Packaging material production quality intelligent monitoring method based on industrial internet of things

Abstract

The invention discloses a packaging material production quality intelligent monitoring method based on the industrial Internet of Things, and particularly relates to the field of the industrial Internet of Things, and the method comprises the steps: S1, equipment deployment, S2, monitoring division, S3, raw material qualification analysis, S4, packaging production feature monitoring, S5, packaging production feature analysis, and S6, adaptive adjustment. The industrial internet is used for connecting all using devices, the sensor terminal is used for monitoring the density and purity of production raw materials, then the visual monitoring terminal is used for conducting picture extraction on the surface of a package, package production characteristics are extracted, package stability and integrity analysis is conducted on the basis of the package production characteristics, and the packaging quality is improved. In this way, correlation analysis is conducted in combination with the temperature of production equipment, self-adaptive adjustment and production quality summarization are conducted, and therefore the safety of the production process is guaranteed to a certain degree, a clear direction is provided for follow-up adjustment, material waste caused by low quality of packaging materials is reduced, and the maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of industrial Internet of Things (IoT) technology, and more specifically, to a method for intelligent monitoring of the production quality of packaging materials based on industrial IoT. Background Technology

[0002] With the rapid development of Industrial Internet of Things (IIoT) technology, the packaging material production sector is undergoing a critical transformation from traditional manufacturing to intelligent and networked manufacturing. Among these transformations, the quality monitoring of packaging materials is a core element in ensuring product safety and performance. By monitoring production quality, the sustainable development of materials can be improved, and the IIoT can be used to reconstruct the traditional production model through equipment interconnection, real-time data collection and analysis.

[0003] Traditional methods for monitoring the quality of packaging material production typically rely on offline sampling for analysis. This involves collecting samples from the production process or monitoring points and then transferring them to an independent laboratory for analysis. This intermittent operational mode of production, sampling, offline analysis, and transfer provides data support for quality control.

[0004] However, it still has some drawbacks in actual use. First, the current method requires sampling and testing during data collection. As a result, the testing time for packaging materials is relatively long. During this period, production continues. If quality problems are found during testing, a large number of unqualified products have already been produced, resulting in waste of raw materials, increased production costs, and even the impact on the subsequent supply chain due to batch issues. In addition, sampling and testing are difficult to cover the quality status of the entire production batch. There may be cases where the sampling is qualified but the actual batch is unqualified, making it difficult for offline sampling to fully capture the quality testing of packaging materials. Secondly, in traditional packaging material quality monitoring, offline sampling analysis can easily lead to a delay in the discovery of production quality problems. In this case, quality traceability and adjustment are required. However, existing quality traceability and adjustment tend to shut down the machine for investigation in order to ensure effectiveness. Frequent shutdowns for investigation will extend the entire production cycle. More importantly, the production process may be interrupted during the shutdown investigation. Especially in continuous production, parameter fluctuations are more likely to occur when the production line is restarted, causing greater quality risks. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide an intelligent monitoring method for the production quality of packaging materials based on the Industrial Internet of Things. By dividing the packaging material production process into monitoring points, each monitoring area is obtained, and raw material qualification analysis and production characteristic analysis are performed in each monitoring area. This enables adaptive adjustment of the temperature of the production equipment, thereby improving the production quality of packaging materials and effectively solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: S1: Equipment Deployment: Set up production equipment, sensor terminals, vision monitoring terminals, transmission equipment and control terminals in the production area of ​​the packaging material production line, and connect them through the transmission equipment; S2: Monitoring Division: Based on the production plan, the production line is divided into monitoring links. Thus, each monitoring link is divided into monitoring points before the production of packaging materials, and based on the monitoring points, it is divided into the first monitoring area and the second monitoring area. S3: Raw material qualification analysis: The density and purity of the first monitoring area are monitored using a sensor terminal, and the qualification of the raw materials is analyzed. The qualified raw materials are then transmitted to the second monitoring area. S4: Packaging production feature monitoring: Based on the second monitoring area, each monitoring segment is extruded to form packaging, and the current packaging surface is scanned by a visual monitoring terminal to form a scanned image, thereby extracting the initial packaging production features of each monitoring segment. At the same time, the temperature of the production equipment is monitored by a sensor terminal. S5: Packaging Production Feature Analysis: Based on the initial packaging production feature analysis, the packaging stability and integrity of each monitoring segment are analyzed. The monitoring object for packaging stability is the packaging surface defect feature, and the monitoring object for packaging integrity is the packaging precision feature. S6: Adaptive Adjustment: Based on the analysis results of packaging stability and integrity, a correlation analysis is performed with the temperature of the production equipment. Then, based on the correlation analysis results of packaging production quality, the temperature of the production equipment is adaptively adjusted, and a quality report is generated according to the preset summary method.

[0007] The technical effects and advantages of this invention are as follows: 1. This invention connects various devices using the Industrial Internet, and then monitors the density and purity of production raw materials through sensor terminals. Based on the qualification level of the raw materials, the quality of the raw materials is perceived and adjusted accordingly. On the one hand, it can meet the adjustment needs to the greatest extent, and on the other hand, it can monitor the quality of raw materials before packaging production, which is beneficial to improve production quality and efficiency during normal packaging material production. 2. This invention extracts images of the packaging surface using a visual monitoring terminal, identifies packaging production features, and analyzes packaging stability and integrity based on these features. This is then combined with correlation analysis of production equipment temperature to perform adaptive adjustments and summarize production quality. This approach ensures production process safety to a certain extent. On one hand, it matches adjustment solutions to specific problems, avoiding new issues arising from single adjustments and better controlling production quality. On the other hand, the summary method accurately identifies the root causes of quality risks, providing a clear direction for subsequent adjustments, reducing material waste caused by poor packaging material quality, and lowering maintenance costs. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating the overall structure of the present invention.

[0009] Figure 2 This is a flowchart illustrating the division of monitoring points in this invention.

[0010] Figure 3 This is a schematic diagram of the device connection for the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0012] like Figures 1-3 The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things (IIoT) shown in this invention includes the following steps in its specific implementation: S1: Set up production equipment, sensor terminals, visual monitoring terminals, transmission equipment and control terminals in the production area of ​​the packaging material production line, and connect them through the transmission equipment.

[0013] In a more specific application of the present invention, the production equipment is used to process raw materials into packaging materials during the production of packaging materials. Specifically, it can be a raw material screening machine or an extruder. The raw material screening machine is used to screen qualified raw materials, and the extruder is used to heat and melt plastic particles and then extrude them through a die to form initial packaging such as films and sheets.

[0014] Sensor terminals are used to sense the state of packaging materials during the production process. Specifically, they can be density sensors, purity sensors, and temperature sensors. The density sensor can be a buoyancy-type density meter, deployed above the conveyor belt at the outlet of the raw material screening machine, to collect the particle density of each batch of raw materials in real time. The purity sensor can be a near-infrared spectral sensor, installed near the density sensor, to identify impurities in the raw materials through spectral analysis. The temperature sensor can be installed on the surface of the production equipment to monitor the temperature of the production equipment during the production process.

[0015] Visual monitoring terminals are used to visualize and monitor surface defects and appearance of packaging through image acquisition and analysis, overcoming the limitations of traditional sensors in shape detection. Specifically, they can be a type of camera such as a high-definition line scan camera, an industrial camera, or a 3D vision sensor. When determining the camera selection, it is necessary to select one based on the packaging material production process and the actual suitability of the camera. High-definition line scan cameras can identify minute defects such as scratches, crystal points, and bubbles during packaging extrusion molding. Industrial cameras can detect problems such as misregistration, missing prints, and color deviations through image comparison. 3D vision sensors can identify three-dimensional defects such as packaging deformation and packaging dents.

[0016] The transmission equipment is used to build data transmission channels, transmitting data collected by each terminal to the monitoring platform in real time and stably. Specifically, it includes 5G base stations and modules, industrial Ethernet switches, and wireless gateways. The 5G base stations and modules are used to transmit high-definition images, using slicing technology to achieve low latency and high bandwidth transmission. The industrial Ethernet switches adopt a redundant ring network structure to transmit data such as equipment status and environment. The wireless gateways are used to wirelessly connect various sensors to realize data transmission. In turn, by utilizing the Industrial Internet through 5G, industrial Ethernet, and wireless networks, intelligent monitoring from raw material preparation to packaging production is realized, making production more precise, efficient, and controllable, improving industrial production efficiency, optimizing resource allocation, and promoting industrial upgrading.

[0017] The control terminal analyzes and controls relevant parameters based on the monitoring data of the above-mentioned equipment during the packaging production process.

[0018] S2: Monitoring Division: Based on the production plan, the production line is divided into monitoring sections. Thus, each monitoring section is divided into monitoring points before the production of packaging materials, and based on the monitoring points, it is divided into the first monitoring area and the second monitoring area.

[0019] It should be explained that production planning needs to be formulated in combination with product type, capacity requirements and process standards. The monitoring links divided according to this plan correspond to the core production processes, specifically including the raw material pretreatment area and the extrusion molding area.

[0020] In this embodiment, it should be specifically noted that the interval monitoring process is divided into monitoring points before the packaging materials are produced, as detailed below: A1: Obtain the starting and ending stations and section lengths corresponding to each monitoring link in the packaging material production line, and then evenly distribute each monitoring point within the corresponding starting and ending stations of each monitoring link.

[0021] It should be added that the monitoring links of the packaging material production line are divided into raw material pretreatment section and extrusion molding section according to the production plan. The start and end stations of each monitoring link are defined by each production equipment. For example, the start station of the raw material pretreatment section is the raw material warehouse outlet and the end station is the extruder inlet. The start station of the extrusion molding section is the extruder inlet and the end station is the traction machine outlet.

[0022] A2: Arrange the monitoring points in the order of the monitoring links of each section from the starting station to the ending station, and conduct production monitoring in the order of the monitoring points. Record the monitoring parameters of each monitoring point. The monitoring parameters are the core parameters corresponding to each monitoring link. For example, the core parameter corresponding to the raw material pretreatment section can be the raw material humidity, and the core parameter corresponding to the extrusion molding section can be the production equipment temperature. The raw material humidity and the production equipment temperature can be collected by setting humidity sensors and temperature sensors in the production equipment and connecting them to the sensor terminal.

[0023] A3: Calculate the similarity of the monitoring data of each pair of adjacent monitoring points and compare it with the similarity threshold preset by the system. For example, the similarity threshold is 85%. If the similarity of the monitoring data of adjacent monitoring points meets the similarity threshold, then the two points are assigned to the same monitoring segment, and the latter monitoring point in the adjacent monitoring points is combined with the next monitoring point to form a new pair of adjacent monitoring points to continue the similarity calculation. If the similarity threshold is not met, then the former monitoring point in the adjacent monitoring points is taken as the end point of the current monitoring segment, and the latter monitoring point in the adjacent monitoring points is combined with the next monitoring point to form a new pair of adjacent monitoring points to continue the similarity calculation.

[0024] For example, in the extrusion molding section, the temperature fluctuation difference between the extruders of adjacent monitoring points 1 and 2 is 2℃, and the data similarity is 90%, which meets the threshold. The line between the two points is classified as the same monitoring segment. However, the temperature fluctuation difference between the extruders of monitoring points 2 and 3 is 8℃, and the data similarity is only 70%, which does not meet the threshold. Therefore, monitoring point 2 is taken as the end point of the current monitoring segment, and monitoring point 3 and the monitoring point of the next section form a new pair to continue the calculation. Finally, the division of the monitoring segments of the entire production line is completed, and the precise section monitoring of the production line is realized.

[0025] A4: Repeat the above operation until the similarity calculation of the last monitoring point is performed. Finally, the entire interval monitoring link is divided into monitoring segments and integrated into the first monitoring area and the second monitoring area. The first monitoring area is obtained by dividing the raw material pretreatment interval into monitoring segments, and the second monitoring area is obtained by dividing the extrusion molding interval into monitoring segments.

[0026] S3: Raw material qualification analysis: The density and purity of the first monitoring area are monitored using a sensor terminal, and the qualification of the raw materials is analyzed. The qualified raw materials are then transmitted to the second monitoring area.

[0027] In this embodiment, it should be specifically explained that the raw material qualification analysis includes density and purity analysis, wherein the density analysis is specifically performed as follows: The density sensor of the sensor terminal is used to collect the raw material particle density of each monitoring segment in the first monitoring area according to the preset sampling frequency. Then, the maximum and minimum raw material particle densities are removed and the average value is taken as the density monitoring value of the first monitoring area. The density monitoring value corresponding to the first monitoring area is compared with the standard density range. If the density monitoring value corresponding to the first monitoring area is within the standard density range, it is marked as normal density. If the density monitoring value corresponding to the first monitoring area is outside the standard density range, it is marked as abnormal density, and the density deviation is extracted.

[0028] It should be explained that the preset sampling frequency is set before the production equipment is used. For example, the preset sampling frequency is once per second.

[0029] Before conducting a raw material rationality analysis, the operator needs to confirm the raw material type and match the corresponding standard density range according to the raw material type. For example, if the raw material type is confirmed to be polyethylene, the standard density range of 0.915-0.925 g / cm³ can be obtained through the Industrial Internet.

[0030] The purity analysis process is as follows: The purity sensor of the sensor terminal is used to collect the raw material spectrum of each monitoring segment in the first monitoring area, and compare it with the reference spectrum of the standard pure raw material. The impurity content is calculated by the difference of spectral characteristic peaks. The impurity content corresponding to the first monitoring area is compared with the impurity threshold. For example, the impurity threshold is 0.1%. If the impurity content is less than or equal to the impurity threshold, it is marked as normal purity. If the impurity content is greater than the impurity threshold, it is marked as abnormal purity, and the purity deviation is extracted.

[0031] It should be further noted that the raw material qualification analysis is based on the results of density and purity analysis, as detailed below: When the density and purity are normal, the density fluctuation value of the raw material particle density of each monitoring segment corresponding to the first monitoring area is calculated by the variance formula. At the same time, the peak area difference rate between the raw material spectrum and the reference spectrum of each monitoring segment corresponding to the first monitoring area is calculated and the average value is calculated to obtain the average peak area difference rate. The raw material qualification coefficient is calculated by adding the density fluctuation value and the average peak area difference rate to the exponential terms and taking the average value. Specifically, it is expressed as follows: , Where Hd represents the raw material qualification coefficient of the first monitoring area, and Md and Sd represent the density fluctuation value and average peak area difference rate of the first monitoring area, respectively. The smaller the density fluctuation value and the lower the average peak area difference rate of the first monitoring area, the larger the raw material qualification coefficient and the higher the degree of raw material qualification. Set a raw material qualification threshold. For example, the raw material qualification threshold is 0.7. Compare the raw material qualification coefficient with the raw material qualification threshold, and transmit the raw material to the second monitoring area when the raw material qualification coefficient is greater than or equal to the raw material qualification threshold. When density and purity are abnormal, the raw materials are remixed, the mixer speed is increased, and the purity and density of each monitoring segment in the first monitoring area are re-analyzed until the purity and density are normal. Then the raw material qualification coefficient is calculated and the qualified raw materials are transferred to the second monitoring area.

[0032] It should be added that when the raw material qualification coefficient is less than the raw material qualification threshold, it means that although the purity and density are normal, the particle size of the raw material may be deviated and insufficient for packaging material production. Therefore, it is necessary to notify the operator through the transmission equipment to replace and inspect the raw material.

[0033] It needs to be explained that the purity and density of raw materials directly affect the subsequent packaging material production process. If the purity of the raw materials is not up to standard and there are too many impurities, the hardness of the impurities may be higher than that of the equipment parts in the second monitoring area, which will lead to screw wear, die head scratches, and production equipment shutdown. On the other hand, impurities are easy to carbonize in the high-temperature molten state, resulting in black spot defects in the extruded film. If the density of the raw materials fluctuates too much, it will lead to unstable feed rate of the extruder. Low-density raw material particles are loose and the feed rate per unit volume is small, while high-density raw material particles are compact and the feed rate per unit volume is large, which can easily lead to a sudden increase in the pressure of the extruder barrel.

[0034] S4: Packaging Production Feature Monitoring: Based on the second monitoring area, each monitoring segment is extruded to form packaging, and the current packaging surface is scanned by a visual monitoring terminal to form a scanned image, thereby extracting the initial packaging production features of each monitoring segment. At the same time, the temperature of the production equipment is monitored by a sensor terminal.

[0035] In this embodiment, the process of extracting the initial production characteristics of each monitoring segment in the packaging needs to be specifically explained as follows: The scanned images are divided into several monitoring images based on the deployed monitoring segments. Each monitoring image corresponds to a monitoring segment, and specifically, the center point of each monitoring image is the deployed monitoring segment.

[0036] The packaging surface defect features and packaging precision features are extracted from each monitoring image as the initial packaging production features for each monitoring segment. The packaging surface defect features include the type of surface defect, the number of surface defects, and the area of ​​surface defects. For example, the packaging surface defect features include, but are not limited to, bubbles, scratches, wrinkles, etc.

[0037] It should be explained that packaging precision features reflect the geometric regularity of packaging extrusion molding, while packaging surface defect features reflect the surface integrity of packaging extrusion molding. The reason for extracting features reflecting the geometric regularity and surface integrity of packaging extrusion molding from scanned images as initial packaging production features is that these features both affect the functional adaptability and quality stability of the packaging. By extracting these features, the production quality requirements of the packaging can be comprehensively evaluated, providing accurate guidance for subsequent production quality optimization.

[0038] It should be added that when performing initial packaging production feature extraction on scanned images, preprocessing operations such as image enhancement, noise reduction, and background segmentation are required, which can significantly improve the accuracy and reliability of subsequent feature extraction.

[0039] S5: Packaging Production Feature Analysis: Based on the initial packaging production feature analysis, the packaging stability and integrity of each monitoring segment are analyzed. The monitoring object for packaging stability is the surface defect feature of the packaging, and the monitoring object for packaging integrity is the packaging precision feature.

[0040] In this embodiment, it is necessary to specifically explain the packaging stability and integrity of each monitoring segment based on the initial packaging production characteristics analysis. The specific analysis of the packaging stability of each monitoring segment is as follows: Based on the surface defect features of each monitoring segment, the surface defect area of ​​the packaging is extracted, and the surface defect areas of each monitoring segment are summed to obtain the surface defect degree coefficient of the corresponding packaging in the second monitoring area.

[0041] The packaging surface defect types are extracted based on their characteristics, and influence factors are set for different defect types. For example, the influence factor for bubbles in packaging surface defects is 0.2, the influence factor for scratches in packaging surface defects is 0.8, and the influence factor for wrinkles in packaging surface defects is 0.5.

[0042] It needs to be explained that bubbles are usually caused by insufficient melting of raw materials or untimely venting of the die during packaging extrusion molding. The presence of small bubbles may slightly reduce the barrier properties of the packaging, but the impact on the appearance integrity and basic function of the packaging is usually small, so the impact of bubbles is relatively minor. Scratches, on the other hand, are linear defects caused by friction between the packaging and the rough surface of the equipment during cooling and shaping or traction transport. Shallow scratches may only affect the appearance of the packaging, but deeper scratches can damage the surface structure of the packaging, leading to uneven ink adhesion or poor interlayer bonding during subsequent processing, and even tearing due to stress concentration at the scratches during use. Therefore, scratches have a greater impact. The impact of wrinkles is greater than that of bubbles. Wrinkles are folds formed by uneven cooling temperature distribution or traction tension fluctuations in the packaging section, resulting in inconsistent local shrinkage of the material. Slight wrinkles can affect the flatness of the packaging surface, while more severe wrinkles can cause dimensional deviations during bag making, reducing the yield. However, scratches usually directly damage the structural integrity of the packaging surface and increase the risk of functional failure. It can be seen that the impact of wrinkles is smaller than that of scratches but larger than that of bubbles. By setting the range of 0 to 1 as the range of impact factors, the impact factor of bubbles in packaging surface defects is 0.2, the impact factor of scratches in packaging surface defects is 0.8, and the impact factor of wrinkles in packaging surface defects is 0.5.

[0043] The impact factors of different defect types are multiplied by the number of surface defects based on the characteristics of packaging surface defects to obtain the impact coefficient of a single type of packaging defect. The impact degree of packaging defects corresponding to each defect type is then accumulated to calculate the impact coefficient of packaging defects for each monitoring segment.

[0044] Based on the packaging surface defect severity coefficient and packaging defect impact coefficient, the packaging surface stability coefficient is calculated by comparing the packaging defect impact coefficient of each monitoring segment with the area of ​​the scanned image, as specifically expressed as: , Where W represents the packaging surface stability coefficient corresponding to each monitoring segment, Us represents the packaging defect influence coefficient corresponding to each monitoring segment, St represents the packaging surface defect degree coefficient corresponding to each monitoring segment, and S represents the area of ​​the scanned image. When the ratio of the packaging defect influence coefficient to the scanned image area is smaller, the packaging defect influence coefficient is smaller, and the packaging surface stability coefficient is larger. At this time, the packaging surface stability of the monitoring segment is better.

[0045] It should be further explained that the specific analysis of the packaging integrity of each monitoring segment is as follows: A gray-level co-occurrence matrix is ​​constructed based on the packaging accuracy characteristics of each monitoring segment. The gray-level co-occurrence matrix is ​​used to describe the spatial distribution relationship of pixel pairs with different gray values ​​in the image. It can quantify the surface texture of the packaging and thus analyze the surface accuracy of the packaging. In the specific implementation, the construction process of the gray-level co-occurrence matrix is ​​an existing technology and will not be described in detail here. Based on the constructed gray-level co-occurrence matrix, feature parameters such as energy, contrast, correlation, and entropy are extracted. Energy reflects the uniformity of the image, and a larger value indicates a smoother packaging surface. Contrast reflects the gray-level difference between adjacent pixels, and the clarity of the packaging image is described by the local variation of gray-level values. Correlation is the linear correlation between pixel pairs, and a value close to 1 indicates regularity in texture. Entropy reflects the complexity of image information, and a larger value indicates more types of defects and a more disordered distribution. The feature parameters are then weighted and averaged to calculate the packaging surface integrity coefficient.

[0046] It should be added that these feature parameters are dimensionless. The lower the energy, the more uneven the gray value distribution on the packaging surface, indicating a smoother packaging surface; the lower the contrast, the fewer gray value changes on the packaging surface, and the smoother the surface; the higher the correlation, the more uniform the gray value distribution on the packaging surface, and the smoother the surface; the lower the entropy, the simpler the gray value distribution on the packaging surface, and therefore the smoother the surface. The weight values ​​of the parameters are determined by the direction of their influence on the integrity of the packaging surface, and a weighted average is then used to calculate the packaging surface integrity coefficient.

[0047] S6: Adaptive Adjustment: Based on the analysis results of packaging stability and integrity, a correlation analysis is performed with the temperature of the production equipment. Then, based on the correlation analysis results of packaging production quality, the temperature of the production equipment is adaptively adjusted, and a quality report is generated according to the preset summary method.

[0048] In this embodiment, the specific method for correlation analysis of packaging production quality needs to be explained in detail as follows: A two-dimensional coordinate system is constructed with the production area temperature of each monitoring segment as the horizontal axis and the packaging surface stability coefficient as the vertical axis. Thus, the production area temperature and packaging surface stability coefficient of each monitoring segment are marked in the constructed two-dimensional coordinate system to form a packaging surface stability change curve. A two-dimensional coordinate system is constructed with the production area temperature of each monitoring segment as the horizontal axis and the packaging surface integrity coefficient as the vertical axis. Thus, the production area temperature and packaging surface integrity coefficient of each monitoring segment are marked at each point in the constructed two-dimensional coordinate system to form a packaging surface integrity change curve. The correlation coefficient between the stable change curve of the packaging surface and the intact change curve of the packaging surface is calculated to obtain the correlation between the packaging stability and integrity in the second monitoring area.

[0049] It should be added that the correlation coefficient can specifically be the Pearson correlation coefficient. When calculating the correlation coefficient between the stable change curve of the packaging surface and the intact change curve of the packaging surface, the production equipment temperature of the stable change curve and the intact change curve of the packaging surface is precisely matched to ensure that the two curves have corresponding data at the same temperature point. Thus, the two curves are divided into two arrays according to the corresponding values ​​at the same temperature point. Then, the mean and standard deviation of the two arrays are calculated separately. Finally, the Pearson correlation coefficient is calculated using the mean and standard deviation. The correlation obtained by the calculation is denoted as r. If r > 0, it indicates that the two are positively correlated; if r < 0, it indicates that the two are negatively correlated; the closer |r| is to 1, the higher the correlation.

[0050] It needs further explanation that the adaptive adjustment of the production area temperature based on the packaging production quality correlation analysis results specifically involves comparing the correlation degree between the packaging stability and integrity of the second monitoring area with a preset positive correlation degree. For example, the positive correlation degree is 0.7. Monitoring segments with a correlation degree lower than the positive correlation degree are selected, and the temperature of the production equipment at this time is extracted and compared with the standard temperature to obtain the temperature deviation. If the temperature deviation is negative, the adjustment method is determined to be to increase the temperature; if the temperature deviation is positive, the adjustment method is determined to be to decrease the temperature. Then, the required temperature adjustment is obtained according to the degree of temperature deviation, specifically expressed as follows: , Where Qt represents the required temperature adjustment, ΔT represents the temperature deviation, and sign(ΔT) is a sign function that acts on the temperature change ΔT to determine the sign of ΔT. The value is 1, 0, or -1. If sign(ΔT) = 1, it means that the current production equipment temperature is higher than the required production equipment temperature and the temperature needs to be lowered. If sign(ΔT) = -1, it means that the current production equipment temperature is lower than the required production equipment temperature and the temperature needs to be raised. If sign(ΔT) = 0, then no temperature adjustment is needed. t0 represents the temperature deviation threshold, which can be determined through packaging production experience. For example, t0 = 2.

[0051] The preset summary method is a combination of text and image summaries. The summary content is transmitted to the user terminal through the industrial internet. The summary content includes production characteristics, monitoring data, defect statistics, etc., and is uploaded to the blockchain. If quality problems are found later, the entire production process data of the packaging can be quickly retrieved through the traceability code on the product, thereby locating the cause of the problem.

[0052] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent monitoring of packaging material production quality based on the Industrial Internet of Things, characterized in that, include: S1: Equipment Deployment: Set up production equipment, sensor terminals, vision monitoring terminals, transmission equipment and control terminals in the production area of ​​the packaging material production line, and connect them through the transmission equipment; S2: Monitoring Division: Based on the production plan, the production line is divided into monitoring links. Thus, each monitoring link is divided into monitoring points before the production of packaging materials, and based on the monitoring points, it is divided into the first monitoring area and the second monitoring area. S3: Raw material qualification analysis: The density and purity of the first monitoring area are monitored using a sensor terminal, and the qualification of the raw materials is analyzed. The qualified raw materials are then transmitted to the second monitoring area. S4: Packaging production feature monitoring: Based on the second monitoring area, each monitoring segment is extruded to form packaging, and the current packaging surface is scanned by a visual monitoring terminal to form a scanned image, thereby extracting the initial packaging production features of each monitoring segment. At the same time, the temperature of the production equipment is monitored by a sensor terminal. S5: Packaging Production Feature Analysis: Based on the initial packaging production feature analysis, the packaging stability and integrity of each monitoring segment are analyzed. The monitoring object for packaging stability is the packaging surface defect feature, and the monitoring object for packaging integrity is the packaging precision feature. S6: Adaptive Adjustment: Based on the analysis results of packaging stability and integrity, a correlation analysis is performed with the temperature of the production equipment. Then, based on the correlation analysis results of packaging production quality, the temperature of the production equipment is adaptively adjusted, and a quality report is generated according to the preset summary method.

2. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The aforementioned interval monitoring process is divided into monitoring points before the production of packaging materials, as detailed below: A1: Obtain the starting and ending stations and section lengths corresponding to each monitoring link of the packaging material production line, and then evenly deploy each monitoring point within the corresponding starting and ending stations of each monitoring link. A2: Arrange the monitoring points in the order from the starting station to the ending station of each monitoring link, and conduct production monitoring in the order of the monitoring points. Record the monitoring parameters of each monitoring point, where the monitoring parameters are the core parameters corresponding to each monitoring link. A3: Calculate the similarity of the monitoring data of each pair of adjacent monitoring points and compare it with the system's preset similarity threshold. If the similarity of the monitoring data of adjacent monitoring points meets the similarity threshold, then these two points are assigned to the same monitoring segment, and the latter monitoring point in the adjacent monitoring points is combined with the next monitoring point to form a new pair of adjacent monitoring points for further similarity calculation. If the similarity does not meet the similarity threshold, then the former monitoring point in the adjacent monitoring points is taken as the end point of the current monitoring segment, and the latter monitoring point in the adjacent monitoring points is combined with the next monitoring point to form a new pair of adjacent monitoring points for further similarity calculation. A4: Repeat the above operation until the similarity calculation of the last monitoring point is performed. Finally, the entire interval monitoring link is divided into monitoring segments and integrated into the first monitoring area and the second monitoring area.

3. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The raw material qualification analysis is based on the results of density and purity analysis, as detailed below: When the density and purity are normal, the density fluctuation value of the raw material particle density of each monitoring segment corresponding to the first monitoring area is calculated by the variance formula. At the same time, the peak area difference rate between the raw material spectrum and the reference spectrum of each monitoring segment corresponding to the first monitoring area is calculated and the average value is calculated to obtain the average peak area difference rate. The raw material qualification coefficient is calculated by adding the density fluctuation value and the average peak area difference rate to the exponential terms and taking the average value. Specifically, it is expressed as follows: Hd represents the raw material qualification coefficient of the first monitoring area, and Md and Sd represent the density fluctuation value and average peak area difference rate of the first monitoring area, respectively. Set a raw material qualification threshold, compare the raw material qualification coefficient with the raw material qualification threshold, and transmit the raw material to the second monitoring area when the raw material qualification coefficient is greater than or equal to the raw material qualification threshold.

4. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The initial production features of each monitoring segment in the packaging are extracted as follows: The scanned images are divided into several monitoring images according to the deployed monitoring segments. Each monitoring image corresponds to a monitoring segment, and the center point of each monitoring image is the deployed monitoring segment. Packaging surface defect features and packaging precision features are extracted from each monitoring image as the initial packaging production features for each monitoring segment. The packaging surface defect features include surface defect type, surface defect quantity, and surface defect area.

5. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The packaging stability of each monitoring segment is analyzed in detail as follows: Based on the surface defect features of each monitoring segment, the surface defect area of ​​the packaging is extracted, and the surface defect areas of each monitoring segment are summed to obtain the surface defect degree coefficient of the corresponding packaging in the second monitoring area. Extract the surface defect characteristics of the packaging to identify the types of surface defects, and set the influencing factors for different defect types; The impact factors of different defect types are multiplied by the number of surface defects based on the characteristics of packaging surface defects to obtain the impact coefficient of a single type of packaging defect. The impact degree of packaging defects corresponding to each defect type is then accumulated to calculate the impact coefficient of packaging defects for each monitoring segment. Based on the packaging surface defect severity coefficient and packaging defect impact coefficient, the packaging surface stability coefficient is calculated by comparing the packaging defect impact coefficient of each monitoring segment with the area of ​​the scanned image, as specifically expressed as: Where W represents the packaging surface stability coefficient corresponding to each monitoring segment, Us represents the packaging defect influence coefficient corresponding to each monitoring segment, St represents the packaging surface defect degree coefficient corresponding to each monitoring segment, and S represents the area of ​​the scanned image.

6. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The packaging integrity of each monitoring segment is analyzed in detail as follows: A gray-level co-occurrence matrix is ​​constructed based on the packaging accuracy characteristics of each monitoring segment. The gray-level co-occurrence matrix is ​​used to describe the spatial distribution relationship of pixel pairs with different gray values ​​in the image, which can quantify the surface texture of the packaging and thus analyze the surface accuracy of the packaging. Based on the constructed gray-level co-occurrence matrix, feature parameters such as energy, contrast, correlation, and entropy are extracted, and a weighted average is performed on the feature parameters to calculate the packaging surface integrity coefficient.

7. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The specific method for the correlation analysis of packaging production quality is as follows: A two-dimensional coordinate system is constructed with the production area temperature of each monitoring segment as the horizontal axis and the packaging surface stability coefficient as the vertical axis. Thus, the production area temperature and packaging surface stability coefficient of each monitoring segment are marked in the constructed two-dimensional coordinate system to form a packaging surface stability change curve. A two-dimensional coordinate system is constructed with the production area temperature of each monitoring segment as the horizontal axis and the packaging surface integrity coefficient as the vertical axis. Thus, the production area temperature and packaging surface integrity coefficient of each monitoring segment are marked at each point in the constructed two-dimensional coordinate system to form a packaging surface integrity change curve. The correlation coefficient between the stable change curve of the packaging surface and the intact change curve of the packaging surface is calculated to obtain the correlation between the packaging stability and integrity in the second monitoring area.

8. The intelligent monitoring method for packaging material production quality based on the Industrial Internet of Things as described in claim 1, characterized in that: The adaptive adjustment specifically involves comparing the correlation between packaging stability and integrity in the second monitoring area with a preset positive correlation. Monitoring segments with a correlation lower than the positive correlation are then selected. The temperature of the production equipment at this point is extracted and compared with a standard temperature to obtain the temperature deviation. If the temperature deviation is negative, the adjustment method is determined to be increasing the temperature; if the temperature deviation is positive, the adjustment method is determined to be decreasing the temperature. The required temperature adjustment is then determined based on the degree of temperature deviation. Specifically: Where Qt represents the required temperature adjustment, ΔT represents the temperature deviation, sign(ΔT) represents the sign function, which acts on the temperature change ΔT to determine the sign of ΔT, and takes values ​​of 1, 0, and -1. If sign(ΔT) = 1, it means that the current production equipment temperature is greater than the required production equipment temperature, and the temperature needs to be lowered. If sign(ΔT) = -1, it means that the current production equipment temperature is less than the required production equipment temperature, and the temperature needs to be raised. If sign(ΔT) = 0, then no temperature adjustment is needed. t0 represents the temperature deviation threshold.

Images