaining abnormal volume packets parameters in the abnormal volume packet parameter set unchanged, and the type of volume packet defect is monitored. If a package defect occurs and the package defect type does not change, it is determined that there is no causal relationship between each abnormal parameter in the current abnormal package parameter subset and the package defect type, and each abnormal package parameter in the current abnormal package parameter subset is removed from the abnormal package parameter set. Conversely, if the causal relationship is not found, then a causal relationship is determined.

6. The system according to claim 3, characterized in that, The specific process for compensating for the root causes of various package defect types under different production conditions is as follows: Each abnormal package parameter retained in the abnormal package parameter set of each package defect type in each historical monitoring group is recorded as the causal parameter of each package defect type under different production conditions. Retrieve the devices corresponding to each package defect type from the database; For all historical monitoring groups, obtain the values ​​of each causal parameter and the usage time of the corresponding equipment for each mark of defect type of each package during the historical monitoring of each package from the package quality inspection report at the time of each mark's historical monitoring. For each type of package defect, the usage time of the corresponding equipment during the historical monitoring of each mark is compared. The historical monitoring of each mark with the same equipment usage time is divided into the same group to obtain each group. The average value of each causal parameter in each group is calculated to obtain the limit value of each causal parameter of each package defect type under different usage times of its corresponding equipment. The limit value of each causal parameter is plotted on a curve to obtain the functional relationship between the usage time of the corresponding equipment and the limit value of each causal parameter of each package defect type. Under different production conditions, compensation is made for each causal parameter limit for each type of package defect based on the functional relationship between the equipment usage time corresponding to each package defect type and the limit value of each causal parameter, as well as the equipment usage time corresponding to each package defect type.

7. The system according to claim 1, characterized in that, The central control module includes: a defect impact analysis unit and a collaborative control unit; The defect impact analysis unit is used to obtain the current package defect type and, based on the current package defect type and the current equipment distribution in the workshop, predict whether other equipment will be affected by the current package defect type. The coordinated control unit is used to mark each device affected by the current package defect type as a marked device, and mark the device corresponding to the current package defect type as the main device. Based on the current equipment distribution and package defect type in the workshop, the unit coordinates and controls the operation of the main device and each marked device.

8. The system according to claim 7, characterized in that, The specific process for predicting whether other devices will be affected by the current type of package defect is as follows: Obtain the current package defect type and the usage duration of the main equipment, and retrieve the equipment distribution in the workshop from the database for each historical monitoring session; Based on the current package defect type, the equipment distribution in the workshop, and the usage time of the main equipment, obtain various similar historical monitoring data; The database is used to obtain the devices affected during each similar historical monitoring period and their corresponding usage durations. The devices affected during each similar historical monitoring period are compared to obtain the set of affected devices and the usage duration threshold of each device in the set of affected devices. The current usage time of each device in the affected device set is compared with the corresponding usage time threshold. Devices whose current usage time is greater than or equal to the corresponding usage time threshold are affected by the current package defect type.

9. The system according to claim 7, characterized in that, The specific process for coordinated control of the operation of the main device and each marking device is as follows: The simulation platform simulates and models the current production conditions and the distribution and operation of various equipment in the workshop in three dimensions, and obtains the abnormal parameters of the current main equipment and each marked equipment according to the type of package defect. In the simulation platform, based on the abnormal parameters of the current master device and each marker device, and the standard value range of the abnormal parameters of the master device and each marker device, several parameter adjustment groups are set. The parameter adjustment groups are composed of the abnormal parameter values ​​of the master device and each marker device. Each parameter adjustment group is applied sequentially to the three-dimensional model of the simulation platform for simulation adjustment, and each roll package parameter is monitored in real time during the simulation adjustment process to obtain the roll package quality change coefficient under each parameter adjustment group. Compare the roll quality change coefficients of each parameter adjustment group, and take the abnormal parameter values ​​of the main equipment and each marking equipment in the parameter adjustment group with the largest roll quality change coefficient as the target values ​​of each abnormal parameter of the main equipment and each marking equipment. Adjust the values ​​of all abnormal parameters of the current master device and each marked device to the corresponding target values ​​of abnormal parameters.

10. The system according to claim 9, characterized in that, The specific process for obtaining the abnormal parameters of the current master device and each marked device is as follows: Retrieve the impact status of each marked device during each similar historical monitoring from the database, and retrieve each similar historical monitoring data for each marked device; The abnormal parameters of each tagging device in similar historical monitoring are obtained from the database, and the abnormal parameters are clustered to obtain the abnormal parameters of each tagging device at present. Obtain the abnormal parameters of the current master device in the same way as obtaining the abnormal parameters of each tagged device.

Images