CN118938847A - Wet wipes quality control system based on artificial intelligence - Google Patents
Wet wipes quality control system based on artificial intelligence Download PDFInfo
- Publication number
- CN118938847A CN118938847A CN202411433065.5A CN202411433065A CN118938847A CN 118938847 A CN118938847 A CN 118938847A CN 202411433065 A CN202411433065 A CN 202411433065A CN 118938847 A CN118938847 A CN 118938847A
- Authority
- CN
- China
- Prior art keywords
- data
- quality
- wet
- production
- adjustment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003908 quality control method Methods 0.000 title claims abstract description 24
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 15
- 238000004519 manufacturing process Methods 0.000 claims abstract description 94
- 230000002159 abnormal effect Effects 0.000 claims abstract description 28
- 238000005457 optimization Methods 0.000 claims abstract description 12
- 238000007405 data analysis Methods 0.000 claims abstract description 8
- 238000011156 evaluation Methods 0.000 claims abstract description 7
- 230000000007 visual effect Effects 0.000 claims abstract description 7
- 230000007547 defect Effects 0.000 claims description 34
- 238000004458 analytical method Methods 0.000 claims description 28
- 238000010438 heat treatment Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 17
- 238000012795 verification Methods 0.000 claims description 17
- 238000012544 monitoring process Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000007619 statistical method Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 239000000463 material Substances 0.000 claims description 4
- 230000035945 sensitivity Effects 0.000 claims description 4
- 208000012826 adjustment disease Diseases 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000010191 image analysis Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 4
- 238000007689 inspection Methods 0.000 abstract description 4
- 230000008713 feedback mechanism Effects 0.000 abstract description 2
- 230000007246 mechanism Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000010972 statistical evaluation Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32252—Scheduling production, machining, job shop
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Factory Administration (AREA)
Abstract
The invention relates to the technical field of quality control, in particular to an artificial intelligence-based wet tissue quality control system, which comprises an abnormal data analysis module, a production adjustment module, a visual evaluation module and an optimization feedback module, wherein the system is used for collecting wet tissue quality data and machine running state data based on sensors in a wet tissue production line. The invention can accurately identify the tiny deviation on the production line and quickly make adjustment by continuously collecting the production data and analyzing in real time, the automatic reaction mechanism obviously reduces the dependence on manual inspection, improves the production efficiency and the consistency of products, the anomaly detection and real-time adjustment ensure the quality standard of each batch of products, effectively reduces the rejection rate and the production cost, the visual evaluation application further enhances the control on the appearance quality of the products, ensures that each wet towel meets strict quality requirements, and the optimization feedback mechanism utilizes the deep data analysis to provide accurate basis for continuously improving the production flow.
Description
Technical Field
The invention relates to the technical field of quality control, in particular to a wet tissue quality control system based on artificial intelligence.
Background
The wet towel quality control system is a system specially used for monitoring and guaranteeing quality standards in the wet towel production process, and mainly used for guaranteeing consistency and safety of quality of wet towels of each batch, including multiple indexes such as humidity, fiber density and sterile state.
Traditional systems rely on periodic manual inspection and partial automation equipment to delay response when handling sudden quality problems on high-speed production lines, for example, the identification of wet tissue humidity and thickness deviation is not timely, resulting in the quality of the whole batch of products being affected, data analysis in traditional systems is limited to simple statistical evaluation, deep learning and real-time feedback capabilities are lacking, the capability of rapid decision in the production process is limited, such technical gaps lead to production efficiency reduction, production cost and resource waste are increased, and the lack of real-time visual monitoring also often leads to the failure of product appearance defects to be timely identified, thereby affecting the product quality.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides an artificial intelligence-based wet tissue quality control system.
In order to achieve the above purpose, the present invention adopts the following technical scheme: an artificial intelligence based wet wipe quality control system, the system comprising:
The abnormal data analysis module collects wet towel quality data and machine operation state data based on sensors in the wet towel production line, and generates abnormal index information by collecting wet towel humidity and thickness information and combining production speed and temperature state to identify abnormal indexes in the data;
The production adjustment module adjusts key control parameters of the wet towel production line based on the abnormal index information, wherein the key control parameters comprise parameters of a humidity controller, a heating controller and a speed controller, evaluates the adjusted control parameters, verifies the adjustment reaction effect through simulating the production condition, and generates an adjustment verification result;
The visual evaluation module captures wet tissue images on a production line by utilizing a camera based on the adjustment verification result, performs feature extraction on the captured images, evaluates texture and folding quality of the wet tissue, classifies each wet tissue by comparing with quality standards, and generates quality classification records;
And the optimization feedback module is used for counting unqualified wet tissues marked based on the quality classification record, analyzing the proportion and defect type of the unqualified wet tissues, adjusting and optimizing production line control parameters, data monitoring and image analysis parameters according to analysis results, and generating a control parameter tuning log.
The invention improves that the abnormal index information is obtained by the following steps:
Based on sensors in a wet towel production line, collecting real-time data, including wet towel humidity, wet towel thickness, production speed and temperature state, performing data primary arrangement and standardization, and generating a primary sensor data set;
Based on the preliminary sensor dataset, each data point is normalized using a Z-score, using the formula:
，
Determining the degree of deviation in the data, generating a normalized set of data points, wherein, The value of the observation is represented by a value,Is the mean value of the data set,Is the standard deviation of the data set,Is an adjustment factor for optimizing the standard deviation effect,Z-score representing data point;
identifying abnormal value of the standardized data point set, and identifying data points with Z score absolute value greater than 2, namely And marking the data points as abnormal, and obtaining abnormal index information.
The invention is improved in that the key control parameter adjusting step comprises the following steps:
Analyzing the abnormal index information, including humidity, heating and speed data, determining deviation from the current production standard, and generating deviation analysis information;
based on the deviation analysis information, the formula is adopted:
，
adjusting parameters of the humidity controller, the heating controller and the speed controller to obtain new control parameter settings, wherein, Indicating that a new parameter setting is to be made,Is the old parameter set-up and,Is the sensitivity coefficient of the target humidity,Is the standard deviation of the history deviation and,Is the target humidity level of the air in the air,Is the current humidity.
The invention improves that the acquisition steps of the adjustment verification result are as follows:
Performing simulated production test by applying new control parameters, monitoring key production indexes including wet towel humidity, heating efficiency and speed, and performing real-time data collection to obtain production test data;
Based on the production test data, statistical analysis is performed, the validity of control parameter adjustment is verified, and the formula is adopted:
，
an adjustment verification effect is obtained, wherein, Is to adjust the score of the difference between the front and rear average values,AndRespectively represents the average production index after adjustment and before adjustment,AndRepresenting the variance of the corresponding sample and,AndRepresenting the new and old sample sizes.
The invention improves that the steps for extracting the characteristics of the captured image specifically comprise:
based on the adjustment verification result, configuring a camera to capture a wet tissue image on the production line, and adjusting the light and the camera angle setting to generate a wet tissue original image;
Processing the wet tissue original image by using an image processing technology, wherein the processing comprises contrast adjustment, graying and noise filtering, optimizing the analysis applicability of the image and generating optimized image data;
and extracting the characteristics of the optimized image data, wherein the formula is adopted:
，
Calculating sharpness index of image texture Wherein, the method comprises the steps of, wherein,AndIs the width and height of the image and,AndRespectively the images are atAndGradient in direction.
The invention improves that the quality classification record is obtained by the following steps:
Evaluating the texture and folding characteristics of each wet tissue by utilizing the data extracted by the characteristics, comparing with quality standards and classifying to generate a preliminary quality rating result;
based on the preliminary quality rating result, the formula is adopted:
，
Calculating the mass fraction of each wet tissue A quality classification record is obtained, wherein,Is the score of the texture clarity of the wet wipe,Representing a fold quality score of the sheet,Is a target threshold value for quality control,AndIs the weight coefficient of the weight of the object,Is a constant for normalizing the result.
The invention improves the steps of analyzing the proportion and defect type of the unqualified wet tissues specifically as follows:
based on the quality classification record, unqualified wet tissue information is extracted from production data, wherein the unqualified wet tissue information comprises defect information and quality scores of each wet tissue, and an unqualified wet tissue list is obtained;
based on the list of rejected wet wipes, the formula is adopted:
，
Calculating the proportion of the total number of the unqualified wet tissues to the production process to obtain the proportion information of the unqualified wet tissues, wherein, The percentage of the reject product is indicated,Is the number of unqualified wet tissues,Is the total wet tissue quantity produced;
And analyzing the defect types, including distribution conditions of textures and folding defects and occurrence rate of the defects in the unqualified wet tissues, based on the unqualified wet tissue proportion information, and generating a defect type analysis result.
The invention improves that the control parameter tuning log is obtained by the following steps:
Determining key control parameters affecting the product quality according to the analysis result, including the running speed, the heating temperature and the humidity level of the production line, and collecting the current setting and the performance of the parameters to obtain a parameter list to be adjusted;
Based on the parameter list to be adjusted, carrying out refinement adjustment, and adopting the formula:
，
an optimal parameter setting is obtained, wherein, Is the value of the parameter after the optimization,Is the current parameter setting and is used to determine,Is the target mass of the material to be processed,Representing a deviation of the current product quality from a target quality;
Based on the optimal parameter setting, monitoring and tracking the adjustment effect through real-time data, analyzing whether each parameter change optimizes the product quality, and continuously recording each item of adjusted data to obtain a control parameter tuning log.
Compared with the prior art, the invention has the advantages and positive effects that:
In the invention, the production data are continuously collected and analyzed in real time, the micro deviation on the production line can be accurately identified and the adjustment is quickly made, the automatic reaction mechanism obviously reduces the dependence on manual inspection, improves the production efficiency and the product consistency, the anomaly detection and the real-time adjustment ensure the quality standard of each batch of products, effectively reduces the rejection rate and the production cost, the visual evaluation application further enhances the control on the appearance quality of the products, ensures that each wet tissue meets strict quality requirements, and the optimization feedback mechanism utilizes the deep data analysis to provide accurate basis for continuously improving the production flow, thereby fundamentally improving the control efficiency.
Drawings
FIG. 1 is a system flow diagram of the present invention;
FIG. 2 is a flowchart for obtaining anomaly index information in the present invention;
FIG. 3 is a flow chart of the adjustment of key control parameters according to the present invention;
FIG. 4 is a flowchart of the method for obtaining the verification result according to the present invention;
FIG. 5 is a flow chart of feature extraction of a captured image in the present invention;
FIG. 6 is a flow chart of the acquisition of quality classification records in the present invention;
FIG. 7 is a flow chart of an analysis of the proportion and defect type of rejected wet wipes in accordance with the invention;
fig. 8 is a flowchart of the control parameter tuning log acquisition in the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," etc. indicate or are based on the orientation or positional relationship shown in the drawings, merely to facilitate description of the invention and simplify description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Referring to fig. 1, the present invention provides a technical solution: the wet tissue quality control system based on artificial intelligence comprises:
The abnormal data analysis module collects wet towel quality data and machine operation state data based on sensors in the wet towel production line, and generates abnormal index information by collecting wet towel humidity and thickness information and combining production speed and temperature state to identify abnormal indexes in the data;
The production adjustment module adjusts key control parameters of the wet towel production line based on abnormal index information, including parameters of a humidity controller, a heating controller and a speed controller, evaluates the adjusted control parameters, verifies the adjustment reaction effect through simulating production conditions, and generates an adjustment verification result;
the visual evaluation module captures wet tissue images on a production line by utilizing a camera based on adjustment verification results, performs feature extraction on the captured images, evaluates texture and folding quality of the wet tissues, classifies each wet tissue by comparing with quality standards, and generates quality classification records;
The optimization feedback module is used for counting unqualified wet tissues based on the quality classification record, analyzing the proportion and defect type of the unqualified wet tissues, adjusting and optimizing production line control parameters, data monitoring and image analysis parameters according to analysis results, and generating a control parameter tuning log.
The adjustment verification results comprise parameter adjustment effects, production stability and adjustment response time, the quality classification records comprise texture qualification ratings, folding integrity inspection, qualified product identification results and unqualified product screening results, and the control parameter adjustment log comprises parameter adjustment records, defect type analysis results and optimization effect evaluation results.
Referring to fig. 2, the steps for obtaining the anomaly index information specifically include:
Based on sensors in a wet towel production line, collecting real-time data, including wet towel humidity, wet towel thickness, production speed and temperature state, performing data primary arrangement and standardization, and generating a primary sensor data set;
Based on the sensor device on the wet towel production line for data acquisition, the deployment of the sensors relates to various key production nodes, including humidity control, thickness measurement points, speed monitoring points and temperature sensing areas, the sensors of each point are used for capturing specific types of data, for example, the humidity sensors are installed near the discharge holes to ensure that the humidity of products meets standard specifications, the thickness measurement is performed immediately after the cutting machine so as to confirm the consistency of each batch of products, the speed sensors record the running speed of materials to ensure the continuity and efficiency of the production flow, the temperature sensors are arranged in the drying area and the cooling area to monitor key heat treatment processes, the collected data are subjected to arrangement and format standardization to ensure the consistency and the usability of all the data, and thus a primarily arranged sensor data set is generated.
Based on the preliminary sensor dataset, each data point is normalized using the Z-score, using the formula:
，
Determining the degree of deviation in the data, generating a normalized set of data points, wherein, The value of the observation is represented by a value,Is the mean value of the data set,Is the standard deviation of the data set,Is an adjustment factor for optimizing the standard deviation effect,Is a constant to prevent zero-divide errors,Z-score representing data point;
There is a set of data collected:
calculate the average value thereof And standard deviationIs provided withAndThe calculation process is as follows:
Calculation of ，
A Z-score is applied to each data point,
For a pair ofThe Z score of (2) is:
；
For a pair of The Z score of (2) is:
；
For a pair of The Z score of (2) is:
；
For a pair of The Z score of (2) is:
；
For a pair of The Z score of (2) is:
；
results show no data points All data points are not identified as anomalies, effectively helping to identify behavior within the normal fluctuation range of the data.
Abnormal value identification is carried out on the standardized data point set, and the data points with Z scores of which the absolute values are more than 2 are identified, namelyMarking the data points as abnormal, and obtaining abnormal index information;
according to the abnormal value identification process of the standardized data point set, the data is subjected to the standardized processing to obtain As a standard for anomaly detection, the Z-score herein is the result of the deviation of the data points from the mean of the collective data by the standard deviation adjustment, during which each data point is calculated separately and compared to a threshold value, the identified anomaly value is marked and recorded, which is then used for further analysis, such as quality control and production adjustment, to ensure consistency of product quality and optimization of production efficiency.
Referring to fig. 3, the key control parameter adjustment steps specifically include:
analyzing the abnormal index information, including humidity, heating and speed data, determining deviation from the current production standard, and generating deviation analysis information;
the abnormal index information is analyzed, data of humidity, heating and speed are focused, the data are displayed and the deviation of production standard is set, and the deviation of control parameters is confirmed to be large by carrying out statistical analysis on the centralized trend and the decentralized trend of the data, for example, the humidity controller has inaccurate reading caused by sensor faults or environmental changes, the speed controller influences the accuracy of the humidity controller due to mechanical abrasion, a deviation analysis report is generated, and the difference between the current value and the target value of each control parameter is listed in the report in detail, so that the basis is provided for the parameter adjustment of the next step.
Based on the deviation analysis information, the formula is adopted:
，
adjusting parameters of the humidity controller, the heating controller and the speed controller to obtain new control parameter settings, wherein, Indicating that a new parameter setting is to be made,Is the old parameter set-up and,Is the sensitivity coefficient of the target humidity,Is the standard deviation of the history deviation and,Is the target humidity level of the air in the air,Is the current humidity.
Monitoring parameters of current humidity controller50, Target humidity30% Of the current humidity35% Standard deviation of history deviation3 Percent,5;
Calculating the ratio of sensitivity to standard deviation:
；
Calculating an adjustment value of the target and the current humidity difference:
；
calculating new control parameters:
；
the results indicate that the adjusted control parameter is 41.65, which will help the humidity controller to be closer to the target setting, reducing product errors.
Referring to fig. 4, the steps for obtaining the adjustment verification result specifically include:
Performing simulated production test by applying new control parameters, monitoring key production indexes including wet towel humidity, heating efficiency and speed, and performing real-time data collection to obtain production test data;
The simulation production test is carried out, the test is mainly aimed at the newly adjusted control parameters, the performances of the humidity controller, the heating controller and the speed controller are particularly concerned, the data of key performance indexes are collected, the data are used for evaluating the influence and the applicability of the new parameters, the change of each index is monitored and recorded in real time, the accuracy and the integrity of the data are ensured, meanwhile, the obtained data can reflect the actual production state through the data collecting equipment, the data report containing the humidity level, the heating efficiency and the machine speed adjusting reaction is generated, the data report is taken as the basis for evaluating the effect of the new control parameters, and necessary input information is provided for the next data analysis.
Based on the production test data, statistical analysis is performed to verify the validity of the adjustment of the control parameters, and the formula is adopted:
，
an adjustment verification effect is obtained, wherein, Is to adjust the score of the difference between the front and rear average values,AndRespectively represents the average production index after adjustment and before adjustment,AndRepresenting the variance of the corresponding sample and,AndRepresenting the new and old sample sizes.
Average production speeds before and after adjustment are respectivelyPer hourSample variance per hour is respectivelyAndSample sizes are respectivelyAnd；
The variance of each set of data is calculated divided by the sample size:
，
；
Calculating denominator:
；
Calculating a final value:
；
the result shows that the production speed after adjustment is obviously improved relative to that before adjustment, which indicates that the setting of the new control parameters effectively improves the production efficiency.
Referring to fig. 5, the steps of feature extraction on the captured image specifically include:
based on the adjustment verification result, configuring a camera to capture a wet tissue image on the production line, and adjusting the light and the camera angle setting to generate a wet tissue original image;
The high-resolution camera is configured to capture wet tissue images on a production line, so that the definition of the images is ensured, a foundation is laid for subsequent feature extraction, a high-quality original image is generated, the details of the images are fully shown by adjusting the light rays and angles of the camera and optimizing the image capturing conditions, the importance of subsequent image processing and analysis is ensured, and the obtained image data directly influences the effect and accuracy of feature extraction, so that the strategy and method of the next image preprocessing are determined.
Processing the wet tissue original image by using an image processing technology, wherein the processing comprises contrast adjustment, graying and noise filtering, optimizing the analysis applicability of the image and generating optimized image data;
The method comprises the steps of processing an original image by using an image processing technology, wherein the image processing technology comprises contrast adjustment, graying processing and noise removal, the quality of image data is optimized by processing, the image data is more suitable for detailed feature analysis, optimized image data is obtained, important features in the image, such as edges and textures, are enhanced, the method is very critical to machine learning algorithm recognition and classification of objects in the image, and the optimized image is used for more accurate feature extraction, so that the efficiency and accuracy of the whole quality detection are improved.
Extracting features of the optimized image data, and adopting the formula:
，
Calculating sharpness index of image texture Wherein, the method comprises the steps of, wherein,AndIs the width and height of the image and,AndRespectively the images are atAndGradient in direction.
The following data were collected:
The image size is Pixel width sumPixel height, gradient for one point is calculatedAndSubstituting the formula:
calculating the square root of the sum of the squares of the gradients of each pixel point:
；
the average of the total gradients was calculated:
；
the results indicate that the image has a texture clarity score of 25, representing the average clarity level of the image texture, providing a quantitative method to evaluate the texture and surface quality of the wet wipe.
Referring to fig. 6, the steps for obtaining the quality classification record specifically include:
evaluating the texture and folding characteristics of each wet tissue by utilizing the data extracted by the characteristics, comparing with quality standards and classifying to generate a preliminary quality rating result;
after texture and folding feature data are extracted from the wet tissues, the data are processed, each feature is quantized, necessary basis is provided for quality evaluation, by comparing the quality standard with a machine learning model, each wet tissue can be effectively classified, the scoring of each feature data is included, the scoring is compared with a set quality threshold, the scoring includes calculating texture definition, folding uniformity and other relevant quality indexes, each wet tissue is endowed with a quality grade according to the indexes, the grade reflects the comprehensive quality of the wet tissue, and only products meeting the standard can enter the next step of sales or further processing, and a preliminary quality rating result is generated.
Based on the preliminary quality rating result, the formula is adopted:
，
Calculating the mass fraction of each wet tissue A quality classification record is obtained, wherein,Is the score of the texture clarity of the wet wipe,Representing a fold quality score of the sheet,Is a target threshold value for quality control,AndIs a weight coefficient, reflects the importance of the difference feature in the quality assessment,Is a constant for normalizing the result, preventing the score from being too high or too low;
The following data were collected:
；
；
；
And ；
；
Calculating a weighted sum of texture and fold scores:
；
Calculating mass fraction:
；
The result shows that the comprehensive mass fraction is 8.3, reflects the performance of the wet tissue in texture and folding quality, and provides a basis for quality classification by matching with the quality standard, thereby ensuring the quality control and improved effectiveness of the product.
Referring to fig. 7, the steps of analyzing the proportion and defect type of the unqualified wet tissues are specifically as follows:
based on the quality classification record, unqualified wet tissue information is extracted from production data, wherein the unqualified wet tissue information comprises defect information and quality scores of each wet tissue, and an unqualified wet tissue list is obtained;
Based on the existing quality classification records, all wet tissue data marked as unqualified are extracted, including defect information and rating results of each wet tissue, and the extraction process of the data involves data query and data screening, so that detailed information of all unqualified products is ensured to be accurately acquired from a large amount of production data, and a data set containing specific defect types and numbers is generated.
Based on the list of rejected wet wipes, the formula is adopted:
，
Calculating the proportion of the total number of the unqualified wet tissues to the production process to obtain the proportion information of the unqualified wet tissues, wherein, The percentage of the reject product is indicated,Is the number of unqualified wet tissues,Is the total wet tissue quantity produced;
the following data were collected, total production 5000 Wet tissues, wherein the wet tissues are rejected300 A;
Calculating the disqualification ratio:
，
The results indicate that 6% of the wet wipes in the current manufacturing lot do not meet the quality standards, indicating quality control issues that need to be addressed during the manufacturing process.
Analyzing defect types, including distribution conditions of textures and folding defects and occurrence rate of the defects in the unqualified wet tissues, based on the unqualified wet tissue proportion information, and generating defect type analysis results;
the obtained unqualified proportion data is utilized to further analyze defect types in the unqualified wet tissues, the occurrence frequency of each defect type is analyzed through a classification statistical method, the defect types comprise a plurality of types such as unclear textures and incorrect folding, the statistics of each defect is not only based on quantity, but also the proportion of each defect in the unqualified wet tissues is considered, so that the main production problem is revealed, the accuracy and the reliability of the result are ensured by means of data processing and analysis technology in the analysis process, and an analysis log comprising the main defect types and descriptions is generated.
Referring to fig. 8, the control parameter tuning log obtaining steps specifically include:
according to the analysis result, determining key control parameters affecting the product quality, including the running speed, heating temperature and humidity level of the production line, and collecting the current setting and performance of the parameters to obtain a parameter list to be adjusted;
On the basis of analyzing unqualified wet tissues, control parameter data related to quality defects, such as humidity setting, temperature control and linear speed adjustment, are collected, the parameters directly influence the qualification rate of products, the current control parameters are integrated through a systematic data acquisition method, the set value and real-time performance of each parameter are recorded, basic data are provided for subsequent adjustment, and a parameter list to be optimized is generated.
Based on the parameter list to be adjusted, carrying out refinement adjustment, and adopting the formula:
，
an optimal parameter setting is obtained, wherein, Is the value of the parameter after the optimization,Is the current parameter setting and is used to determine,Is the target mass of the material to be processed,Representing a deviation of the current product quality from a target quality;
The following data were collected:
50;
90%;
is-10%;
Calculating new parameters:
，
the results indicate that by tuning, the new control parameters should be tuned to 55.55 in order to improve the quality of the production.
Based on the optimal parameter setting, monitoring a tracking adjustment effect through real-time data, analyzing whether each parameter change optimizes the product quality, and continuously recording each item of adjusted data to obtain a control parameter tuning log;
After adjusting the control parameters, a control test is implemented to verify the effectiveness of the new parameter setting, the parameter adjustment effect is continuously tracked through data monitoring, the product quality is obviously improved through each adjustment, a control parameter adjustment log is formed through a real-time data feedback adjustment strategy, the whole process from parameter analysis to optimization implementation is recorded by the log, and support is provided for continuously improving the production line efficiency and the product quality.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.
Claims (8)
1. An artificial intelligence based wet wipe quality control system, the system comprising:
The abnormal data analysis module collects wet towel quality data and machine operation state data based on sensors in the wet towel production line, and generates abnormal index information by collecting wet towel humidity and thickness information and combining production speed and temperature state to identify abnormal indexes in the data;
The production adjustment module adjusts key control parameters of the wet towel production line based on the abnormal index information, wherein the key control parameters comprise parameters of a humidity controller, a heating controller and a speed controller, evaluates the adjusted control parameters, verifies the adjustment reaction effect through simulating the production condition, and generates an adjustment verification result;
The visual evaluation module captures wet tissue images on a production line by utilizing a camera based on the adjustment verification result, performs feature extraction on the captured images, evaluates texture and folding quality of the wet tissue, classifies each wet tissue by comparing with quality standards, and generates quality classification records;
And the optimization feedback module is used for counting unqualified wet tissues marked based on the quality classification record, analyzing the proportion and defect type of the unqualified wet tissues, adjusting and optimizing production line control parameters, data monitoring and image analysis parameters according to analysis results, and generating a control parameter tuning log.
2. The artificial intelligence-based wet tissue quality control system according to claim 1, wherein the step of obtaining the abnormality index information specifically includes:
Based on sensors in a wet towel production line, collecting real-time data, including wet towel humidity, wet towel thickness, production speed and temperature state, performing data primary arrangement and standardization, and generating a primary sensor data set;
Based on the preliminary sensor dataset, each data point is normalized using a Z-score, using the formula:
，
Determining the degree of deviation in the data, generating a normalized set of data points, wherein, The value of the observation is represented by a value,Is the mean value of the data set,Is the standard deviation of the data set,Is an adjustment factor for optimizing the standard deviation effect,Z-score representing data point;
identifying abnormal value of the standardized data point set, and identifying data points with Z score absolute value greater than 2, namely And marking the data points as abnormal, and obtaining abnormal index information.
3. The artificial intelligence based wet wipe quality control system of claim 1, wherein the step of adjusting the key control parameters is specifically:
Analyzing the abnormal index information, including humidity, heating and speed data, determining deviation from the current production standard, and generating deviation analysis information;
based on the deviation analysis information, the formula is adopted:
，
adjusting parameters of the humidity controller, the heating controller and the speed controller to obtain new control parameter settings, wherein, Indicating that a new parameter setting is to be made,Is the old parameter set-up and,Is the sensitivity coefficient of the target humidity,Is the standard deviation of the history deviation and,Is the target humidity level of the air in the air,Is the current humidity.
4. The artificial intelligence based wet tissue quality control system according to claim 1, wherein the step of obtaining the adjustment verification result specifically comprises:
Performing simulated production test by applying new control parameters, monitoring key production indexes including wet towel humidity, heating efficiency and speed, and performing real-time data collection to obtain production test data;
Based on the production test data, statistical analysis is performed, the validity of control parameter adjustment is verified, and the formula is adopted:
，
an adjustment verification effect is obtained, wherein, Is to adjust the score of the difference between the front and rear average values,AndRespectively represents the average production index after adjustment and before adjustment,AndRepresenting the variance of the corresponding sample and,AndRepresenting the new and old sample sizes.
5. The artificial intelligence based wet wipe quality control system of claim 1, wherein the step of feature extraction of the captured image is specifically:
based on the adjustment verification result, configuring a camera to capture a wet tissue image on the production line, and adjusting the light and the camera angle setting to generate a wet tissue original image;
Processing the wet tissue original image by using an image processing technology, wherein the processing comprises contrast adjustment, graying and noise filtering, optimizing the analysis applicability of the image and generating optimized image data;
and extracting the characteristics of the optimized image data, wherein the formula is adopted:
，
Calculating sharpness index of image texture Wherein, the method comprises the steps of, wherein,AndIs the width and height of the image and,AndRespectively the images are atAndGradient in direction.
6. The artificial intelligence based wet wipe quality control system of claim 1, wherein the step of obtaining the quality classification record is specifically:
Evaluating the texture and folding characteristics of each wet tissue by utilizing the data extracted by the characteristics, comparing with quality standards and classifying to generate a preliminary quality rating result;
based on the preliminary quality rating result, the formula is adopted:
，
Calculating the mass fraction of each wet tissue A quality classification record is obtained, wherein,Is the score of the texture clarity of the wet wipe,Representing a fold quality score of the sheet,Is a target threshold value for quality control,AndIs the weight coefficient of the weight of the object,Is a constant for normalizing the result.
7. The artificial intelligence based wet wipe quality control system of claim 1, wherein the step of analyzing the proportion and defect type of the rejected wet wipes is specifically:
based on the quality classification record, unqualified wet tissue information is extracted from production data, wherein the unqualified wet tissue information comprises defect information and quality scores of each wet tissue, and an unqualified wet tissue list is obtained;
based on the list of rejected wet wipes, the formula is adopted:
，
Calculating the proportion of the total number of the unqualified wet tissues to the production process to obtain the proportion information of the unqualified wet tissues, wherein, The percentage of the reject product is indicated,Is the number of unqualified wet tissues,Is the total wet tissue quantity produced;
And analyzing the defect types, including distribution conditions of textures and folding defects and occurrence rate of the defects in the unqualified wet tissues, based on the unqualified wet tissue proportion information, and generating a defect type analysis result.
8. The artificial intelligence based wet tissue quality control system according to claim 1, wherein the step of obtaining the control parameter tuning log specifically comprises:
Determining key control parameters affecting the product quality according to the analysis result, including the running speed, the heating temperature and the humidity level of the production line, and collecting the current setting and the performance of the parameters to obtain a parameter list to be adjusted;
Based on the parameter list to be adjusted, carrying out refinement adjustment, and adopting the formula:
，
an optimal parameter setting is obtained, wherein, Is the value of the parameter after the optimization,Is the current parameter setting and is used to determine,Is the target mass of the material to be processed,Representing a deviation of the current product quality from a target quality;
Based on the optimal parameter setting, monitoring and tracking the adjustment effect through real-time data, analyzing whether each parameter change optimizes the product quality, and continuously recording each item of adjusted data to obtain a control parameter tuning log.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411433065.5A CN118938847A (en) | 2024-10-15 | 2024-10-15 | Wet wipes quality control system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411433065.5A CN118938847A (en) | 2024-10-15 | 2024-10-15 | Wet wipes quality control system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118938847A true CN118938847A (en) | 2024-11-12 |
Family
ID=93352702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411433065.5A Pending CN118938847A (en) | 2024-10-15 | 2024-10-15 | Wet wipes quality control system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118938847A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119169010A (en) * | 2024-11-21 | 2024-12-20 | 青岛理工大学 | Dynamic production line measurement and control method and system based on visual recognition |
| CN119717729A (en) * | 2024-12-19 | 2025-03-28 | 东台派瑞斯五金有限公司 | Intelligent control system for feed production |
| CN119845824A (en) * | 2025-03-19 | 2025-04-18 | 珠海东美纸业有限公司 | Sanitary towel comfort level optimizing system based on AI intelligent analysis |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20210002142A (en) * | 2019-06-25 | 2021-01-07 | (주)아이리녹스 | Smart plant construction system for tissue paper manufacturing |
| CN112557406A (en) * | 2021-02-19 | 2021-03-26 | 浙江大胜达包装股份有限公司 | Intelligent inspection method and system for paper product production quality |
| CN117115147A (en) * | 2023-10-19 | 2023-11-24 | 山东华盛创新纺织科技有限公司 | Textile detection method and system based on machine vision |
| CN118550263A (en) * | 2024-07-30 | 2024-08-27 | 山东固风科技有限公司 | Control system and method for silicone sealant production |
| CN118707908A (en) * | 2024-06-11 | 2024-09-27 | 广州市鑫富塑胶有限公司 | A method and system for upgrading and optimizing production parameters of glue needle gun |
| CN118747479A (en) * | 2024-07-03 | 2024-10-08 | 元光亿科技(上海)有限公司 | A digital twin object processing system and method |
-
2024
- 2024-10-15 CN CN202411433065.5A patent/CN118938847A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20210002142A (en) * | 2019-06-25 | 2021-01-07 | (주)아이리녹스 | Smart plant construction system for tissue paper manufacturing |
| CN112557406A (en) * | 2021-02-19 | 2021-03-26 | 浙江大胜达包装股份有限公司 | Intelligent inspection method and system for paper product production quality |
| CN117115147A (en) * | 2023-10-19 | 2023-11-24 | 山东华盛创新纺织科技有限公司 | Textile detection method and system based on machine vision |
| CN118707908A (en) * | 2024-06-11 | 2024-09-27 | 广州市鑫富塑胶有限公司 | A method and system for upgrading and optimizing production parameters of glue needle gun |
| CN118747479A (en) * | 2024-07-03 | 2024-10-08 | 元光亿科技(上海)有限公司 | A digital twin object processing system and method |
| CN118550263A (en) * | 2024-07-30 | 2024-08-27 | 山东固风科技有限公司 | Control system and method for silicone sealant production |
Non-Patent Citations (2)
| Title |
|---|
| 何玉青，等: "《光学与光电检测系统》", 31 August 2023, 国防工业出版社, pages: 282 - 283 * |
| 甘平: "《统计学》", 30 November 2014, 北京理工大学出版社, pages: 108 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119169010A (en) * | 2024-11-21 | 2024-12-20 | 青岛理工大学 | Dynamic production line measurement and control method and system based on visual recognition |
| CN119169010B (en) * | 2024-11-21 | 2025-03-25 | 青岛理工大学 | Dynamic production line measurement and control method and system based on visual recognition |
| CN119717729A (en) * | 2024-12-19 | 2025-03-28 | 东台派瑞斯五金有限公司 | Intelligent control system for feed production |
| CN119845824A (en) * | 2025-03-19 | 2025-04-18 | 珠海东美纸业有限公司 | Sanitary towel comfort level optimizing system based on AI intelligent analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118938847A (en) | Wet wipes quality control system based on artificial intelligence | |
| US20130193123A1 (en) | Method for inspecting the quality of a solder joint | |
| CN117350967B (en) | Optical detection-based wafer appearance defect detection system and method | |
| CN119148659B (en) | Intelligent shoe production line upper spraying control method and system | |
| CN111402236A (en) | Hot-rolled strip steel surface defect grading method based on image gray value | |
| CN117686516A (en) | Automatic chip appearance defect detection system based on machine vision | |
| CN117115169A (en) | Intelligent recognition method for abnormal deformation of surface of die-casting die of automobile part | |
| CN118037137A (en) | Method for determining product quality accident number based on convolutional neural network | |
| CN116664540A (en) | Surface defect detection method of rubber sealing ring based on Gaussian line detection | |
| CN119304387B (en) | An automated monitoring method and system for metal sheet laser cutting operations | |
| CN118864464B (en) | Intelligent management system for textile artware | |
| CN119780587A (en) | A Fault Diagnosis Method and System for Photovoltaic Inverter | |
| CN118657764B (en) | Equipment defect detection method and system based on image recognition | |
| CN117952482B (en) | Product quality accident grading method and system based on convolutional neural network | |
| CN118839228A (en) | Industrial flow operation monitoring system based on data analysis | |
| KR101782366B1 (en) | Vision inspection method based on learning data using normalization of sample scale | |
| CN119004991A (en) | Method for testing adhesive force of flat wire | |
| CN117929418A (en) | Integrated circuit defect detection method and system | |
| CN116772718A (en) | An online detection method and system for cigarette laser drilling based on linear array vision | |
| CN118076929A (en) | In particular, using self-organizing maps to monitor the production of material panels, in particular engineered wood panels | |
| CN118708914B (en) | Rotating equipment fault diagnosis method and device | |
| CN118278653B (en) | A device defect detection system based on visual robot | |
| CN117654907B (en) | Automatic eliminating method and system for strip detector | |
| CN119559173B (en) | A PCB production line defect detection analysis method and system based on recognition model | |
| CN119722655B (en) | New energy vehicle cylinder cover detection system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |