Product quality supervision system for leather shoe processing

Abstract

The invention relates to the technical field of quality data monitoring, in particular to a product quality supervision system for leather shoe processing, which comprises a monitoring data analysis module for acquiring multiple data such as running speed, pressure and temperature of a forming machine from a leather shoe production line, analyzing a logarithmic relationship between the number of closed boxes at data points and the size by using a box counting method, the complexity of data change in the leather shoe forming process is calculated, and the stability index of the leather shoe forming process is generated. According to the method, the logarithmic relationship between the number and the size of the closed boxes of key parameters on the leather shoe production line is analyzed through a box counting method, so that the system can perform quantitative analysis on the complexity of data change, and potential production fluctuation and quality deviation are effectively identified. Through analysis of fluctuation parameters and identification of a periodic abnormal mode, the precision and the reaction speed of production process control are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of quality data monitoring, and in particular to a product quality monitoring system for leather shoe processing. Background Art

[0002] The field of quality data monitoring technology collects and analyzes data generated during the manufacturing process in real time to ensure that product quality meets preset standards. Use real-time monitoring and automated analysis tools to ensure that product quality meets preset standards and specifications. Continuous monitoring of data on the production line can help identify anomalies in the production process, thereby achieving timely intervention and quality control.

[0003] Among them, the product quality supervision system for leather shoe processing integrates a variety of sensors and monitoring tools to track and record key quality parameters in the leather shoe processing process in real time. By continuously monitoring and analyzing the data on the production line, the system can promptly detect any quality problems that may arise in the production process and provide necessary feedback to correct these problems. The purpose of the system is to ensure that leather shoes meet all quality requirements before leaving the production line and entering the market.

[0004] Although existing technologies integrate a variety of sensors and monitoring tools to track and record key quality parameters in the leather shoe processing process, they mainly focus on data collection and feedback. When faced with data fluctuations and abnormal patterns, the system is difficult to promptly identify and prevent production problems caused by parameter fluctuations, resulting in reduced production efficiency and increased scrap rates. In addition, production lines tend to be slow to respond to repetitive quality problems, affecting overall production quality and cost control. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a product quality supervision system for leather shoe processing.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A product quality supervision system for leather shoe processing comprises:

[0007] The monitoring data analysis module obtains multiple data such as the operating speed, pressure and temperature of the molding machine from the leather shoe production line, uses the box counting method to analyze the logarithmic relationship between the number and size of the closed boxes of the data points, and calculates the complexity of the data changes in the leather shoe molding process to generate the stability index of the leather shoe molding process;

[0008] The quality stability monitoring module uses the stability index of the leather shoe forming process to perform trend analysis, identify periodic abnormal patterns in the operating parameter data of the forming machine, and analyze the degree of deviation between the abnormal patterns and the quality standards of the leather shoe forming process, evaluate the quality stability, and generate leather shoe quality information;

[0009] The quality fluctuation prediction module uses the Bayesian network to analyze and calculate the quality fluctuation in the future leather shoe molding process based on the leather shoe quality information and historical quality deviation data, identifies the quality parameters that have the greatest impact on the molding process, and obtains the deviation analysis results;

[0010] The quality optimization response module adjusts the operating parameters of the molding machine in real time according to the deviation analysis results, and the operating parameters include temperature, pressure or speed to obtain the leather shoe molding quality optimization results.

[0011] As a further solution of the present invention, the step of obtaining the logarithmic relationship between the number and size of the closed boxes of the analysis data points is specifically as follows:

[0012] Collect the data of running speed V, pressure P and temperature T from the molding machine monitoring system, using the formula:

[0013]

[0014] Calculate the number of boxes N required for the data points;

[0015] Among them, max() is used to return the maximum value in a given data set, min() is used to return the minimum value in a given data set, δV, δP, δT are the box sizes of velocity, pressure, and temperature dimensions;

[0016] Based on the number of boxes N required for the data points, the formula is used:

[0017]

[0018] Calculate the average size S of each box;

[0019] Among them, σ V , σ P , σ T is the standard deviation of velocity, pressure, and temperature data;

[0020] Based on the number of boxes N required for the data points and the average size S of each box, the formula is used:

[0021]

[0022] Calculate the ratio R of the number of enclosed boxes of data points to the logarithm of their size;

[0023] Among them, log() is the natural logarithm function.

[0024] As a further solution of the present invention, the steps for obtaining the stability index of the leather shoe forming process are specifically as follows:

[0025] According to the ratio R of the number of enclosed boxes of the data point to the logarithm of the size, the formula is adopted:

[0026]

[0027] Calculate the complexity D of data changes;

[0028] Among them, C1 represents the constant coefficient used for adjustment, α1, β1, and γ1 represent the normalized weight coefficients of the velocity, pressure, and temperature data ranges, respectively;

[0029] According to the complexity D of the data change, the formula is adopted:

[0030]

[0031] Calculate the impact factor G;

[0032] Among them, V var1 , P var1 、T var1 Represent the fluctuation range of speed, pressure and temperature respectively;

[0033] The impact factor G is compared with the industry standard threshold θ1 using the formula:

[0034] Idx=|G-Θ1|

[0035] Determine the stability of the forming process and obtain the stability index Idx of the leather shoe forming process.

[0036] As a further solution of the present invention, the step of obtaining the periodic abnormal pattern in the operating parameter data of the identifying molding machine is specifically as follows:

[0037] According to the stability index Idx of the leather shoe forming process, the formula is adopted:

[0038]

[0039] By setting thresholds to determine the changes that are considered abnormal at various levels, the initial abnormal analysis indicator M1 is obtained;

[0040] Where, e is the base of natural logarithm;

[0041] According to the initial abnormal analysis indicator M1, the formula is adopted:

[0042]

[0043] Calculate the sum of the deviations between M1 and its time series average to obtain the cyclical abnormal fluctuation index M2;

[0044] Among them, M 1i represents the M1 value at the i-th time point, n is the total number of data points in the time series, M 1jrepresents the M1 value at the jth time point;

[0045] According to the periodic abnormal fluctuation index M2, the threshold θ2 is applied again to determine whether there is a periodic abnormality, using the formula:

[0046] Q=M2-θ2

[0047] Calculate the abnormal pattern recognition result Q;

[0048] If Q is greater than zero, it indicates an abnormal pattern, otherwise it is a normal periodic change.

[0049] As a further solution of the present invention, the steps of obtaining the leather shoe quality information are specifically as follows:

[0050] According to the abnormal pattern recognition result Q, the quality standard threshold is set to θ3, using the formula:

[0051] O=|Q-Θ3|

[0052] Calculate the deviation between Q and the quality standard threshold Θ3 to obtain the degree of deviation O;

[0053] According to the degree of deviation O, the formula is used:

[0054]

[0055] Calculate the quality stability index L and obtain the quality information of leather shoes;

[0056] Among them, k1 is an empirical parameter used to adjust the sensitivity of S, and μ1 is another empirical parameter used to adjust the reference point.

[0057] As a further solution of the present invention, the step of obtaining the quality fluctuation in the future leather shoe forming process is specifically as follows:

[0058] According to the quality stability index L and the historical quality deviation data H, the formula is adopted:

[0059]

[0060] Initialize the node value B of the Bayesian network;

[0061] Among them, α2 is a weight parameter used to adjust the influence ratio of current data and historical data in quality prediction, and c2 is a constant used to adjust the influence of historical data;

[0062] Using the node value B of the initialized Bayesian network, the formula is:

[0063]

[0064] Calculate the estimated value W of future quality fluctuations;

[0065] Among them, β2 is the adjustment coefficient, which is used to adjust the influence of network output on prediction, γ2 is the parameter for adjusting the smoothness of network output, and δ2 is used to adjust the sensitivity of network output to the node value B of the Bayesian network.

[0066] As a further solution of the present invention, the step of obtaining the deviation analysis result is specifically:

[0067] According to the estimated value W of future quality fluctuation, the formula is adopted:

[0068]

[0069] Calculate the total impact U of each quality parameter on future quality fluctuations;

[0070] Among them, f k is the historical impact record of the kth quality parameter, σ k is the uncertainty coefficient associated with the kth quality parameter, g is the total number of parameters;

[0071] According to the total influence U of each quality parameter on future quality fluctuations, the formula is adopted:

[0072]

[0073] Calculate the quality parameter value Z that has the greatest impact on the molding process and obtain the deviation analysis results;

[0074] Among them, U k is the deviation value of the kth quality parameter, r k is the correlation factor used to weight the influence of each parameter.

[0075] As a further solution of the present invention, the steps for obtaining the molding quality optimization result are specifically as follows:

[0076] According to the quality parameter value Z which has the greatest impact on the molding process, the formula is adopted:

[0077]

[0078] Calculate the adjustment factor A;

[0079] Among them, τ3 is the acceptable threshold of mass deviation;

[0080] According to the adjustment coefficient A, the formula is adopted:

[0081] T′=T×A

[0082] and

[0083] P′=P×A

[0084] and

[0085] V′=V×A

[0086] The adjusted temperature T′, pressure P′ and speed V′ are calculated to obtain the molding quality optimization result.

[0087] Compared with the prior art, the advantages and positive effects of the present invention are:

[0088] In the present invention, the box counting method is used to analyze the logarithmic relationship between the number of closed boxes and the size of the key parameters on the leather shoe production line, so that the system can quantitatively analyze the complexity of data changes and effectively identify potential production fluctuations and quality deviations. By analyzing the volatility parameters and identifying the periodic abnormal patterns, the accuracy and response speed of the production process control are optimized. In addition, the Bayesian network analysis is performed using historical quality deviation data to predict possible quality fluctuations in the future, thereby achieving forward-looking quality control. The risk of production interruption and defective products caused by parameter deviation is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 is a system flow chart of the present invention;

[0090] Figure 2 A flow chart of the steps for obtaining the logarithmic relationship between the number and size of closed boxes for analyzing data points of the present invention;

[0091] Figure 3 The flowchart of the steps for obtaining the stability index of the leather shoe forming process of the present invention;

[0092] Figure 4 A flow chart of the steps for obtaining periodic abnormal patterns in the operating parameter data of a molding machine according to the present invention;

[0093] Figure 5 This is a flow chart of the steps for obtaining the quality information of leather shoes of the present invention;

[0094] Figure 6 A flow chart of the steps for calculating the quality fluctuations in the future leather shoe molding process of the present invention;

[0095] Figure 7 A flowchart of the steps for obtaining the deviation analysis results of the present invention;

[0096] Figure 8 The present invention is a flowchart of the steps for obtaining the molding quality optimization results. DETAILED DESCRIPTION

[0097] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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.

[0098] 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" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0099] See also Figure 1 , a product quality supervision system for leather shoes processing includes:

[0100] The monitoring data analysis module obtains multiple data such as the operating speed, pressure and temperature of the molding machine from the leather shoe production line, uses the box counting method to analyze the logarithmic relationship between the number and size of the closed boxes of the data points, and calculates the complexity of the data changes in the leather shoe molding process to generate the stability index of the leather shoe molding process;

[0101] The quality stability monitoring module uses the stability index of the leather shoe forming process to perform trend analysis, identify the periodic abnormal patterns in the operating parameter data of the forming machine, and analyze the degree of deviation between the abnormal patterns and the quality standards of the leather shoe forming process, evaluate the quality stability, and generate leather shoe quality information;

[0102] The quality fluctuation prediction module uses Bayesian network to analyze and calculate the quality fluctuation in the future leather shoe molding process based on the leather shoe quality information and historical quality deviation data, identifies the quality parameters that have the greatest impact on the molding process, and obtains the deviation analysis results;

[0103] The quality optimization response module adjusts the operating parameters of the molding machine in real time according to the deviation analysis results. The operating parameters include temperature, pressure or speed, and obtains the optimization results of the leather shoe molding quality.

[0104] The stability indicators of the leather shoe molding process include the data density index, the coefficient of variation and the box dimension of the data set. The leather shoe quality information includes the recognition frequency of abnormal patterns, the intensity of periodic anomalies and the deviation assessment index from the established quality standards. The deviation analysis results include the predicted quality fluctuation range, the sensitivity score of key influencing parameters and the potential risk level. The molding quality optimization results include the stability index of the adjusted operating parameters and the quality consistency comparison before and after optimization.

[0105] See also Figure 2 , the steps to obtain the logarithmic relationship between the number and size of closed boxes of the analyzed data points are as follows:

[0106] Collect the data of running speed V, pressure P and temperature T from the molding machine monitoring system, using the formula:

[0107]

[0108] Calculate the number of boxes N required for the data points;

[0109] Among them, max() is used to return the maximum value in a given data set, min() is used to return the minimum value in a given data set, δV, δP, δT are the box sizes of velocity, pressure, and temperature dimensions;

[0110] Based on the number of boxes N required for the data points, the formula is used:

[0111]

[0112] Calculate the average size S of each box;

[0113] Among them, σ V , σ P , σ T is the standard deviation of velocity, pressure, and temperature data;

[0114] Based on the number of boxes N required for the data points and the average size S of each box, the formula is adopted:

[0115]

[0116] Calculate the ratio R of the number of enclosed boxes of data points to the logarithm of their size;

[0117] Among them, log() is the natural logarithm function;

[0118] The calculation process is as follows:

[0119] The running speed V, pressure P and temperature T data were collected from the shoe forming machine within one hour.

[0120] Assume the data is as follows:

[0121] Speed ​​V: range 100-200m / min;

[0122] Pressure P: range 10-50kPa;

[0123] Temperature T: range 150-200℃;

[0124] The box size was selected as follows: δV = 10 m / min, δP = 5 kPa, δT = 10 °C.

[0125] Calculate the number of boxes using the formula:

[0126]

[0127] The result N = 400 represents the minimum number of boxes required to cover all data points during the entire monitoring period.

[0128] Assume that the standard deviations of velocity, pressure, and temperature data are σ V =20m / min,σ P =10kPa,σ T =12℃.

[0129] Use the formula:

[0130]

[0131] Logarithmic relationship calculation, using the formula:

[0132]

[0133] Assume that the benchmark value is set to 2 (ideal case, indicating perfectly evenly distributed data). The actual calculated value R≈1.86 is lower than the benchmark value, indicating that the data distribution is not perfectly even, and there may be a central tendency or that the data in some areas is sparse.

[0134] See also Figure 3 , the specific steps for obtaining the stability index of the leather shoe forming process are:

[0135] According to the ratio R of the number of enclosed boxes of data points to the logarithm of their size, the formula is:

[0136]

[0137] Calculate the complexity D of data changes;

[0138] Among them, C1 represents the constant coefficient used for adjustment, α1, β1, and γ1 represent the normalized weight coefficients of the velocity, pressure, and temperature data ranges, respectively, which are used to adjust the influence of different parameters in the complexity calculation;

[0139] According to the complexity D of data changes, the formula is used:

[0140]

[0141] Calculate the impact factor G;

[0142] Among them, V var1 , P var1 、Tvar1 Represent the fluctuation range of speed, pressure and temperature respectively;

[0143] Compare the impact factor G with the industry standard threshold Θ1 using the formula:

[0144] Idx=|G-Θ1|

[0145] Determine the stability of the forming process and obtain the stability index Idx of the leather shoe forming process;

[0146] The calculation process is as follows:

[0147] Assume that the logarithmic relationship value R obtained from the shoe molding machine is 1.86, the adjustment coefficient C1 is set to 0.1, and the normalized weight coefficients α1, β1, and γ1 are assumed to be 0.33 respectively, indicating that the speed, pressure, and temperature have an equal impact on complexity. Formula:

[0148]

[0149] Assume that the complexity benchmark value is 0.05, which is used to indicate a medium level of complexity. If the actual complexity is lower than this value, it indicates that the process is relatively simple; if it is higher than this value, it indicates that the process is relatively complex. The resulting complexity D≈0.0348. This value is small, indicating that the data changes in the leather shoe molding process are not particularly complex within the observed time window.

[0150] Assuming the fluctuation range V var1 , P var1 、T var1 They are 10, 5, and 15 respectively. They represent the maximum range of change of speed, pressure, and temperature within a certain period of time.

[0151] formula:

[0152]

[0153] The influence factor G ≈ 0.0333 was obtained. This low value indicates that the molding process has a large variation during the observation period.

[0154] Assume that the industry standard threshold Θ1 is set to 0.05. Formula:

[0155] Idx=|G-Θ1|

[0156] Idx=|0.0333-0.05|≈0.0167

[0157] The calculated Idx≈0.0167. The difference shows that there is a small gap between the actual stability index and the industry standard, indicating that the stability of the molding process is slightly lower than the expected level.

[0158] See also Figure 4 , the steps for obtaining the periodic abnormal pattern in the molding machine operation parameter data are as follows:

[0159] According to the stability index Idx of the leather shoe forming process, the formula is adopted:

[0160]

[0161] By setting thresholds to determine the changes that are considered abnormal at various levels, the initial abnormal analysis indicator M1 is obtained;

[0162] Where, e is the base of natural logarithm;

[0163] According to the initial abnormal analysis indicator M1, the formula is used:

[0164]

[0165] Calculate the sum of the deviations between M1 and its time series average to obtain the cyclical abnormal fluctuation index M2;

[0166] in, represents the M1 value at the i-th time point, n is the total number of data points in the time series, Represents the M1 value at the jth time point, which is used to calculate the time series average of M1;

[0167] According to the periodic abnormal fluctuation index M2, the threshold θ2 is applied again to determine whether there is a periodic anomaly, using the formula:

[0168] Q=M2-θ2

[0169] Calculate the abnormal pattern recognition result Q;

[0170] Among them, if Q is greater than zero, it indicates that there is an abnormal pattern, otherwise it is considered to be a normal periodic change;

[0171] The calculation process is as follows:

[0172] Assuming that the known stability indicator Idx is approximately 0.0167, use the logistic regression model to determine whether it is abnormal:

[0173]

[0174] Assume Θ2=0.5.

[0175]

[0176] The result M1=0.381 represents the preliminary abnormality assessment result, and the closer it is to 1, the higher the possibility of abnormality.

[0177] formula:

[0178]

[0179] Suppose there is a time series data set: M1 = [0.38, 0.40, 0.39, 0.37, 0.36], with a total of n = 5 data points.

[0180] average value:

[0181]

[0182] M2=0.04

[0183] Assume 0.1 as the reference value. If it exceeds 0.1, it indicates that the periodic abnormal fluctuation is obvious, which may indicate that there is a problem with the operation of the equipment or that adjustments are needed. The result M2 = 0.04 indicates that the abnormal fluctuation is very small and the periodic abnormality is not obvious.

[0184] formula:

[0185] Q=M2-θ2

[0186] Assume Θ2 is the same as above, which is 0.5.

[0187] Q = 0.04-0.5 = -0.46

[0188] Because the Q value is less than zero, it means there is no obvious periodic anomaly. Conclusion Through the above steps, it can be seen that the current molding machine operating parameters show normal operation under the set monitoring system, and no significant periodic anomaly is detected. This result means that the production line is currently within the normal operating range and no adjustment is required.

[0189] See also Figure 5 , the specific steps for obtaining leather shoes quality information are:

[0190] According to the abnormal pattern recognition result Q, the quality standard threshold is set to Θ3, using the formula:

[0191] O=|Q-Θ3|

[0192] Calculate the deviation between Q and the quality standard threshold Θ3 to obtain the degree of deviation O;

[0193] According to the degree of deviation O, the formula is used:

[0194]

[0195] Calculate the quality stability index L and obtain the quality information of leather shoes;

[0196] Among them, k1 is an empirical parameter used to adjust the sensitivity of S, and μ1 is another empirical parameter used to adjust the reference point;

[0197] The calculation process is as follows:

[0198] formula:

[0199] O=|Q-Θ3|

[0200] Assume that the set threshold θ3 of the quality standard is 0.1, which represents the maximum deviation value that will be accepted.

[0201] O=|-0.46-0.1|=|-0.56|=0.56

[0202] Here, the value of O is 0.56, which means that there is a large deviation between the actual monitored abnormal indicators and the set threshold.

[0203] formula:

[0204]

[0205] Fixed empirical parameters k1 = 0.5 and μ1 = 0.05 were used. The parameters were set based on historical data and statistical analysis to adjust sensitivity and reference points.

[0206]

[0207] Here, the value of L is 0.936, indicating that the current quality stability is very high.

[0208] Assuming the benchmark stability index is 0.9, it is used to determine whether improvement measures need to be taken. L = 0.936 compared to the benchmark value of 0.9, indicating that the current quality control is within the acceptable range, the production process is relatively stable, and no immediate intervention is required.

[0209] See also Figure 6 , the steps for calculating the quality fluctuation in the future leather shoe molding process are as follows:

[0210] According to the quality stability index L and the historical quality deviation data H, the formula is used:

[0211]

[0212] Initialize the node value B of the Bayesian network;

[0213] Among them, α2 is a weight parameter used to adjust the influence ratio of current data and historical data in quality prediction, and c2 is a constant used to adjust the influence of historical data;

[0214] Using the node value B of the initialized Bayesian network, the formula is:

[0215]

[0216] Calculate the estimated value W of future quality fluctuations;

[0217] Among them, β2 is the adjustment coefficient, which is used to adjust the influence of network output on actual prediction, γ2 is the parameter for adjusting the smoothness of network output, and δ2 is used to adjust the sensitivity of network output to the node value B of the Bayesian network;

[0218] The calculation process is as follows:

[0219] formula:

[0220]

[0221] Assume H=0.2, α2=0.7.

[0222]

[0223] B=0.6552+0.3×0.5831≈0.6552+0.1749=0.8301

[0224] The result B = 0.8301 is used to initialize the Bayesian network, which represents the quality situation of the integrated current and historical data.

[0225] formula:

[0226]

[0227] Assume β2=1.2, γ2=1.0, δ2=0.5.

[0228]

[0229] The result W=0.7042 indicates the estimated degree of future quality fluctuation.

[0230] Assuming the benchmark quality stability index is W = 0.5, it is used to determine whether the quality fluctuation is beyond the normal range. W = 0.7042 compared with the benchmark value of 0.5 indicates that the expected quality fluctuation is slightly higher than the normal range, and future quality control and risk management may need attention.

[0231] See also Figure 7 , the specific steps for obtaining the deviation analysis results are:

[0232] According to the estimated value W of future quality fluctuations, the formula is used:

[0233]

[0234] Calculate the total impact U of each quality parameter on future quality fluctuations;

[0235] Among them, f k is the historical impact record of the kth quality parameter, σ kis the uncertainty coefficient associated with the kth quality parameter, which is used to reflect the variability or risk level of each parameter, and g is the total number of parameters;

[0236] According to the total impact U of each quality parameter on future quality fluctuations, the formula is adopted:

[0237]

[0238] Calculate the quality parameter value Z that has the greatest impact on the molding process and obtain the deviation analysis results;

[0239] Among them, U k is the deviation value of the kth quality parameter, r k is the relevance factor, which is used to weight the influence of each parameter to ensure that the identified parameters actually have a significant impact on the molding process;

[0240] The calculation process is as follows:

[0241] Assume there are three mass parameters: velocity (f1), pressure (f2), and temperature (f3). Using the formula:

[0242]

[0243] g=3 is the total number of parameters.

[0244] Assuming the historical impact records f1 = 0.8, f2 = 0.65, f3 = 0.75 and the uncertainty coefficient σ1 =

[0245] 0.05, σ2=0.03, σ3=0.04.

[0246]

[0247] Total Contribution:

[0248] U=1.597+1.355+0.916=3.868

[0249] formula:

[0250]

[0251] Assume that all r k =1.

[0252]

[0253]

[0254] Z=max(1.597,1.355,0.916)=1.597

[0255] It shows that the speed parameter f1 has the greatest influence on the forming process.

[0256] Assuming the reference value is 1.5, it means that any parameter effect exceeding this value needs special attention. Z = 1.597 shows that the speed parameter exceeds the reference value, indicating that it has a significant impact on the stability of product quality. Measures need to be taken to reduce its fluctuation or adjust related control parameters to ensure the optimization and quality control of the molding process.

[0257] See also Figure 8 , the specific steps for obtaining the molding quality optimization results are:

[0258] According to the quality parameter value Z which has the greatest impact on the molding process, the formula is used:

[0259]

[0260] Calculate the adjustment factor A;

[0261] Among them, τ3 is the acceptable threshold of quality deviation, which is used to standardize the adjustment range;

[0262] According to the adjustment factor A, the formula is used:

[0263] T′=T×A

[0264] and

[0265] P′=P×A

[0266] and

[0267] V′=V×A

[0268] Calculate the adjusted temperature T′, pressure P′ and speed V′ to obtain the molding quality optimization result;

[0269] The calculation process is as follows:

[0270] formula:

[0271]

[0272] Assume Z = 1.597, τ3 = 1.5.

[0273]

[0274] A=1-1.0647≈-0.0647

[0275] The adjustment coefficient A is less than zero, which means that in this case, it may be necessary to reversely adjust the operating parameters to reduce the deviation of the molding quality, that is, the temperature, pressure and speed may need to be reduced in actual operation.

[0276] formula:

[0277] T′=T×A

[0278] P′=P×A

[0279] V′=V×A

[0280] Assume the original operation parameters are:

[0281] T=150℃, P=200bar, V=60m / min.

[0282] T′=150×-0.0647≈-9.705℃

[0283] P′=200×-0.0647≈-12.94bar

[0284] V′=60×-0.0647≈-3.882m / min

[0285] These negative values ​​represent the need for reverse adjustments in the theoretical model, but in practice this means reducing the original setpoints.

[0286] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A product quality supervision system for leather shoe processing, characterized in that: The system comprises: The monitoring data analysis module obtains multiple data such as the operating speed, pressure and temperature of the molding machine from the leather shoe production line, uses the box counting method to analyze the logarithmic relationship between the number and size of the closed boxes of the data points, and calculates the complexity of the data changes in the leather shoe molding process to generate the stability index of the leather shoe molding process; The quality stability monitoring module uses the stability index of the leather shoe forming process to perform trend analysis, identify periodic abnormal patterns in the operating parameter data of the forming machine, and analyze the degree of deviation between the abnormal patterns and the quality standards of the leather shoe forming process, evaluate the quality stability, and generate leather shoe quality information; The quality fluctuation prediction module uses the Bayesian network to analyze and calculate the quality fluctuation in the future leather shoe molding process based on the leather shoe quality information and historical quality deviation data, identifies the quality parameters that have the greatest impact on the molding process, and obtains the deviation analysis results; The quality optimization response module adjusts the operating parameters of the molding machine in real time according to the deviation analysis results, and the operating parameters include temperature, pressure or speed to obtain the leather shoe molding quality optimization results.

2. The product quality supervision system for leather shoe processing according to claim 1 is characterized in that: The steps for obtaining the logarithmic relationship between the number and size of the closed boxes of the analysis data points are specifically as follows: Collect the data of running speed V, pressure P and temperature T from the molding machine monitoring system, using the formula: Calculate the number of boxes N required for the data points; Among them, max() is used to return the maximum value in a given data set, min() is used to return the minimum value in a given data set, δV, δP, δT are the box sizes of velocity, pressure, and temperature dimensions; Based on the number of boxes N required for the data points, the formula is used: Calculate the average size S of each box; Among them, σ V , σ P , σ T is the standard deviation of velocity, pressure, and temperature data; Based on the number of boxes N required for the data points and the average size S of each box, the formula is used: Calculate the ratio R of the number of enclosed boxes of data points to the logarithm of their size; Where log( ) is the natural logarithm function.

3. The product quality supervision system for leather shoe processing according to claim 2 is characterized in that: The steps for obtaining the stability index of the leather shoe forming process are specifically as follows: According to the ratio R of the number of enclosed boxes of the data point to the logarithm of the size, the formula is adopted: Calculate the complexity D of data changes; Among them, C1 represents the constant coefficient used for adjustment, α1, β1, and γ1 represent the normalized weight coefficients of the velocity, pressure, and temperature data ranges, respectively; According to the complexity D of the data change, the formula is adopted: Calculate the impact factor G; Among them, V var1 , P var1 , T var1 Represent the fluctuation range of speed, pressure and temperature respectively; The impact factor G is compared with the industry standard threshold θ1 using the formula: Idx=|G-Θ1| Determine the stability of the forming process and obtain the stability index Idx of the leather shoe forming process.

4. The product quality supervision system for leather shoe processing according to claim 3 is characterized in that: The steps of obtaining the periodic abnormal pattern in the operating parameter data of the molding machine are specifically as follows: According to the stability index Idx of the leather shoe forming process, the formula is adopted: By setting thresholds to determine the changes that are considered abnormal at various levels, the initial abnormal analysis indicator M1 is obtained; Where, e is the base of natural logarithm; According to the initial abnormal analysis indicator M1, the formula is adopted: Calculate the sum of the deviations between M1 and its time series average to obtain the cyclical abnormal fluctuation index M2; in, represents the M1 value at the i-th time point, n is the total number of data points in the time series, represents the M1 value at the jth time point; According to the periodic abnormal fluctuation index M2, the threshold θ2 is applied again to determine whether there is a periodic abnormality, using the formula: Q=M2-θ2 Calculate the abnormal pattern recognition result Q; If Q is greater than zero, it indicates an abnormal pattern, otherwise it is a normal periodic change.

5. The product quality supervision system for leather shoe processing according to claim 4 is characterized in that: The steps for obtaining the leather shoes quality information are specifically as follows: According to the abnormal pattern recognition result Q, the quality standard threshold is set to θ3, using the formula: O=|Q-Θ3| Calculate the deviation between Q and the quality standard threshold Θ3 to obtain the degree of deviation O; According to the degree of deviation O, the formula is used: Calculate the quality stability index L and obtain the quality information of leather shoes; Among them, k1 is an empirical parameter used to adjust the sensitivity of S, and μ1 is another empirical parameter used to adjust the reference point.

6. The product quality supervision system for leather shoe processing according to claim 5 is characterized in that: The steps for calculating the quality fluctuation in the future leather shoe forming process are specifically as follows: According to the quality stability index L and the historical quality deviation data H, the formula is adopted: Initialize the node value B of the Bayesian network; Among them, α2 is a weight parameter used to adjust the influence ratio of current data and historical data in quality prediction, and c2 is a constant used to adjust the influence of historical data; Using the node value B of the initialized Bayesian network, the formula is: Calculate the estimated value W of future quality fluctuations; Among them, β2 is the adjustment coefficient, which is used to adjust the influence of network output on prediction, γ2 is the parameter for adjusting the smoothness of network output, and δ2 is used to adjust the sensitivity of network output to the node value B of the Bayesian network.

7. The product quality supervision system for leather shoe processing according to claim 6 is characterized in that: The steps for obtaining the deviation analysis results are specifically as follows: According to the estimated value W of future quality fluctuation, the formula is adopted: Calculate the total impact U of each quality parameter on future quality fluctuations; Among them, f k is the historical impact record of the kth quality parameter, σ k is the uncertainty coefficient associated with the kth quality parameter, g is the total number of parameters; According to the total influence U of each quality parameter on future quality fluctuations, the formula is adopted: Calculate the quality parameter value Z that has the greatest impact on the molding process and obtain the deviation analysis results; Among them, U k is the deviation value of the kth quality parameter, r k is the correlation factor used to weight the influence of each parameter.

8. The product quality supervision system for leather shoe processing according to claim 7 is characterized in that: The steps for obtaining the molding quality optimization result are specifically as follows: According to the quality parameter value Z which has the greatest impact on the molding process, the formula is adopted: Calculate the adjustment factor A; Among them, τ3 is the acceptable threshold of mass deviation; According to the adjustment coefficient A, the formula is adopted: T′=T×A and P′=P×A and V′=V×A The adjusted temperature T′, pressure P′ and speed V′ are calculated to obtain the molding quality optimization result.

Images