Analysis method and system for cigarette pack inspection equipment

Abstract

The invention discloses an analysis method and system for cigarette pack inspection equipment, and the method comprises the steps: S1, obtaining multi-source heterogeneous data in the operation process of the cigarette pack inspection equipment, the multi-source heterogeneous data comprising at least two of equipment production data, equipment state signals, equipment process data, equipment monitoring data and equipment energy consumption data; s2, based on a preset state judgment rule set, processing the multi-source heterogeneous data obtained in the step S1, and automatically identifying and marking the real-time operation state of the equipment; the operation states at least comprise production, idling, standby, shutdown and alarm; s3, the data with the state marks and the multi-source heterogeneous data are stored in an associated mode; and S4, executing at least one correlation analysis based on the data after correlation storage, and outputting decision information for guiding production or equipment maintenance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial data acquisition and intelligent analysis, in particular to an analysis method and system of a cigarette label inspection equipment based on multi-source data fusion. BACKGROUND

[0002] In the packaging and printing industry, the printing quality of cigarette labels (cigarette packaging boxes) is extremely high, so the subsequent inspection process is crucial. At present, most of the cigarette label inspection equipment on the market lacks deep data acquisition and analysis capabilities, and there are the following technical bottlenecks: The existing ordinary data acquisition system can only acquire and display basic data such as production rate and vehicle speed, while the state of the equipment (such as idling, debugging), process parameters (such as template information, air pressure), environmental data (temperature and humidity), and energy consumption data are not effectively acquired or correlated with each other.

[0003] The judgment of the running state of the equipment depends on manual experience, and it is difficult to accurately and automatically distinguish the microscopic states such as "production", "idling", and "standby", which leads to inaccurate calculation of key indicators such as equipment overall efficiency (OEE) and difficulty in quantifying and managing time waste in the production process.

[0004] The acquired data is only used for basic monitoring and recording, and lacks deep analysis. It is difficult to establish a correlation model between product quality defects and production process parameters, it is difficult to realize predictive warning of equipment failure, and it is difficult to accurately calculate energy costs under different production states. Data cannot be converted into effective decision-making information to guide production optimization. SUMMARY

[0005] In order to solve the above problems, the embodiments of the present application provide an analysis method and system of a cigarette label inspection equipment, which aims to realize full-dimensional self-perception of the equipment, and through correlation analysis of multi-source data, to provide data-driven decision support for production efficiency improvement, product quality control, predictive maintenance of equipment, and energy consumption optimization.

[0006] In a first aspect, the embodiments of the present application provide an analysis method of a cigarette label inspection equipment, which comprises: S1, acquiring multi-source heterogeneous data in the running process of the cigarette label inspection equipment, wherein the multi-source heterogeneous data comprises at least two of equipment production data, equipment state signals, equipment process data, equipment monitoring data, and equipment energy consumption data; S2, processing the multi-source heterogeneous data acquired in step S1 based on a preset state judgment rule set, automatically identifying and marking the real-time running state of the equipment; the running state at least includes production, idling, standby, shutdown, and alarm; S3, storing the data with state marks and the multi-source heterogeneous data in association; S4. Based on the stored data, performing at least one correlation analysis, and outputting decision information for guiding production or equipment maintenance.

[0007] Preferably, the set of preset state determination rules in step S2 includes: If the equipment speed is greater than zero and the production count continues to increase, it is determined and marked as a "production" state. If the equipment speed is greater than zero and the production count does not change, it is determined and marked as an "idling" state.

[0008] Preferably, the correlation analysis in step S4 includes efficiency analysis, quality traceability analysis, predictive maintenance analysis, and energy consumption optimization analysis.

[0009] Preferably, the efficiency analysis specifically includes: Based on the cumulative time of each state mark, the comprehensive efficiency and time utilization rate of the equipment are calculated.

[0010] Preferably, the quality traceability analysis specifically includes: Establish a correlation model between the bad sheet rate data in a specific time period and the process parameters and equipment operation parameters in the same period to locate the key factors causing quality defects.

[0011] Preferably, the predictive maintenance analysis specifically includes: Based on the temperature and current time series data of the key components of the equipment, a health assessment model is constructed, and maintenance warning information is generated when the data characteristics deviate from the preset threshold.

[0012] Preferably, the energy consumption optimization analysis specifically includes: Correlate energy consumption data with equipment state data to quantitatively calculate energy loss in idle and standby non-production states.

[0013] Preferably, in step S4, when multiple correlation analyses are performed simultaneously, the analysis results of the multiple correlation analyses are fused to form comprehensive decision information.

[0014] Preferably, forming comprehensive decision information at least includes: Fusing the analysis results of the predictive maintenance analysis with the efficiency analysis and the quality traceability analysis to form comprehensive decision information. Simultaneously performing efficiency analysis, quality traceability analysis, and energy consumption optimization analysis, and fusing the analysis results to form comprehensive decision information.

[0015] In a second aspect, the present application also provides an analysis system for a cigarette label inspection equipment, which is used to implement the analysis method of the cigarette label inspection equipment according to claim 1, and the system includes: A data acquisition module is used to implement step S1. a state recognition module for implementing step S2 and having a configurable rule engine built-in; a data storage module for implementing step S3 and containing a storage space for storing data classified by data category / subject; a data analysis module for implementing step S4 and containing a calculation unit for performing different correlation analysis; a visual display module for graphically displaying decision information through a comprehensive dashboard.

[0016] The analysis method and system of the cigarette label inspection equipment have the following beneficial effects: 1. Fine and automatic perception of equipment state is realized. The rule engine is used to convert low-dimensional data into high-dimensional state information, replacing subjective judgment relying on human experience, and greatly improving the accuracy and efficiency of state monitoring.

[0017] 2. Information silos are broken down, and unified data assets are built. Through multi-source data collection and subject-oriented correlation storage, isolated data is integrated into structured data assets that can be used for in-depth analysis, laying a solid foundation for upper-layer intelligent applications.

[0018] 3. Data is converted into executable decision-making wisdom. Through various correlation analysis models, the internal relationships between data are actively mined, realizing the transition from "post-event statistics" to "in-event warning" and "pre-event prediction", and directly driving the improvement of production efficiency, product quality and equipment reliability.

[0019] 4. Fusion intelligent decision-making across analysis dimensions is realized. The application first systematically integrates analysis results of different dimensions such as quality, efficiency, energy consumption and equipment health, realizes dynamic optimization of globally optimal process parameters and production intelligent scheduling based on equipment state, solves the decision conflict problem caused by single-point analysis, and realizes system-level production optimization.

[0020] 5. The technical solution has high scalability and universality. Although the application is designed for cigarette label inspection equipment, the core architecture based on software and algorithms can be easily migrated to other similar industrial manufacturing and detection equipment, and has wide application prospects. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0022] Figure 1A flowchart of an analysis method of a cigarette label inspection equipment provided in an embodiment of the present application is shown in the figure. Figure 2 A structural block diagram of an analysis system of a cigarette label inspection equipment provided in an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described in combination with the accompanying drawings in the embodiments of the present application.

[0024] In the following description, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance. The following description provides a plurality of embodiments of the present application, and different embodiments can be replaced or combined, so that the present application can also be considered to include all possible combinations of the same and / or different embodiments described. Therefore, if one embodiment includes features A, B and C, and another embodiment includes features B and D, the present application should also be considered to include one or more embodiments of all other possible combinations of A, B, C and D, although the embodiment may not be explicitly described in the following content.

[0025] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes can be made to the functions and arrangements of the described elements without departing from the scope of the present application. Various examples can appropriately omit, replace or add various processes or components. For example, the described methods can be executed in different order from the described order, and various steps can be added, omitted or combined. In addition, features described with respect to some examples can be combined into other examples.

[0026] Referring to Figure 1 , Figure 1 A flowchart of an analysis method of a cigarette label inspection equipment provided in an embodiment of the present application is shown in the figure. In the embodiment of the present application, the method comprises: S1, acquiring multi-source heterogeneous data in a running process of a cigarette label inspection equipment.

[0027] Among them, the multi-source heterogeneous data includes at least two of the device production data, the device state signal, the device process data, the device monitoring data and the device energy consumption data.

[0028] The execution subject of the present application can be a data acquisition system, which can be deployed on a standard industrial server. The specific application object can be, for example, a MK420Qmini cigarette label inspection machine.

[0029] In this application, data acquisition can be achieved through the equipment's OPC UA server interface, which can read production data and status signals such as "total output quantity," "good wafer quantity," "bad wafer quantity," and "real-time machine speed" in real time. Through a general data acquisition gateway deployed on the equipment network, process, environmental, and energy consumption data from standard sensors (such as temperature sensors, humidity sensors, smart meters, and electronic barometers) can be received. All data is timestamped with millisecond-level precision.

[0030] S2. Based on the preset state determination rule set, process the multi-source heterogeneous data obtained in step S1, and automatically identify and mark the real-time operating status of the device; The operating status includes at least production, idling, standby, shutdown, and alarm.

[0031] In this embodiment, a "status recognition module" is deployed in the data aggregation server (such as the MDIS platform). This module loads a predefined set of status determination rules (e.g., described using "IF... THEN..." logic). The rule engine performs logical judgments on the real-time incoming data stream and dynamically labels the device's operating status at each point in time with status tags such as "production" or "idling". This rule set can be configured and expanded through a management interface.

[0032] In one possible implementation, the preset set of state determination rules in step S2 includes: If the equipment speed is greater than zero and the output count continues to increase, it is determined and marked as "production" status; If the equipment speed is greater than zero and the output count remains unchanged, it is determined and marked as "idling".

[0033] Specifically, rule R1 is used to determine production, and can be expressed as follows: IF (Equipment power == ON) AND (Real-time vehicle speed > 0) AND (Production counter continues to increment) THEN Status = "Production"; Rule R2 is used to determine idling, and can be expressed as follows: IF (Equipment power == ON) AND (Real-time speed > 0) AND (Production counter remains unchanged and exceeds threshold T1) THEN state = "idling"; Rule R3 is used to determine standby mode, and can be expressed as follows: IF (Device power == ON) AND (Real-time vehicle speed == 0) AND (No alarm signal) THEN Status = "Standby".

[0034] For example, if the system detects that the vehicle speed is 120 meters per minute, but the production counter does not change within 1 minute, it will automatically mark that time period as "idling".

[0035] S3. Associate and store data with status tags with multi-source heterogeneous data.

[0036] In the embodiments of this application, the tagged data stream is cleaned, transformed, and classified according to a preset thematic data model. For example, a "device status thematic table" (timestamp, status), a "production thematic table" (timestamp, output), and a "process thematic table" (timestamp, air pressure, template number) are established. These tables are associated with timestamps and equipment IDs and stored in a time-series database or data warehouse, transforming the raw data into a well-structured "data asset" that is easy to query and analyze efficiently.

[0037] For example, the tagged data is written to the data storage module. The system establishes the following topic data tables: Status table: timestamp, device ID, status label.

[0038] Production table: timestamp, device ID, total output, number of good films, number of bad films.

[0039] Process sheet: timestamp, equipment ID, vehicle speed, air pressure, ambient temperature, ambient humidity.

[0040] Energy consumption meter: timestamp, device ID, instantaneous power.

[0041] Therefore, all data can be queried across tables by using "timestamp" and "device ID" as the association keys.

[0042] When storing data, partitions can be created based on the subject or category of the data, and space can be allocated reasonably for different subjects, so that each partition of the storage space has a considerable margin of use.

[0043] S4. Based on the data stored in association, perform at least one association analysis and output decision information to guide production or equipment maintenance.

[0044] In the embodiments of this application, various computing units in the "data analysis module" are run in the data analysis server.

[0045] Feasible correlation analysis can include several conventional single analysis forms such as efficiency analysis, quality traceability analysis, predictive maintenance analysis, and energy consumption optimization analysis, and can also include comprehensive decision-making through melting analysis.

[0046] The efficiency analysis specifically includes: calculating the overall efficiency and time utilization of the equipment based on the cumulative time of each state marker.

[0047] For example, this can be used to evaluate overall equipment efficiency. The system reads equipment status marker data over a week and aggregates the time spent in each status. It finds that "production" time accounts for 45%, "idle" time 15%, "standby" time 25%, and "downtime" time 15%. The decision information is: the system calculates and displays an OEE of 45%, clearly indicating that "idle" and "standby" are the main sources of wasted time. Based on this, the production manager optimizes production planning and material distribution to reduce non-production time.

[0048] Quality traceability analysis specifically includes: establishing a correlation model between defective film rate data within a specific time period and process parameters and equipment operating parameters during the same period, in order to locate the key factors leading to quality defects.

[0049] For example, to pinpoint the root cause of a spike in the defective film rate, the system detected an abnormal defective film rate between 2:00 PM and 3:00 PM yesterday. Data from that period and preceding and following times was automatically extracted and correlation calculations were performed. The decision information was as follows: the system output report showed that the defective film rate had the strongest negative correlation with "compressed air pressure" (-0.92). The dashboard indicated that "insufficient exhaust air pressure is suspected to be the main cause." Maintenance personnel checked and repaired the air leaks accordingly, and the defective film rate returned to normal.

[0050] Predictive maintenance analysis specifically includes: building a health assessment model based on temperature and current time-series data of key equipment components, and generating maintenance warning information when data characteristics deviate from preset thresholds.

[0051] For example, in the case of main motor bearing wear early warning, if the system monitors the main motor temperature and finds that its baseline value shows a monotonous upward trend from 70℃, reaching 78℃ within a week and exceeding the limit multiple times momentarily, the decision information is: the system generates a "Level 1 Warning," indicating "the main motor temperature is abnormally high, and it is recommended to check the bearings and cooling system." The maintenance team completes the maintenance during planned downtime, avoiding unplanned downtime.

[0052] Energy consumption optimization analysis specifically includes: linking energy consumption data with equipment status data, and quantitatively calculating energy losses during idling, standby, and non-production states.

[0053] For example, to reduce non-value-added energy consumption, the system analyzes energy consumption data and finds that the power consumption of equipment in "idle" mode is as high as 18kW, accounting for more than 20% of the total energy consumption. The decision information is: the system quantifies the waste of "idle energy consumption" and suggests the formulation of operating procedures requiring operators to switch to a low-power "standby" mode when waiting for more than 5 minutes. After implementation, the total weekly energy consumption decreased by 10%.

[0054] This approach connects isolated data points into meaningful "information chains" and "decision maps." For example, quality traceability models link "results" (defective parts) with "processes" (process parameters), enabling the root cause of quality problems to be located; predictive maintenance models link "current status" (temperature trends) with "future risks" (failures), enabling proactive intervention in maintenance work.

[0055] Taking quality traceability analysis in a single analysis format as an example, its detailed process is as follows: Step S41: The user or system monitors that the defective film rate is abnormally high within the time period T_a.

[0056] Step S42: The system automatically extracts historical data from T_a and the previous normal time period T_n, including the defective film rate, vehicle speed, air pressure, and ambient humidity.

[0057] Step S43: Call a correlation analysis algorithm (such as Pearson correlation coefficient calculation) to analyze the correlation strength between the defective film rate and various process and operating parameters.

[0058] Step S44: The system sorts and outputs the factors with the highest correlation. For example, the output results are: the negative correlation with "compressed air pressure" is -0.92 (strong correlation), and the positive correlation with "ambient humidity" is 0.75.

[0059] Step S45: The result is highlighted on the dashboard in the form of a report, indicating that "insufficient compressed air pressure" is the primary suspected cause of the increased defect rate, and instructing maintenance personnel to immediately check the air circuit system.

[0060] For example, when an operator discovers an abnormal defect rate in a batch of products, they select the abnormal time period on the dashboard. The system automatically extracts data on "defect rate," "vehicle speed," "air pressure," and "ambient humidity" for that time period and the normal time periods before and after it. The system then uses Python's pandas library to calculate the Pearson correlation coefficient between the defect rate and the other parameters.

[0061] In one possible instance, analysis showed that the defective film rate had a correlation coefficient of -0.89 (strong negative correlation) with "compressed air pressure" and a correlation coefficient of 0.45 with "ambient humidity." The system highlighted "Insufficient exhaust air pressure is the primary suspect" on the dashboard. After inspection, maintenance personnel confirmed a faulty air pressure valve; after replacement, the defective film rate returned to normal.

[0062] Regarding efficiency analysis, the implementation steps can be found below: The system reads the status data for a specified month and aggregates the cumulative hours of "production," "idle," "standby," and "downtime." Then, it calculates the overall equipment efficiency (OEE) as follows: OEE = (production time / total planned time) × (theoretical cycle output / production time) × (good wafers / total output). In this embodiment, the theoretical cycle output is a fixed value.

[0063] In one possible instance, the system calculated the OEE for that month to be 65%. The Kanban board displayed the time distribution as a pie chart, with "idle time" accounting for 18%. After the production manager discovered this, an investigation revealed that it was due to frequent paper roll docking. The roll loading process was then optimized, and the "idle time" time decreased to 10% the following month, with the OEE improving to 70%.

[0064] The implementation steps for predictive maintenance analysis can be found below: The system continuously monitors the "main motor temperature". A control chart is established using historical data from the previous three months as a baseline, with an upper limit set at 85℃. Simultaneously, a simple linear regression is used to fit the temperature's trend over time.

[0065] In one possible instance, the system monitored the main motor temperature slowly rising from 70°C to 78°C over a week, and the trend model predicted it would reach the warning threshold of 85°C in five days. The system automatically generated a predictive maintenance work order, indicating "Main motor overheating risk; inspection of bearings and cooling system recommended." The maintenance team conducted an inspection during a planned downtime, discovered and replaced the worn bearing, preventing unplanned downtime and more serious equipment damage.

[0066] The implementation steps for energy consumption optimization analysis can be found below: The system associates the instantaneous power data in the "Energy Consumption Meter" with the status labels in the "Status Table" by timestamp. It calculates the average power and total power consumption under the three states of "production", "idle", and "standby".

[0067] In one possible example, the analysis report showed that the equipment's average power consumption was 28kW when "in production" and as high as 20kW when "idling." Although "idling" time accounted for only 15%, it accounted for 21% of the total energy consumption. The system calculated that "idling energy consumption" was the main source of waste. Based on this, management implemented a policy of "shutting down the machine if non-production exceeds 10 minutes," resulting in an 8% reduction in total monthly electricity consumption.

[0068] When multiple correlation analyses are performed simultaneously, the results of the multiple correlation analyses are merged to form comprehensive decision-making information.

[0069] Preferably, the formation of comprehensive decision-making information includes at least the following two fusion forms: The results of predictive maintenance analysis are integrated with those of efficiency analysis and quality traceability analysis to form comprehensive decision-making information; Simultaneously, efficiency analysis, quality traceability analysis, and energy consumption optimization analysis are performed, and the analysis results are integrated to form comprehensive decision-making information.

[0070] In one specific embodiment, efficiency analysis, quality traceability analysis, and energy consumption optimization analysis are performed simultaneously, and the analysis results are integrated to form comprehensive decision-making information, specifically including: The safety range of process parameters for ensuring product quality is determined based on the aforementioned quality traceability analysis; Within the safe range of the process parameters, a multi-objective optimization algorithm is used to find the optimal setting value of the process parameters that can simultaneously improve the overall equipment efficiency (OEE) and reduce the energy consumption per unit output, and this setting value is output as the decision information.

[0071] For example, dynamic process parameter optimization seeks the optimal balance between efficiency and energy consumption while maintaining stable quality.

[0072] The system simultaneously calls the quality analysis results (determining that the safe speed range for stable defect rate of the current product is 115-130m / min), the efficiency analysis results (calculating the current OEE), and the energy consumption analysis results (calculating the energy consumption per unit output). Within the safe range of 115-130 m / min, the system performed simulation calculations using a multi-objective optimization algorithm. It was found that when the vehicle speed was set to 125 m / min, the system could achieve the highest OEE and the lowest energy consumption per unit output while ensuring that the defective film rate did not exceed the standard. The decision information is as follows: The system automatically sends 125m / min as the "recommended speed" to the operator or directly to the equipment controller to achieve dynamic optimization of process parameters.

[0073] In one specific embodiment, predictive maintenance analytics is integrated with efficiency analytics and quality traceability analytics to form comprehensive decision-making information, specifically including: Based on the health scores of key equipment components obtained from the predictive maintenance analysis, it is determined whether the health scores are lower than the threshold set for a specific production task; If so, then by combining the historical data from the aforementioned quality traceability analysis and efficiency analysis, the risk of defective products and the loss of production efficiency when producing the specific product under the current equipment health condition can be predicted. Based on the prediction results, production scheduling suggestions are generated. These suggestions include scheduling high-precision, high-value orders to equipment with higher health or arranging maintenance windows for equipment with low health. These scheduling suggestions are then output as decision information.

[0074] For example, predictive production scheduling based on equipment health can select the most suitable production equipment for high-value orders and mitigate quality risks.

[0075] The system receives the output of predictive maintenance analysis (main roller health score of No. 1 is 65%) and calls the historical model of quality analysis and efficiency analysis (the model shows that when the health of No. 1 is below 80%, the risk of defective sheets in the production of high-end products increases by 3 times and the speed needs to be reduced by 5%). At this time, the planning system issues a high-value "Chunghwa" cigarette label order. The system determines that the health status of machine 1 does not meet the production requirements of the order and automatically generates a production scheduling suggestion: "It is recommended to adjust the production of the 'Chunghwa' cigarette label order from machine 1 to machine 2, which has a health status of 95%, and generate a preventive maintenance work order for machine 1." The decision information is as follows: The production scheduler adopted the suggestion, effectively avoiding a potential large-scale quality accident, and reasonably arranged equipment maintenance.

[0076] Of course, in practice, individual analysis forms can be added according to actual needs, and integrated decision analysis between individual analysis results can be added according to correlation to solve specific production problems, such as the balance between quality and efficiency, and the balance between energy consumption and efficiency.

[0077] In one possible embodiment, the availability and continuity of energy consumption data are utilized as "sentinels" for early equipment failures, enabling coordinated early warning of energy efficiency and equipment status. Specifically: Real-time comparison of the output of energy consumption analysis (such as instantaneous power and energy consumption per unit output) with the output of efficiency analysis (such as equipment status and output).

[0078] Establish a baseline model of "state-energy consumption" under normal conditions. For example, the power baseline for "production state" is 25kW±2kW; the power baseline for "idle state" is 8kW±1kW.

[0079] Decision information: When the system detects that the equipment is in "production" state and the output is normal, but the real-time power is continuously higher than the baseline limit, it triggers an "energy efficiency anomaly" warning.

[0080] This early warning system will work in conjunction with the predictive maintenance system to automatically create a check task for relevant components (such as the main motor and transmission system), indicating that "the equipment may have increased mechanical resistance or decreased electrical efficiency," thereby achieving earlier and lower-cost fault warnings.

[0081] In a specific instance, the system detected that during normal production, the power of inspection machine No. 3 slowly increased from 25kW to 28kW, but the machine speed and output remained unchanged. The system triggered an energy efficiency warning. After receiving the work order, maintenance personnel inspected the machine and found that the paper feed pulley bearing was experiencing early jamming. Timely intervention prevented an unplanned downtime due to bearing seizure.

[0082] By integrating analytical dimensions such as quality, efficiency, energy consumption, and predictive maintenance, industrial artificial intelligence can move from "descriptive analysis" to "decision optimization," shifting the mindset from triggering global optimization rather than local optimization, and from reactive remediation to proactive intervention, transforming data insights into automated, closed-loop decision-making.

[0083] Based on this, decision-making information or other information can be displayed, making the decision-making information visual and interactive.

[0084] Specifically, a comprehensive data dashboard (visualization module) is developed via a web interface. This dashboard retrieves analysis results from the data storage module and displays them in the form of OEE indicator cards, status time distribution pie charts, quality defect Pareto charts, key parameter trend curves, and warning information lists. Users can click on any chart element (such as a period with a high defect rate) to drill down and view all relevant raw data and process parameters for that period, achieving a transparent and traceable analysis loop.

[0085] According to the second aspect of this application, please refer to Figure 2 Furthermore, an analysis system for a cigarette label inspection device is provided to implement the method described in the first aspect. The system includes: The data acquisition module is used to implement step S1; it can be written in Java or Go language, and serves as a resident service. It is responsible for communicating with devices and sensors with different protocols to achieve stable data acquisition and initial caching.

[0086] The status recognition module implements step S2 and has a built-in configurable rule engine; it can be deployed as a microservice with an embedded rule engine. It subscribes to the message queue (such as Kafka) of the data acquisition module, performs real-time calculations, and outputs results with status tags.

[0087] The data storage module is used to implement step S3 and includes storage space that can be classified and stored according to data category / topic; it adopts a hybrid architecture that combines time-series database and relational database to store real-time monitoring data and dimension table information respectively.

[0088] The data analysis module is used to implement step S4 and includes computing units for performing different correlation analyses; it can be built based on the Spark or Flink framework and is used to process correlation analysis of batch historical data; it also includes a lightweight real-time computing unit for quickly calculating metrics such as OEE.

[0089] The visualization module is used to graphically display decision-making information through a comprehensive dashboard. It is developed using front-end frameworks such as Vue or React, renders data dashboards using graphics libraries such as ECharts, and provides rich interactive functions, such as time range selection, data drill-down, and alert confirmation.

[0090] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0091] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0092] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0093] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. An analytical method for a cigarette label inspection device, characterized in that, The method includes: S1. Acquire multi-source heterogeneous data during the operation of the cigarette label inspection equipment, wherein the multi-source heterogeneous data includes at least two of the following: equipment production data, equipment status signals, equipment process data, equipment monitoring data, and equipment energy consumption data. S2. Based on a preset set of status determination rules, process the multi-source heterogeneous data obtained in step S1 to automatically identify and mark the real-time operating status of the equipment; the operating status includes at least production, idling, standby, shutdown, and alarm. S3. Associate and store the data with status tags with the multi-source heterogeneous data; S4. Based on the data stored in association, perform at least one association analysis and output decision information to guide production or equipment maintenance.

2. The method according to claim 1, characterized in that, The preset state determination rule set mentioned in step S2 includes: If the equipment speed is greater than zero and the output count continues to increase, it is determined and marked as "production" status; If the equipment speed is greater than zero and the output count remains unchanged, it is determined and marked as "idling".

3. The method according to claim 1, characterized in that, The correlation analysis mentioned in step S4 includes efficiency analysis, quality traceability analysis, predictive maintenance analysis, and energy consumption optimization analysis.

4. The method according to claim 3, characterized in that, The efficiency analysis specifically includes: Based on the cumulative time of each state flag, calculate the overall efficiency and time utilization of the device.

5. The method according to claim 3, characterized in that, The quality traceability analysis specifically includes: Establish a correlation model between defective film rate data within a specific time period and process parameters and equipment operating parameters during the same period to identify key factors leading to quality defects.

6. The method according to claim 3, characterized in that, The predictive maintenance analysis specifically includes: Based on the temperature and current time-series data of key equipment components, a health assessment model is constructed, and maintenance warning information is generated when the data characteristics deviate from the preset threshold.

7. The method according to claim 3, characterized in that, The energy consumption optimization analysis specifically includes: By linking energy consumption data with equipment status data, energy loss during idling, standby, and non-production states can be quantitatively calculated.

8. The method according to claim 3, characterized in that, Step S4 also includes fusing the results of multiple correlation analyses when multiple correlation analyses are performed simultaneously to form comprehensive decision information.

9. The method according to claim 8, characterized in that, Information for forming a comprehensive decision includes at least: The results of predictive maintenance analysis are integrated with those of efficiency analysis and quality traceability analysis to form comprehensive decision-making information; Simultaneously, efficiency analysis, quality traceability analysis, and energy consumption optimization analysis are performed, and the analysis results are integrated to form comprehensive decision-making information.

10. An analysis system for a cigarette label inspection device, characterized in that, The system is used to implement the analytical method of the cigarette label inspection equipment as described in claim 1, the system comprising: The data acquisition module is used to implement step S1; The status recognition module is used to implement step S2 and has a built-in configurable rule engine; The data storage module is used to implement step S3 and includes storage space that can be classified and stored according to data category / topic; The data analysis module is used to implement step S4 and includes computing units for performing different correlation analyses; The visualization module is used to graphically display decision-making information through a comprehensive dashboard.

Images