Intelligent shoe production line upper spraying control method and system

Abstract

The present invention relates to the field of intelligent spray control technology, specifically to an intelligent shoe production line upper spray control method and system, comprising the following steps: collecting spray speed, spray amount, pressure and coating quality information, analyzing the influence of parameters on coating uniformity, coverage and drying speed according to the current spraying effect, and generating a parameter influence data set. The present invention calculates the influence of each parameter on the coating quality, and determines their priority in the control process in turn, and then compares the historical records with the current monitoring data, which can not only identify abnormal fluctuations, but also predict the potential abnormal conditions and locations of key objects in the future based on these data, and identify fault modes. The predictability and response speed of the system are improved. Combined with image processing technology, the coating thickness of the upper spray area is further analyzed through image comparison, which can more accurately detect thin coating, thick coating and missed spray areas, making the spraying process more stable and efficient.

Description

Technical Field

[0001] The present invention relates to the field of intelligent spraying control technology, and in particular to a method and system for controlling spraying of shoe uppers on an intelligent shoemaking production line. Background Art

[0002] Intelligent spray control technology integrates robotics, sensor technology, computer vision, and artificial intelligence algorithms to achieve precise control of the spray process. Intelligent spray systems automatically adjust spray parameters such as flow rate, pressure, and nozzle speed and path based on the shape and size of the workpiece to ensure uniformity and quality of the coating. They are widely used in industries such as automotive, aviation, furniture, and electronics to improve production efficiency, reduce material waste, and enhance the working environment.

[0003] The intelligent shoe production line's upper spraying control method is a type of intelligent spraying technology used in footwear manufacturing. During the upper manufacturing process, the spraying system is intelligently controlled to precisely apply a coating or pattern to the upper material. This method adjusts spraying parameters, such as spray speed, pressure, and color mixing, based on the material and design requirements of the individual upper, achieving efficient and consistent production results.

[0004] The existing technology mainly adapts to the material and design requirements of different uppers by adjusting parameters such as spraying speed, pressure and color mixing during the upper manufacturing process. However, the limitation of this method is that the adjustment of spraying parameters is mostly preset values, which makes the system susceptible to instantaneous fluctuations or sudden abnormalities in actual production. For example, during the production process, abnormal fluctuations in spraying pressure or sudden changes in speed may cause uneven coating, and inconsistent spraying is also prone to occur, especially when the upper material and design requirements change, the system is difficult to respond in time. In addition, although the existing technology can adjust the spraying parameters through intelligent control, it is difficult to predict potential problems in the future based on historical data analysis, which increases uncertainty and downtime in production, and affects the stability of production efficiency and product quality. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method and system for controlling the spraying of shoe uppers in an intelligent shoemaking production line.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a method for controlling the spraying of shoe uppers in an intelligent shoe production line, comprising the following steps:

[0007] S1: Collect information on spraying speed, spraying volume, pressure, and coating quality. Based on the current spraying effect, analyze the impact of the parameters on coating uniformity, coverage, and drying speed, and generate a parameter impact data set:

[0008] S2: Based on the parameter impact data set, calculate the parameter weight values, compare the size of each weight, arrange the weight values ​​in order to determine the priority in the control process, and generate a parameter weight table:

[0009] S3: Based on the parameter weight table, according to the currently monitored spraying pressure, nozzle status, and speed change information, the instantaneous fluctuation is compared with the historical records, abnormal fluctuations are identified, and a spraying behavior trend analysis result is generated;

[0010] S4: Based on the spraying behavior trend analysis results, predict the occurrence time and location of potential abnormalities in the future, identify the failure mode of the spraying equipment under specified conditions, and generate abnormal pattern recognition results;

[0011] S5: The acquisition camera captures images of the sprayed shoe upper. By comparing the thickness consistency of the sprayed areas, the image is analyzed for thin coating, thick coating, and missed coating areas to generate the spraying image quality assessment results:

[0012] S6: Based on the spray image quality assessment results and abnormal pattern recognition results, simulate the influence of various spray parameters on the quality of the shoe upper coating, adjust and update the speed and pressure parameters of the spray equipment according to the simulation results, and generate a spray optimization control result.

[0013] The improvements of the present invention are that the parameter influence data set includes a set of standardized values ​​of parameters, an influence index of parameters on spraying uniformity and coverage range, and an influence coefficient of parameters on drying speed; the parameter weight table includes a weighted numerical arrangement of spraying speed, spraying amount and pressure; the spraying behavior trend analysis results include a pressure change curve, a nozzle switch status diagram and a speed change trend diagram; the abnormal pattern recognition results include a time warning of potential faults, a location identification, and a classification of spraying equipment fault types; the spraying image quality assessment results include a coating thickness distribution diagram, a regional uniformity evaluation, the coordinates of the missed spraying area, and the proportion data of thin and thick coating areas; the spraying optimization control results include an adjusted spraying speed range, an optimal pressure setting value, and a spraying angle configuration.

[0014] The present invention has been improved in that the spraying speed, spraying amount, pressure and coating quality information are collected, and the influence of the parameters on the coating uniformity, coverage and drying speed is analyzed based on the current spraying effect. The specific steps for generating the parameter influence data set are as follows:

[0015] S101: Collect spray speed, spray volume, pressure and coating quality data, eliminate erroneous readings and outliers, verify data integrity, and generate raw data sets;

[0016] S102: Based on the original data set, adjust the data to the same scale, eliminate the difference between measurement units, and generate a processed data set;

[0017] S103: Based on the processed data set, according to the current coating uniformity, coverage and drying speed data, the correlation between each parameter and the spraying effect is analyzed, the influence of the parameter on the coating characteristics is calculated, and a parameter influence data set is generated.

[0018] The present invention is improved in that, based on the parameter influence data set, the weight values ​​of the parameters are calculated, and the size of each weight is compared, and the weight values ​​are arranged in sequence to determine the priority in the control process. The specific steps of generating the parameter weight table are as follows:

[0019] S201: extracting the influence intensity of each parameter from the parameter influence data set, determining the relative importance of each parameter in affecting the spraying quality, and generating an initial parameter effect relationship table;

[0020] S202: Based on the initial parameter-effect relationship table, assign a weight to each parameter, match the weight with the contribution of the parameter to the spraying effect, and generate adjusted parameter weight data;

[0021] S203: Integrate the adjusted parameter weight data, sort the parameter weight values ​​from high to low, determine the priority of the parameters in the spray control strategy, and generate a parameter weight table.

[0022] The present invention has been improved in that, based on the parameter weight table, according to the currently monitored spray pressure, nozzle status, and speed change information, the instantaneous fluctuation is compared with the historical records, abnormal fluctuations are identified, and the specific steps of generating the spray behavior trend analysis results are as follows:

[0023] S301: monitoring the spraying pressure, nozzle state, and speed parameter change data with high weights in the parameter weight table, and recording the real-time fluctuation of the data to generate a real-time monitoring data set;

[0024] S302: Based on the real-time monitoring data set, compare the historical data records to identify abnormal fluctuations such as pressure drop, nozzle switch instability, or speed mutation, mark the abnormal events, and generate an abnormal fluctuation mark set;

[0025] S303: Analyze the frequency and pattern of abnormal fluctuations using the abnormal fluctuation mark set, determine the impact of abnormal fluctuations on spraying effects, and generate spraying behavior trend analysis results.

[0026] The present invention has been improved in that, based on the spraying behavior trend analysis results, the time and location of future potential anomalies are predicted, and the failure mode of the spraying equipment under specified conditions is identified. The specific steps of generating the abnormal pattern recognition results are as follows:

[0027] S401: creating a data sequence based on the spraying behavior trend analysis result, marking abnormal data as a reference point for future analysis, and generating an abnormal data sequence;

[0028] S402: Classifying the fluctuation information in the abnormal data sequence, associating each type of fluctuation with the spraying problem it may cause, determining the spraying defects caused by each abnormal type, and generating an abnormal fluctuation classification result;

[0029] S403: Based on the abnormal fluctuation classification result, predict the occurrence time and location of potential abnormalities in the future, identify failure modes under various spraying conditions, and generate abnormal pattern recognition results.

[0030] The improvement of the present invention is that the acquisition camera captures the sprayed shoe upper image, and by comparing the thickness consistency of the sprayed area, the thin coating, thick coating and missed spraying areas in the image are analyzed to generate the spraying image quality assessment result. The specific steps are as follows:

[0031] S501: collecting continuous image information of the sprayed shoe upper continuously photographed by the camera to generate an original image data set;

[0032] S502: grayscale processing is performed on the original image dataset to remove background noise in the image and optimize edge details of the sprayed area in the image to generate a processed image dataset;

[0033] S503: Identify pixel density and distribution differences in the processed image data set, compare thickness consistency of the spraying area, identify thin coating, thick coating and missed spraying areas, and generate a spraying image quality assessment result.

[0034] The present invention has been improved in that, based on the spray image quality assessment results and abnormal pattern recognition results, the effects of various spray parameters on the quality of the shoe upper coating are simulated, and the speed and pressure parameters of the spray equipment are adjusted and updated according to the simulation results. The specific steps for generating the spray optimization control results are as follows:

[0035] S601: Based on the spraying image quality assessment result and the abnormal pattern recognition result, perform three-dimensional digital modeling to simulate the effects of various spraying parameters, including pressure and speed, on the spraying of the shoe upper, and generate a three-dimensional simulation effect data set;

[0036] S602: Determine the effects of various spraying speed and pressure settings on coating uniformity and coverage by evaluating the three-dimensional simulation effect data set, and obtain the optimal parameter configuration by comparing and selecting spraying parameters;

[0037] S603: Apply the optimal parameter configuration to the spraying equipment, adjust and update the parameters, so that each spraying reaches the preset quality standard, and generate a spraying optimization control result.

[0038] Intelligent shoe production line upper spraying control system, the system includes:

[0039] The data acquisition and processing module collects information on spray speed, spray volume, pressure, and coating quality, formats and cleans the collected data, analyzes the impact of parameters on coating uniformity, coverage, and drying speed, and generates a parameter impact data set;

[0040] The parameter analysis and weight configuration module calculates the weight values ​​of the parameters according to the influence size based on the parameter influence data set, arranges the weight data, and generates a parameter weight table;

[0041] The abnormality monitoring and trend analysis module monitors the real-time fluctuation of key parameters based on the parameter weight table, compares and analyzes the detected fluctuations with historical data, identifies potential abnormal fluctuation patterns, and generates spraying behavior trend analysis results;

[0042] The fault prediction and pattern recognition module builds a data model based on the spraying behavior trend analysis results to predict the time and location of potential abnormalities in the future, determines the failure mode of the spraying equipment under specified conditions, and generates abnormal pattern recognition results;

[0043] Based on the abnormal pattern recognition results, the parameter optimization and control module simulates the influence of various spraying parameters on the quality of the shoe upper coating according to the image of the sprayed shoe upper taken by the camera, determines the optimal spraying speed and pressure parameters, adjusts and updates the settings of the spraying equipment, and generates a spray optimization control result.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are:

[0045] In the present invention, by calculating the influence of each parameter on the coating quality, and determining their priority in the control process in turn, and then comparing historical records with current monitoring data, not only can abnormal fluctuations be identified, but also potential abnormal conditions and locations of key objects in the future can be predicted based on these data, and fault modes can be identified. The predictability and response speed of the system are improved. Combined with image processing technology, the coating thickness of the upper spray area is further analyzed through image comparison, which can more accurately detect thin coating, thick coating and missed spray areas, making the spraying process more stable and efficient. Finally, based on these evaluation data, the influence of multiple spraying parameters on the coating quality is simulated, so that the system can automatically adjust and optimize the spraying parameters to ensure the uniformity and consistency of the upper coating, and improve product quality and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 The present invention proposes a flow chart of the upper spraying control method of the intelligent shoe production line;

[0047] Figure 2This is a detailed flow chart of step S1 of the present invention;

[0048] Figure 3 This is a detailed flow chart of step S2 of the present invention;

[0049] Figure 4 This is a detailed flow chart of step S3 of the present invention;

[0050] Figure 5 This is a detailed flow chart of step S4 of the present invention;

[0051] Figure 6 This is a detailed flow chart of step S5 of the present invention;

[0052] Figure 7 This is a detailed flow chart of step S6 of the present invention;

[0053] Figure 8 The present invention proposes a module diagram of the upper spraying control system of an intelligent shoe production line. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction, and therefore should not be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0056] See also Figure 1 The present invention provides a technical solution: a method for controlling the spraying of shoe uppers in an intelligent shoe production line, comprising the following steps:

[0057] S1: Collect information on spraying speed, spraying volume, pressure, and coating quality. Based on the current spraying effect, analyze the impact of the parameters on coating uniformity, coverage, and drying speed, and generate a parameter impact data set:

[0058] S2: Based on the parameter impact data set, calculate the parameter weight values, compare the size of each weight, arrange the weight values ​​in order to determine the priority in the control process, and generate a parameter weight table:

[0059] S3: Based on the parameter weight table, according to the currently monitored spray pressure, nozzle status, and speed change information, the real-time fluctuation situation is compared with the historical records, abnormal fluctuations are identified, and the spray behavior trend analysis results are generated;

[0060] S4: Based on the spraying behavior trend analysis results, predict the time and location of potential abnormalities in the future, identify the failure mode of the spraying equipment under specified conditions, and generate abnormal pattern recognition results;

[0061] S5: The acquisition camera captures images of the sprayed shoe upper. By comparing the thickness consistency of the sprayed areas, the image is analyzed for thin coating, thick coating, and missed coating areas to generate the spraying image quality assessment results:

[0062] S6: Based on the spray image quality assessment results and abnormal pattern recognition results, simulate the impact of various spray parameters on the quality of the shoe upper coating, adjust and update the speed and pressure parameters of the spray equipment according to the simulation results, and generate spray optimization control results.

[0063] The parameter influence data set includes a set of standardized parameter values, an index of the influence of the parameters on spray uniformity and coverage, and an influence coefficient of the parameters on drying speed. The parameter weight table includes the weighted numerical arrangement of spray speed, spray volume, and pressure. The spray behavior trend analysis results include a pressure change curve, a nozzle switch status diagram, and a speed change trend diagram. The abnormal pattern recognition results include time warning of potential faults, location identification, and classification of spray equipment fault types. The spray image quality assessment results include a coating thickness distribution diagram, regional uniformity evaluation, coordinates of the missed spray area, and the ratio data of thin and thick coating areas. The spray optimization control results include the adjusted spray speed range, optimal pressure setting value, and spray angle configuration.

[0064] See also Figure 2 , collect spraying speed, spraying amount, pressure and coating quality information, and analyze the impact of parameters on coating uniformity, coverage and drying speed based on the current spraying effect. The specific steps to generate parameter impact data set are as follows:

[0065] S101: Collect spray speed, spray volume, pressure and coating quality data, eliminate erroneous readings and outliers, verify data integrity, and generate raw data sets;

[0066] During the collection process, the original collection of spray speed, spray volume, pressure and coating quality data is often accompanied by noise and errors. To ensure data quality, data cleaning technology is first applied to eliminate obvious erroneous readings and outliers. This includes using statistical methods to identify and process outliers, such as using box plots to identify and eliminate data points that are outside the quartile range. Next, data integrity checks are used to check whether each necessary field has a value or whether the field data type conforms to the predefined format, thereby effectively cleaning and verifying the data to ensure the accuracy and reliability of subsequent analysis.

[0067] S102: Based on the original data set, adjust the data to the same scale, eliminate the differences between measurement units, and generate a processed data set;

[0068] After obtaining the raw data set that has been preliminarily cleaned, in order to eliminate the scale differences between the spray speed, spray amount, pressure and coating quality data, all data need to be adjusted to the same scale. The Z-score normalization method can be selected, that is, each eigenvalue of the data is subtracted from its mean and divided by the standard deviation, so that the processed data set has zero mean and unit variance. In addition, by converting the data between different measurement units, such as converting the speed unit from m / s to cm / s, ensure that all data are compared and analyzed under the same standard.

[0069] S103: Based on the processed data set, according to the current coating uniformity, coverage and drying speed data, the correlation between each parameter and the spraying effect is analyzed, the influence of the parameter on the coating characteristics is calculated, and the parameter influence data set is generated;

[0070] According to the processed data set, the correlation between each parameter and the spraying effect is analyzed according to the formula: , calculate the influence of each parameter on the coating characteristics. Represents the regression coefficient, indicating the parameter and spraying effect The strength and direction of the linear relationship between Representative Observed parameter values, such as spraying speed, Representative Evaluation index values ​​of spraying effects, such as coating uniformity, and All and The average of the observed values. These values ​​are obtained through data collection and then preprocessed to ensure accuracy and consistency.

[0071] Taking the effect of spraying speed on coating uniformity as an example, the spraying speed values ​​collected from the data set are m / s, the corresponding coating uniformity score is .

[0072] First calculate and The mean m / s, Then, each observation is substituted into the formula to calculate the difference :

[0073] calculate :

[0074]

[0075] calculate :

[0076]

[0077] result:

[0078] result It shows that for every unit increase in spray speed, the coating uniformity score will increase by about The calculated regression coefficient It provides a quantitative indicator to evaluate the linear relationship between parameters and spraying effect. For example, in the example of spraying speed and coating uniformity, The value of 0.092 indicates that the spraying speed is positively correlated with the coating uniformity, that is, as the spraying speed increases, the coating uniformity will also improve. The value can be used to determine which parameters are the key factors affecting the spraying effect. If the value is large, it means that it has a significant impact on the spraying effect, and this parameter should be adjusted first to obtain the best spraying quality. On the contrary, if a parameter is If the value is close to zero or very small, it means that it has little impact on the spraying effect, and you can consider focusing your adjustments on more critical parameters. In addition, by comparing the parameter settings in actual spraying operations with the regression analysis results, you can identify possible deviations and deficiencies. For example, if the spraying speed is low in actual operation, but the regression analysis shows that increasing the spraying speed can significantly improve coating uniformity, you can recommend adjusting the spraying speed setting range to achieve higher quality production results.

[0079] See also Figure 3, based on the parameter impact data set, calculate the parameter weight value, compare the size of each weight, and arrange the weight values ​​in sequence to determine the priority in the control process. The specific steps for generating the parameter weight table are as follows:

[0080] S201: extracting the influence intensity of each parameter from the parameter influence data set, determining the relative importance of each parameter in affecting the spraying quality, and generating an initial parameter effect relationship table;

[0081] First, the influence intensity of each parameter is extracted. Data analysis technology is needed to quantitatively analyze the relationship between each parameter and spraying quality. The information entropy of each parameter is calculated to evaluate its influence on spraying quality. Information entropy, as an indicator of data variability, can effectively represent the uncertainty and influence of parameters. The influence intensity of each parameter is determined by the size of its variability in the data set. A higher information entropy means that the parameter has a greater impact on the result during the spraying process. Therefore, the relative importance is determined. These steps are implemented through statistical software and data processing algorithms, such as using Python's Pandas library to obtain the standard deviation and coefficient of variation, and listing each parameter and its influence intensity on the spraying quality.

[0082] S202: Based on the initial parameter effect relationship table, weights are assigned to each parameter, and the weights are matched with the contribution of the parameters to the spraying effect to generate adjusted parameter weight data;

[0083] For weight distribution, according to the formula , calculate the weight of each parameter, where, Representation parameters The weight is used to indicate the relative importance of the parameter in the spraying effect. Representation parameters The information entropy is used to quantify the uncertainty and variability of the parameter during the spraying process. A high information entropy indicates that the change of the parameter during the spraying process has a greater impact on the final effect. is information entropy The natural exponent is used to amplify the information entropy value and make the weight distribution more sensitive to high entropy parameters. Represents all parameters, information entropy The calculation is based on the statistical data analysis of the relationship between each parameter and the spraying quality, and is obtained from the historical data according to the discreteness of the influence of each parameter on the spraying quality.

[0084] If there are four parameters, the information entropy values ​​are .

[0085] Calculate the exponential value of each information entropy:

[0086] Compute the sum of these exponential values:

[0087] Calculate the weight of each parameter:

[0088]

[0089]

[0090]

[0091] The weighting results indicate that the second parameter has the highest information entropy and the greatest impact on spraying performance, making it a priority when developing spraying control strategies. This is consistent with historical data showing that this parameter frequently fluctuates and is highly correlated with spraying quality. This approach ensures that spraying control strategies are more focused on key parameters, optimizing resource allocation and operational efficiency.

[0092] S203: Integrate the adjusted parameter weight data, sort the parameter weight values ​​from high to low, determine the priority of the parameters in the spray control strategy, and generate a parameter weight table;

[0093] After integrating the adjusted parameter weight data, the parameter weight values ​​need to be sorted from high to low. Use data sorting algorithms, such as quick sort or merge sort, to ensure that the parameters are correctly arranged according to their weights. During the sorting process, parameters with high weights are placed at the front of the table, indicating that they have a higher priority in spray control.

[0094] See also Figure 4 Based on the parameter weight table, according to the currently monitored spray pressure, nozzle status, and speed change information, the instantaneous fluctuation is compared with the historical records, abnormal fluctuations are identified, and the specific steps for generating the spray behavior trend analysis results are as follows:

[0095] S301: monitoring the spraying pressure, nozzle status, and speed parameter change data with high weights in the parameter weight table, and recording the real-time fluctuation of the data to generate a real-time monitoring data set;

[0096] High-weight parameters such as spray pressure, nozzle status, and speed changes are key monitoring points. The system monitors the changes in these parameters in real time to capture immediate fluctuations in the data. In particular, the monitoring equipment records the real-time readings of each parameter. The system uses a sliding window averaging algorithm to smooth data fluctuations. The sliding window algorithm eliminates errors caused by short-term fluctuations by smoothing the data within a continuous time period. The length of the sliding window is determined based on historical data and is usually set to the number of data within a certain time window (such as 5-second or 10-second readings). The system continuously calculates new window values ​​and updates the smoothed monitoring data in real time to ensure data accuracy and reliability, generating a real-time monitoring data set. This data set provides the basis for subsequent analysis, allowing the system to respond more quickly to possible equipment failures or process deviations.

[0097] S302: Based on the real-time monitoring data set, compare the historical data records to identify abnormal fluctuations such as pressure drop, nozzle switch instability, or speed mutation, mark the abnormal events, and generate an abnormal fluctuation mark set;

[0098] By comparing historical data, threshold detection algorithms and autoregressive moving average models are used to identify abnormal fluctuations such as sudden pressure drops, unstable nozzle switches, or sudden speed changes. First, through the threshold detection algorithm, when the spraying pressure or speed exceeds the set normal range (for example, the speed exceeds 3 standard deviations or the pressure drops by more than 10%), the system will mark the abnormal event. Next, the system analyzes the time series trend of the data through the ARMA model. The ARMA model predicts normal parameter changes by combining the autoregressive term (past observations) and the moving average term (prediction error term). If the detected change value deviates significantly from the predicted value of the ARMA model, the point is identified as an abnormal event. Threshold detection is used to capture instantaneous fluctuations, while the ARMA model is used to capture more complex abnormal fluctuation behaviors in time series.

[0099] S303: Analyze the frequency and pattern of abnormal fluctuations using the abnormal fluctuation mark set, determine the impact of abnormal fluctuations on spraying effects, and generate spraying behavior trend analysis results;

[0100] Cluster analysis and anomaly detection algorithms are used to analyze the frequency and patterns of fluctuations. K-means clustering is used to classify normal operating conditions from abnormal fluctuations. The K-means algorithm divides data points into several categories based on fluctuation characteristics (such as amplitude and duration), helping to identify typical abnormal patterns. Simultaneously, an anomaly detection algorithm is used to further detect outliers. By randomly selecting features and recursively segmenting the dataset based on feature thresholds, outliers are isolated along shorter paths and marked as outliers. By combining these two algorithms, the system can accurately identify common abnormal fluctuation patterns and analyze the impact of their frequency on spraying results.

[0101] See also Figure 5 ,Based on the results of spraying behavior trend analysis, the time and location of potential anomalies in the future are predicted, and the failure mode of the spraying equipment under specified conditions is identified. The specific steps for generating abnormal pattern recognition results are as follows:

[0102] S401: Based on the spraying behavior trend analysis results, a data sequence is created, abnormal data is marked as a reference point for future analysis, and an abnormal data sequence is generated;

[0103] By processing all collected abnormal fluctuation data, we first create a time series dataset and arrange these fluctuation events in chronological order so that the order and duration of these events can be quickly determined in subsequent analysis. In addition, by marking and classifying abnormal events, we can better define future benchmarks, which become the basis for identifying similar fluctuation behaviors in the future. The generated abnormal data sequence contains different types of fluctuations and corresponding time information. The system marks the abnormal data as a reference for future analysis to help improve the accuracy of the prediction and provide a basis for the handling of anomalies, ensuring that these abnormal events can be quickly responded to in future monitoring.

[0104] S402: Classify the fluctuation information in the abnormal data sequence, associate each type of fluctuation with the spraying problem it may cause, determine the spraying defects caused by each abnormal type, and generate an abnormal fluctuation classification result;

[0105] The classification of fluctuation information in abnormal data sequences is based on the identification of each abnormal type and the correlation analysis of the spraying defects they may cause. The system adopts a decision tree classification algorithm. By training historical data, it learns the characteristics of each abnormal fluctuation, such as pressure fluctuation, speed mutation, etc., and how they are associated with spraying defects such as uneven coating, spray leakage, overspray, etc. During the classification process, the system will automatically assign them to preset categories according to different fluctuation amplitudes, durations and frequencies of abnormal occurrence. The system quickly determines the spraying problem corresponding to each type of fluctuation through the branch structure of the decision tree. The generated abnormal fluctuation classification results clearly associate each fluctuation with the spraying defect, thereby providing a reference for subsequent prediction and problem handling.

[0106] S403: Based on the abnormal fluctuation classification results, predict the occurrence time and location of potential abnormalities in the future, identify failure modes under various spraying conditions, and generate abnormal pattern recognition results;

[0107] For identification of failure modes under various spraying conditions, the formula is used:

[0108]

[0109] Calculating future time The probability of abnormal fluctuations occurring ;

[0110] in, It is the time interval between two consecutive time steps, usually determined by the sampling frequency of the monitoring equipment. This probability value is calculated by using the current parameters and their changing trends to estimate future abnormal risks, and the value range is between 0 and 1. Represents the current moment The probability of abnormal occurrence under the condition is obtained through historical data and current parameter status statistics, usually ranging from 0 to 1. Is an influencing factor The weight coefficient is determined by regression analysis of historical data or machine learning methods. is the current time Next The instantaneous value of each parameter (such as spraying pressure, nozzle status, speed) is obtained by the system monitoring equipment. It is the total number of parameters, which means the number of all relevant parameters (such as spraying pressure, nozzle status, speed, etc.) monitored in the system, obtained through system monitoring. Indicates the The increment of the parameter, that is, from the previous time step to the current time The change in , used to capture short-term fluctuations, is the weight of the incremental term, which is determined by analyzing the correlation between historical abnormal fluctuations and parameter changes, and adjusting the impact of short-term changes in parameters on the probability of abnormal occurrence. represents the rate of change of the parameter, It is The time derivative of a parameter indicates the rate of change of the parameter over time and is calculated by the finite difference method or automatic differentiation method. is the weight of the rate term, which captures the impact of long-term trend changes in parameters on future anomalies and is determined through historical data analysis.

[0111] The weight coefficient obtained from historical data analysis is: , , , .

[0112] Current time point Spraying pressure , nozzle status , whose increment is and , rate of change and If the current abnormality probability .

[0113] Substituting the values ​​into the formula:

[0114]

[0115]

[0116]

[0117] Since the probability cannot exceed 1, the system corrects it to 1. Indicates the future time The probability of abnormal fluctuations is very high, approaching 100%. This prediction allows the system to identify potential fault locations in advance, helping operators perform preventive maintenance and providing an effective early warning mechanism for potential failures in the production process. If the average probability of abnormal fluctuations under similar conditions in the past was 0.8, this predicted value indicates a significantly increased likelihood of fluctuations, necessitating more proactive intervention measures such as preventive maintenance and parameter adjustments.

[0118] See also Figure 6 The acquisition camera captures the sprayed shoe upper image, and by comparing the thickness consistency of the sprayed area, analyzes the thin coating, thick coating and missed spraying areas in the image, and generates the spraying image quality assessment results. The specific steps are as follows:

[0119] S501: collecting continuous image information of the sprayed shoe upper continuously photographed by the camera to generate an original image data set;

[0120] The camera continuously captures the sprayed shoe upper to obtain continuous image information, which is then converted into a raw image dataset. The collected image data includes every step and angle of the spraying process. The image acquisition device in the system uses a high-resolution industrial camera to ensure that every detail is captured. The data acquisition frequency is adjusted according to the spraying speed, usually multiple frames per second. The system stores these images in chronological order and adds spraying parameters such as pressure, speed and temperature. The image dataset includes a complete timestamp and parameter record, providing complete data for subsequent image processing and analysis, ensuring the traceability of the spraying process.

[0121] S502: grayscale processing is performed on the original image dataset to remove background noise in the image and optimize edge details of the sprayed area in the image to generate a processed image dataset;

[0122] First, the image is grayscaled. An image preprocessing algorithm is used to convert the color image into a grayscale image, retaining the important details of the sprayed area. Then, a Gaussian filter algorithm is used to remove background noise and subtle interference in the image. The convolution kernel operation of the algorithm is used to smooth the noise points in the image. Subsequently, the Canny edge detection algorithm is used to detect and optimize the edges of the sprayed area, enhancing the edge details of the sprayed area and ensuring that the outline and boundary of the sprayed area can be accurately identified in subsequent analysis. This dataset provides an optimized image foundation for subsequent quality assessment and detection.

[0123] S503: Identify pixel density and distribution differences in the processed image data set, compare thickness consistency of the spraying area, identify thin coating, thick coating and missed coating areas, and generate a spraying image quality assessment result;

[0124] Identify differences in pixel density and distribution in the processed image dataset, according to the formula: Calculate the thickness consistency of the sprayed area, where Indicates the average value of pixel density differences within the spraying area, which is used to measure the consistency of spraying thickness. The smaller the value, the more uniform the spraying. is the total number of pixels in the spray area, which is automatically determined by the system's image processing and spray area segmentation algorithm. It is The brightness value (grayscale value) of each pixel is extracted through image processing technology. The brightness value represents the relative density of the spray thickness. The larger the value, the greater the relative thickness. It is the average brightness value of all pixels in the spraying area, which is used to represent the average thickness of the area. The calculation formula is: , the overall thickness of the spraying is estimated by the average brightness value of all pixels, :Indicates the The brightness value of each pixel and the average brightness value of the area The absolute difference is used to indicate the deviation of the pixel relative to the overall thickness.

[0125] Assume that there are 5 pixels in the spraying area. The brightness values ​​of these pixels obtained through image processing are: Each brightness value Indicates the spray thickness of the corresponding pixel.

[0126] Calculate average brightness

[0127]

[0128]

[0129] Average brightness value , indicating the standard thickness of the spraying area.

[0130] Using the formula , calculate the difference between each pixel and the mean:

[0131]

[0132]

[0133] According to the formula , sum these differences and calculate the average:

[0134]

[0135] Calculated spray thickness difference , indicating that there are certain differences in the spraying thickness in this area, and the differences are large, and there may be thick or thin coating phenomena.

[0136] Thin coating area: pixel value is The number of pixels is significantly lower than the average , indicating that the pixel may correspond to a thin coating area.

[0137] Thick Paint Area: Pixel value is The number of pixels is significantly higher than the average , indicating that the pixel corresponds to the thick coating area.

[0138] Missing area: If the value of some pixels is close to zero (such as or ), it could be identified as a missed area, but in this example there is no missed area.

[0139] See also Figure 7 Based on the spray image quality assessment results and abnormal pattern recognition results, the effects of various spray parameters on the quality of the shoe upper coating are simulated. The speed and pressure parameters of the spray equipment are adjusted and updated according to the simulation results. The specific steps to generate the spray optimization control results are as follows:

[0140] S601: Based on the spraying image quality assessment results and abnormal pattern recognition results, perform three-dimensional digital modeling to simulate the effects of various spraying parameters, including pressure and speed, on the sprayed shoe upper, and generate a three-dimensional simulation effect data set;

[0141] A three-dimensional model of the shoe upper is generated through three-dimensional digital modeling software. Three-dimensional modeling requires the input of the geometric size and shape data of the shoe upper, which is obtained through a laser scanner to obtain a high-precision geometric mesh model. On this basis, the system will simulate the impact of different pressure, speed and other parameters on the shoe upper coating during the spraying process. Specific parameters such as spraying pressure are collected by sensors on the equipment and transmitted to the system for simulation. The spraying speed is determined by the nozzle movement speed and is also used as an input parameter for the simulation. Based on these parameters, the system will perform simulation calculations in three-dimensional space on the spraying thickness, coating uniformity and shoe upper coverage. A three-dimensional simulation effect data set is generated through different parameter combinations. This data set records the changes in coating thickness, surface flatness, coating adhesion and coverage distribution under different conditions.

[0142] S602: Determine the effects of various spraying speed and pressure settings on coating uniformity and coverage by evaluating the three-dimensional simulation effect data set, and select spraying parameters by comparison to obtain the optimal parameter configuration;

[0143] First, multiple combinations of spray speed and pressure are selected and simulated in sequence. The spray speed is controlled by the movement speed of the nozzle, and the pressure is detected by the sensor of the equipment. A set of simulation data is generated for each combination. These data will record the thickness, uniformity and coverage effect of the coating. The thickness uniformity of the coating can be obtained through statistical methods such as calculating the standard deviation. In addition, the coverage effect can be determined by analyzing the distribution range of the coating and the adhesion of the coating in different areas. The system compares each set of data to determine which combination of spray speed and pressure performs best in uniformity and coverage, and selects the parameters with the smallest standard deviation and the widest coverage as the optimal parameter configuration.

[0144] S603: Apply the optimal parameter configuration to the spraying equipment, adjust and update the parameters so that each spraying reaches the preset quality standard, and generate the spraying optimization control result;

[0145] The parameters are uploaded to the control system of the spraying equipment, and the spraying equipment will automatically adjust the spraying speed and pressure to meet the preset optimal parameters. The adjusted parameters such as spraying speed will control the movement speed of the nozzle through the stepper motor, and the spraying pressure will be regulated in real time through the air pressure regulating device. These parameters are continuously monitored to ensure that various indicators in the spraying process remain within the optimal range. If a parameter deviation is detected during the spraying process, such as a slowdown in the spraying speed or excessive pressure fluctuations, the system will immediately make adjustments through the feedback mechanism to ensure that the coating thickness and coverage effect meet the preset quality standards, and finally record the results of the spray optimization control.

[0146] See also Figure 8 , intelligent shoe production line upper spraying control system, the system includes:

[0147] The data acquisition and processing module collects information on spray speed, spray volume, pressure, and coating quality, formats and cleans the collected data, analyzes the impact of parameters on coating uniformity, coverage, and drying speed, and generates a parameter impact data set;

[0148] The parameter analysis and weight configuration module calculates the weight values ​​of the parameters based on the parameter influence data set, arranges the weight data, and generates a parameter weight table;

[0149] The abnormality monitoring and trend analysis module monitors the real-time fluctuation of key parameters based on the parameter weight table, compares and analyzes the detected fluctuations with historical data, identifies potential abnormal fluctuation patterns, and generates spray behavior trend analysis results;

[0150] The fault prediction and pattern recognition module builds a data model based on the spraying behavior trend analysis results to predict the time and location of potential abnormalities in the future, determine the failure mode of the spraying equipment under specified conditions, and generate abnormal pattern recognition results;

[0151] Based on the abnormal pattern recognition results, the parameter optimization and control module simulates the impact of various spraying parameters on the quality of the shoe upper coating according to the camera's image of the sprayed shoe upper, determines the optimal spraying speed and pressure parameters, adjusts and updates the settings of the spraying equipment, and generates spray optimization control results.

[0152] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. Intelligent shoe production line upper spraying control method, characterized in that: The following steps are involved: Collect information on spraying speed, spraying volume, pressure and coating quality. According to the current spraying effect, analyze the influence of parameters on coating uniformity, coverage and drying speed, and generate parameter influence data set: Based on the parameter impact data set, the weight values ​​of the parameters are calculated, and the size of each weight is compared, and the weight values ​​are arranged in sequence to determine the priority in the control process. The specific steps for generating the parameter weight table are as follows: Extracting the influence intensity of each parameter from the parameter influence data set, determining the relative importance of each parameter in influencing the spraying quality, and generating an initial parameter effect relationship table; Based on the initial parameter effect relationship table, weights are assigned to each parameter, and the weights are matched with the contribution of the parameters to the spraying effect to generate adjusted parameter weight data; Integrate the adjusted parameter weight data, sort the parameter weight values ​​from high to low, determine the priority of the parameters in the spray control strategy, and generate a parameter weight table: Based on the parameter weight table, according to the currently monitored spraying pressure, nozzle status, and speed change information, the instantaneous fluctuation situation is compared with the historical records, abnormal fluctuations are identified, and the spraying behavior trend analysis results are generated; According to the spraying behavior trend analysis results, the specific steps of predicting the occurrence time and location of potential abnormalities in the future, identifying the failure mode of the spraying equipment under specified conditions, and generating abnormal pattern recognition results are as follows: Based on the spraying behavior trend analysis results, a data sequence is created, abnormal data is marked as a reference point for future analysis, and an abnormal data sequence is generated; Classifying the fluctuation information in the abnormal data sequence, associating each type of fluctuation with the spraying problem it may cause, determining the spraying defects caused by each abnormal type, and generating abnormal fluctuation classification results; Based on the abnormal fluctuation classification results, predict the occurrence time and location of potential abnormalities in the future, identify failure modes under various spraying conditions, and generate abnormal pattern recognition results; For identification of failure modes under various spraying conditions, the formula is used: ; Calculating future time The probability of abnormal fluctuations ; in, is the time interval between two consecutive time steps, Represents the current moment The probability of abnormal occurrence under Is an influencing factor The weight coefficient is the current time Next The instantaneous value of a parameter, is the total number of parameters, Indicates The increment of the parameter, is the weight of the incremental term, It is The time derivative of a parameter, is the weight of the rate term; The acquisition camera takes images of the sprayed shoe uppers, compares the thickness consistency of the sprayed areas, analyzes the thin coating, thick coating and missed spraying areas in the image, and generates spraying image quality assessment results: Based on the spraying image quality evaluation results and abnormal pattern recognition results, the specific steps of simulating the influence of various spraying parameters on the quality of the upper coating, adjusting and updating the speed and pressure parameters of the spraying equipment according to the simulation results, and generating the spraying optimization control results are as follows: Based on the spraying image quality assessment result and the abnormal pattern recognition result, three-dimensional digital modeling is performed to simulate the effects of various spraying parameters, including pressure and speed, on the spraying shoe upper, and a three-dimensional simulation effect data set is generated; By evaluating the three-dimensional simulation effect data set, determining the effects of various spraying speed and pressure settings on coating uniformity and coverage, and obtaining the optimal parameter configuration by comparing and selecting spraying parameters; The optimal parameter configuration is applied to the spraying equipment to adjust and update the parameters so that each spraying can meet the preset quality standards and generate a spraying optimization control result.

2. The method for controlling the spraying of the upper of the intelligent shoe production line according to claim 1 is characterized in that: The parameter influence data set includes a set of standardized parameter values, an index of the influence of the parameters on spraying uniformity and coverage, and an influence coefficient of the parameters on drying speed. The parameter weight table includes a weighted numerical arrangement of spraying speed, spraying amount and pressure. The spraying behavior trend analysis results include a pressure change curve, a nozzle switch status diagram and a speed change trend diagram. The abnormal pattern recognition results include a time warning of potential faults, a location identification, and a classification of spraying equipment fault types. The spraying image quality assessment results include a coating thickness distribution diagram, a regional uniformity evaluation, the coordinates of the missed spraying area, and the proportion data of thin and thick coating areas. The spraying optimization control results include an adjusted spraying speed range, an optimal pressure setting value, and a spraying angle configuration.

3. The method for controlling the spraying of the upper of the intelligent shoe production line according to claim 1 is characterized in that: Collect information on spraying speed, spraying volume, pressure and coating quality. According to the current spraying effect, analyze the influence of parameters on coating uniformity, coverage and drying speed. The specific steps to generate parameter influence data set are as follows: Collect spray speed, spray volume, pressure and coating quality data, eliminate false readings and outliers, verify data integrity, and generate raw data sets; Based on the original data set, adjusting the data to the same scale, eliminating the difference between measurement units, and generating a processed data set; Based on the processed data set, according to the current coating uniformity, coverage and drying speed data, the correlation between each parameter and the spraying effect is analyzed, the influence of the parameter on the coating characteristics is calculated, and the parameter influence data set is generated.

4. The method for controlling the spraying of the upper of the intelligent shoe production line according to claim 1 is characterized in that: Based on the parameter weight table, according to the currently monitored spraying pressure, nozzle status, speed change information, the instant fluctuation situation is compared with the historical records, abnormal fluctuations are identified, and the specific steps of generating the spraying behavior trend analysis results are as follows: Monitor the spraying pressure, nozzle state, and speed parameter change data with high weights in the parameter weight table, and record the real-time fluctuation of the data to generate a real-time monitoring data set; Based on the real-time monitoring data set, compare the historical data records, identify abnormal fluctuations such as pressure drop, nozzle switch instability or speed mutation, mark the abnormal events, and generate an abnormal fluctuation mark set; The abnormal fluctuation mark set is used to analyze the frequency and pattern of the abnormal fluctuation, determine the impact of the abnormal fluctuation on the spraying effect, and generate a spraying behavior trend analysis result.

5. The method for controlling the spraying of the upper of the intelligent shoe production line according to claim 1 is characterized in that: The acquisition camera takes the sprayed shoe upper image, compares the thickness consistency of the sprayed area, analyzes the thin coating, thick coating and missed spraying areas in the image, and generates the spraying image quality assessment results in the following specific steps: Collect continuous image information of the sprayed shoe upper continuously shot by the camera to generate a raw image data set; grayscale the original image data set, remove background noise in the image, optimize edge details of the sprayed area in the image, and generate a processed image data set; Identify the pixel density and distribution differences in the processed image data set, compare the thickness consistency of the spraying area, identify the thin coating, thick coating and missed spraying areas, and generate a spraying image quality assessment result.

6. Intelligent shoe production line upper spraying control system, characterized in that: According to the intelligent shoe production line upper spraying control method according to any one of claims 1 to 5, the system comprises: The data acquisition and processing module collects information on spraying speed, spraying volume, pressure and coating quality, formats and cleans the collected data, analyzes the impact of parameters on coating uniformity, coverage and drying speed, and generates parameter impact data sets; The parameter analysis and weight configuration module calculates the weight values ​​of the parameters based on the parameter influence data set according to the influence size, arranges the weight data, and generates a parameter weight table; The abnormal monitoring and trend analysis module monitors the real-time fluctuation of key parameters based on the parameter weight table, compares and analyzes the detected fluctuations with historical data, identifies potential abnormal fluctuation patterns, and generates spraying behavior trend analysis results; The fault prediction and pattern recognition module builds a data model based on the spraying behavior trend analysis results to predict the time and location of potential abnormalities in the future, determines the fault mode of the spraying equipment under specified conditions, and generates abnormal pattern recognition results; Based on the abnormal pattern recognition results, the parameter optimization and control module simulates the influence of various spraying parameters on the quality of the upper coating according to the image of the sprayed upper taken by the camera, determines the optimal spraying speed and pressure parameters, adjusts and updates the settings of the spraying equipment, and generates a spraying optimization control result.

Images