A data processing method for intelligent manufacturing production line of shoes and boots

Abstract

The present invention relates to the field of data processing, and in particular to a data processing method for an intelligent manufacturing production line for footwear, the method comprising: using a DBSCAN algorithm to perform abnormal detection on a main temperature sequence to obtain abnormal temperatures, intercepting multiple data with the abnormal temperatures as the center to construct a main subsequence; obtaining a reference temperature in the same time period as the main subsequence to construct a reference subsequence; calculating the primary noise coefficient of the abnormal temperature; calculating the noise coefficient of the abnormal temperature, the noise coefficient is positively correlated with the product of the primary noise coefficient and the cumulative sum of the Pearson correlation coefficient; when the noise coefficient is greater than a preset coefficient threshold, the corresponding abnormal temperature is a noise abnormal temperature. The present invention analyzes the temperature data change characteristics at multiple locations on the surface of a heating plate, calculates the noise coefficient through multiple dimensions, and facilitates the detection of noise abnormal temperature data through the noise coefficient, thereby effectively improving the accuracy of temperature anomaly detection.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a data processing method for an intelligent manufacturing production line for shoes and boots. Background Art

[0002] The footwear industry is facing rapidly changing market demands and technological advances, especially in the context of the growth of e-commerce and personalized consumer demand. Intelligent manufacturing has emerged, which can significantly improve production efficiency through automation, digitization and intelligence. In the footwear production line, the heat press can evenly heat the footwear material in a short time to achieve the optimal molding temperature, thereby ensuring the quality and consistency of the product. Monitoring the temperature data of the heating plate can effectively avoid overheating of the equipment and ensure the safety of the production process. It can also ensure that different components can be bonded at the appropriate temperature, thereby improving the overall strength and durability of the product. When collecting the temperature data of the heating plate of the heat press, the electromagnetic interference of the surrounding equipment or power lines will affect the signal of the temperature sensor, thereby generating temperature noise data.

[0003] The Chinese patent application document with publication number CN114861788A discloses a load anomaly detection method and system based on DBSCAN clustering, the method comprising: step 1, collecting load anomaly data; step 2, preprocessing the load anomaly data to obtain a reliable data set; dividing the reliable data set into a training set and a test set; step 3, building a DBSCAN clustering model, taking the training set as input and the anomaly type clustering cluster as output, and training the clustering model; step 4, using the trained clustering model to cluster the test set, and using the silhouette coefficient as an evaluation index for the clustering result, to obtain multiple identification type clusters of load anomalies.

[0004] The DBSCAN algorithm can be used to detect abnormal values ​​in the temperature data of the heating plate of the hot press. The DBSCAN algorithm can effectively identify and mark abnormal values ​​and adapt to different data distributions. Since the numerical value of the temperature noise data is relatively extreme, when using the DBSCAN algorithm to detect abnormalities in the heating plate temperature data, it is easy to mistakenly detect the temperature noise data as real abnormal temperature data, resulting in a decrease in the accuracy of temperature anomaly detection. Summary of the invention

[0005] In order to identify abnormal temperature due to noise and improve the accuracy of detecting real abnormal temperature data, the present invention provides a data processing method for an intelligent manufacturing production line for shoes and boots.

[0006] The present invention provides a data processing method for a footwear intelligent manufacturing production line, which adopts the following technical solutions:

[0007] Obtain the main body temperature of the surface of the heating plate of the hot press and the reference temperatures of multiple reference data points at the same time, construct a main body temperature sequence according to the main body temperature at different times, and construct a reference temperature sequence according to the reference temperature of the same reference data point at different times;

[0008] The DBSCAN algorithm is used to detect the abnormal temperature of the main temperature sequence to obtain the abnormal temperature and determine whether the abnormal temperature is a noise abnormal temperature;

[0009] The method for determining the abnormal temperature of noise is as follows: in the main temperature sequence, multiple data are intercepted with the abnormal temperature as the center to construct the main subsequence; in the reference temperature sequence, the reference temperature in the same time period as the main subsequence is obtained to construct the reference subsequence;

[0010] The cumulative sum of the differences between the abnormal temperature and the neighboring data points is calculated, and the primary noise coefficient of the abnormal temperature is calculated. The primary noise coefficient is positively correlated with the cumulative sum of the differences. The Pearson correlation coefficient between the main subsequence to which the abnormal temperature belongs and each reference subsequence in the corresponding time period is calculated, and the cumulative sum of the Pearson correlation coefficients is further obtained.

[0011] The noise coefficient of the abnormal temperature is calculated, and the noise coefficient is positively correlated with the product of the primary noise coefficient and the cumulative sum of the Pearson correlation coefficient; when the noise coefficient is greater than a preset coefficient threshold, the corresponding abnormal temperature is the noise abnormal temperature.

[0012] By analyzing the temperature data change characteristics at multiple positions on the surface of the heating plate, the noise coefficient of each abnormal temperature in the main temperature sequence is obtained. Combined with the coefficient threshold, it is determined whether the abnormal temperature is a noise abnormal temperature. The noise abnormal temperature can be identified, which improves the accuracy of detecting the real abnormal temperature data. The real abnormal temperature is obtained by eliminating the noise abnormal temperature, which facilitates the monitoring of the heating plate temperature of the hot press.

[0013] Preferably, the expression of the noise coefficient is:

[0014] ;

[0015] Where: represents the noise coefficient of the nth abnormal temperature in the main temperature sequence, represents the primary noise factor of the nth abnormal temperature in the main temperature data sequence, represents the nth abnormal temperature in the main temperature sequence, represents the average value of all reference temperatures at the time corresponding to the nth abnormal temperature in the main temperature sequence, γ is a preset hyperparameter, H represents the number of reference subsequences, represents the Pearson correlation coefficient between the main subsequence to which the nth abnormal temperature belongs and the hth reference subsequence in the corresponding time period, and norm represents the normalization function.

[0016] By calculating the difference between the abnormal temperature in the main temperature sequence and the temperature at the same time in the reference temperature sequence, and by calculating the Pearson coefficient between the main subsequence and the reference subsequence, the noise coefficient is calculated from multiple dimensions, which improves the accuracy of the noise coefficient and facilitates the screening of noise abnormal temperatures.

[0017] Preferably, the expression of the noise coefficient is:

[0018] ;

[0019] Where: represents the noise coefficient of the nth abnormal temperature in the main temperature sequence, represents the primary noise factor of the nth abnormal temperature in the main temperature data sequence, represents the number of reference data points, represents the Pearson correlation coefficient between the main subsequence to which the nth abnormal temperature belongs and the reference subsequence of the hth reference data point in the corresponding time period, Represents the normalization function.

[0020] Preferably, the expression of the primary noise coefficient is:

[0021]

[0022] Where: represents the primary noise factor of the nth abnormal temperature in the main temperature series, Indicates the number of data points in the nth abnormal temperature neighborhood when using the DBSCAN algorithm for anomaly detection. Indicates the value of the nth abnormal temperature in the main temperature sequence, Represents the value of the u-th neighboring data point of the n-th abnormal temperature in the main temperature series.

[0023] Preferably, the expression of the primary noise coefficient is:

[0024]

[0025] In the formula, represents the primary noise factor of the nth abnormal temperature in the main temperature series, Indicates the number of data points in the nth abnormal temperature neighborhood when using the DBSCAN algorithm for anomaly detection. Indicates the value of the nth abnormal temperature in the main temperature sequence, Represents the value of the u-th neighboring data point of the n-th abnormal temperature in the main temperature sequence, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs.

[0026] The accuracy of calculating the primary coefficient is improved by calculating the primary noise coefficient from multiple dimensions, including the number of abnormal temperature neighborhood data points, the mutation of the main subsequence, and the difference between the abnormal temperature and the neighborhood data points in the main temperature sequence.

[0027] Preferably, the expression of mutation is:

[0028]

[0029] In the formula, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs, It represents the mean of the difference between all adjacent data points in the main subsequence to which the nth abnormal temperature in the main temperature data sequence belongs. It represents the number of data points with a second-order derivative of 0 in the subsequence of the nth abnormal temperature in the main temperature sequence. Represents the normalization function.

[0030] By calculating the mutation, it is possible to indicate the severity of the data change of the main subsequence to which the abnormal temperature belongs, so as to facilitate the subsequent detection of whether the abnormal temperature is a noise abnormal temperature.

[0031] Preferably, the expression of mutation is:

[0032]

[0033] In the formula, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs, It represents the mean of the difference between all adjacent data points in the main subsequence to which the nth abnormal temperature in the main temperature data sequence belongs. It represents the number of data points with a second-order derivative of 0 in the subsequence of the nth abnormal temperature in the main temperature sequence. Represents the normalization function.

[0034] Preferably, the method of intercepting multiple data and constructing the main subsequence with abnormal temperature as the center is:

[0035] In the main temperature sequence, multiple data are intercepted along the left side of the abnormal temperature, and multiple data are intercepted along the right side of the abnormal temperature, and the number of data intercepted on the left side is the same as the number of data intercepted on the right side. The intercepted data and the abnormal temperature are used as the main subsequence.

[0036] By constructing the main subsequence, it is convenient to compare with the reference subsequence and calculate the noise coefficient of abnormal temperature.

[0037] The present invention has the following technical effects:

[0038] The present invention analyzes the temperature data change characteristics at multiple locations on the surface of the heating plate and calculates the noise coefficient through multiple dimensions. The noise abnormality temperature data can be detected through the noise coefficient, which effectively improves the accuracy of temperature anomaly detection and enhances the system's robustness to electromagnetic interference, thereby improving the monitoring and subsequent control accuracy of the heating plate. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0040] Figure 1 It is a method flow chart of a data processing method for an intelligent manufacturing production line of shoes and boots according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0042] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0043] The embodiment of the present invention discloses a data processing method for a footwear intelligent manufacturing production line, referring to Figure 1 , including the following steps, as follows:

[0044] S1: Obtain the main body temperature of the surface of the hot press heating plate and the reference temperatures of multiple reference data points at the same time, construct a main body temperature sequence according to the main body temperature at different times, and construct a reference temperature sequence according to the reference temperatures of the same data point at different times.

[0045] Temperature sensors are used to collect temperature data on the surface of the heating plate of the hot press in the footwear production line. Specifically, five temperature sensors are used, one of which is placed in the middle to collect the temperature of the middle part of the heating plate, and the temperature of the middle part is used as the main temperature data; the other four temperature sensors are placed in the four directions of the middle temperature sensor, 2 cm away from the main temperature sensor, and each temperature sensor corresponds to a reference data point. The temperature data collected by the four temperature sensors is used as the reference temperature. When the five temperature sensors collect temperature data, all temperature sensors collect temperature data at the same time. The main temperature sequence is constructed using the main temperature data collected at different times, and the reference temperature sequence is constructed using the reference temperatures collected at different times. It should be noted that one reference data point corresponds to one reference temperature sequence, and four reference data points correspond to four reference temperature sequences in total.

[0046] S2: Use the DBSCAN algorithm to perform anomaly detection on the main temperature sequence to obtain abnormal temperature. In the main temperature sequence, multiple data are intercepted with the abnormal temperature as the center to construct a main subsequence; in the reference temperature sequence, the reference temperature in the same time period as the main subsequence is obtained to construct a reference subsequence.

[0047] When using the DBSCAN algorithm to detect anomalies in the main temperature sequence, the neighborhood radius is preset to an empirical value of 20, and the minimum number of neighborhood points required for a point to be considered a core point is preset to an empirical value of 18. The method of intercepting multiple data with abnormal temperature as the center to construct the main subsequence is as follows: in the main temperature sequence, multiple data are intercepted along the left side of the abnormal temperature, and multiple data are intercepted along the right side of the abnormal temperature, and the number of data intercepted on the left side is the same as the number of data intercepted on the right side, and the intercepted data and the abnormal temperature are used as the main subsequence.

[0048] For example, in the main temperature sequence, with the abnormal temperature as the center, 25 data are selected on the left and 25 data are selected on the right, and the main subsequence is constructed using the 50 data on the left and right sides and the abnormal temperature. If the number of data on the left is 20, which is less than 25, we can fill it with data on the right, that is, we select 30 data on the right. The corresponding time period is ( , ), then select ( , ) The reference temperature in each reference temperature sequence within the time period constructs a reference subsequence, the main subsequence Corresponding to four reference subsequences.

[0049] S3: Calculate the cumulative sum of the differences between the abnormal temperature and the neighboring data points, and calculate the primary noise coefficient of the abnormal temperature. The primary noise coefficient is positively correlated with the cumulative sum of the differences.

[0050] In one embodiment, the expression for the primary noise figure is:

[0051]

[0052]

[0053] In the formula, represents the primary noise factor of the nth abnormal temperature in the main temperature series, Indicates the number of data points in the nth abnormal temperature neighborhood when using the DBSCAN algorithm for anomaly detection. Indicates the value of the nth abnormal temperature in the main temperature sequence, Represents the value of the u-th neighboring data point of the n-th abnormal temperature in the main temperature sequence, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs, It represents the mean of the difference between all adjacent data points in the main subsequence to which the nth abnormal temperature in the main temperature data sequence belongs. It represents the number of data points with a second-order derivative of 0 in the subsequence of the nth abnormal temperature in the main temperature sequence. Represents the normalization function.

[0054] In the formula It represents the sum of the numerical differences between the nth abnormal temperature in the main temperature sequence and each data point in the neighborhood. The larger its value is, the greater the difference between the corresponding abnormal temperature and the data points in its neighborhood is. The more prominent the corresponding abnormal temperature is, the greater the possibility that it belongs to potential noise temperature data, and the larger the corresponding primary noise coefficient is. The smaller it is, the fewer the number of data points in the neighborhood of the nth abnormal temperature in the main temperature sequence is, which means that the existence of the corresponding abnormal temperature is more special in its neighborhood data, and the greater the possibility that the corresponding abnormal temperature belongs to potential noise temperature data, the larger the corresponding primary noise coefficient is. The larger it is, the more drastic the data change in the main subsequence to which the corresponding abnormal temperature belongs is, then the possibility that the corresponding abnormal temperature belongs to potential noise temperature data is greater, and the possibility that it belongs to real abnormal temperature data is smaller, and the primary noise coefficient of the corresponding abnormal temperature will be larger.

[0055] in the formula The larger the value is, the more dramatic the change in the values ​​of the data points in the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs is, and the greater the mutation of the main subsequence to which the nth abnormal temperature data point in the main temperature sequence belongs is. The larger it is, the more frequent the numerical changes in the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs are, and the more drastic the numerical changes between the data points in the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs are. The greater the credibility, the greater the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs will be.

[0056] In another embodiment, the expression of mutation is:

[0057]

[0058] In the formula, Indicates the mutation of the surrounding data segment to which the nth abnormal temperature in the main temperature sequence belongs. It represents the mean of the difference between all adjacent data points in the main subsequence to which the nth abnormal temperature in the main temperature data sequence belongs. It represents the number of data points with a second-order derivative of 0 in the subsequence of the nth abnormal temperature in the main temperature sequence. Represents the normalization function.

[0059] S4: Calculate the Pearson correlation coefficient between the main subsequence to which the abnormal temperature belongs and each reference subsequence in the corresponding time period, further obtain the cumulative sum of the Pearson correlation coefficients, and calculate the noise coefficient of the abnormal temperature. The noise coefficient is positively correlated with the product of the primary noise coefficient and the cumulative sum of the Pearson correlation coefficients.

[0060] The primary noise coefficient of each abnormal temperature in the main temperature sequence is obtained by calculation. The primary noise coefficient is calculated based on the main temperature sequence. In the actual collected temperature sequence, if a data point is noise data, then its surrounding data may also be affected by noise and produce numerical drift, which in turn causes the data changes around the noise data point to have the same gradual characteristics as the data around the real abnormal temperature. At the same time, since the temperature at different positions on the heating plate of the hot press at the same time is highly consistent, the connection between the temperature data changes in the main temperature sequence and multiple reference temperature sequences is analyzed, and the primary noise coefficient of each abnormal temperature in the main temperature sequence is optimized to obtain the noise coefficient of each abnormal temperature.

[0061] In one embodiment, the expression of the noise factor is:

[0062] ;

[0063] Where: represents the noise coefficient of the nth abnormal temperature in the main temperature sequence, represents the primary noise factor of the nth abnormal temperature in the main temperature data sequence, represents the nth abnormal temperature in the main temperature sequence, represents the average value of all reference temperatures at the time corresponding to the nth abnormal temperature in the main temperature sequence, and γ is a preset hyperparameter, which exists to prevent When the value of is 0, Indicates the number of reference subsequences. In this embodiment , It represents the Pearson correlation coefficient between the main subsequence to which the nth abnormal temperature belongs and the hth reference subsequence in the corresponding time period, Represents the normalization function.

[0064] in the formula The larger the value is, the more likely it is that the nth abnormal temperature in the main temperature sequence belongs to noise data, and the larger the corresponding noise coefficient will be. The larger it is, the lower the correlation between the nth abnormal temperature in the main temperature sequence and the reference temperature at the same moment, which means that the authenticity of the nth abnormal temperature in the main temperature sequence is lower, and the corresponding abnormal temperature is more likely to be potential noise data, and its corresponding noise coefficient will be larger. The smaller the value is, the lower the correlation between the nth abnormal temperature in the main temperature sequence and the reference temperature at the same moment is, which means that the authenticity of the nth abnormal temperature in the main temperature sequence is lower, and the corresponding abnormal temperature is more likely to be potential noise data, and its corresponding noise coefficient will be larger.

[0065] In another embodiment, the expression of the noise factor is:

[0066] ;

[0067] Where: represents the noise coefficient of the nth abnormal temperature in the main temperature sequence, represents the primary noise coefficient of the nth abnormal temperature in the main temperature data sequence, H represents the number of reference data points, represents the Pearson correlation coefficient between the main subsequence to which the nth abnormal temperature belongs and the reference subsequence of the hth reference data point in the corresponding time period, and norm represents the normalization function.

[0068] S5: When the noise coefficient is greater than a preset coefficient threshold, the corresponding abnormal temperature is the noise abnormal temperature.

[0069] If the noise coefficient of the abnormal temperature in the main temperature sequence is greater than the preset coefficient threshold, it indicates that the corresponding abnormal temperature is a noise abnormal temperature. The coefficient threshold is set manually according to the actual situation. For example, the coefficient threshold is 0.6. All abnormal temperatures in the main temperature sequence are traversed, and the noise abnormal temperatures are eliminated to obtain the real abnormal temperature data, so as to detect the abnormal values ​​in the temperature data of the hot press heating plate, thereby improving the accuracy of the detection.

[0070] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0071] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data processing method for a footwear intelligent manufacturing production line, characterized in that: Includes steps: Obtain the main body temperature of the surface of the heating plate of the hot press and the reference temperatures of multiple reference data points at the same time, construct a main body temperature sequence according to the main body temperature at different times, and construct a reference temperature sequence according to the reference temperature of the same reference data point at different times; The DBSCAN algorithm is used to detect the abnormal temperature of the main temperature sequence to obtain the abnormal temperature and determine whether the abnormal temperature is a noise abnormal temperature; The method for determining the abnormal temperature of noise is as follows: in the main temperature sequence, multiple data are intercepted with the abnormal temperature as the center to construct the main subsequence; in the reference temperature sequence, the reference temperature in the same time period as the main subsequence is obtained to construct the reference subsequence; The accumulated sum of the differences between the abnormal temperature and the neighboring data points is calculated, and the primary noise coefficient of the abnormal temperature is calculated. The primary noise coefficient is positively correlated with the accumulated sum of the differences. Calculate the Pearson correlation coefficient between the main subsequence to which the abnormal temperature belongs and each reference subsequence in the corresponding time period, and further obtain the cumulative sum of the Pearson correlation coefficients; Calculate the noise coefficient of abnormal temperature. The noise coefficient is positively correlated with the product of the primary noise coefficient and the cumulative sum of the Pearson correlation coefficient. When the noise coefficient is greater than a preset coefficient threshold, the corresponding abnormal temperature is the noise abnormal temperature.

2. The data processing method for the intelligent manufacturing production line of footwear according to claim 1 is characterized in that: The expression for noise figure is: ； Where: represents the noise coefficient of the nth abnormal temperature in the main temperature sequence, represents the primary noise factor of the nth abnormal temperature in the main temperature data sequence, represents the nth abnormal temperature in the main temperature sequence, It represents the average value of all reference temperatures at the time corresponding to the nth abnormal temperature in the main temperature sequence. is the preset hyperparameter, represents the number of reference subsequences, It represents the Pearson correlation coefficient between the main subsequence to which the nth abnormal temperature belongs and the hth reference subsequence in the corresponding time period, Represents the normalization function.

3. The data processing method for a footwear intelligent manufacturing production line according to claim 1 is characterized in that: The expression for noise figure is: ； Where: represents the noise coefficient of the nth abnormal temperature in the main temperature sequence, represents the primary noise factor of the nth abnormal temperature in the main temperature data sequence, represents the number of reference data points, represents the Pearson correlation coefficient between the main subsequence to which the nth abnormal temperature belongs and the reference subsequence of the hth reference data point in the corresponding time period, Represents the normalization function.

4. A data processing method for a footwear intelligent manufacturing production line according to claim 2 or 3, characterized in that: The expression for the primary noise figure is: Where: represents the primary noise factor of the nth abnormal temperature in the main temperature series, Indicates the number of data points in the nth abnormal temperature neighborhood when using the DBSCAN algorithm for anomaly detection. Indicates the value of the nth abnormal temperature in the main temperature sequence, Represents the value of the u-th neighboring data point of the n-th abnormal temperature in the main temperature series.

5. A method for processing data of a footwear intelligent manufacturing production line according to claim 2 or 3, characterized in that: The expression for the primary noise figure is: In the formula, represents the primary noise factor of the nth abnormal temperature in the main temperature series, Indicates the number of data points in the nth abnormal temperature neighborhood when using the DBSCAN algorithm for anomaly detection. Indicates the value of the nth abnormal temperature in the main temperature sequence, Represents the value of the u-th neighboring data point of the n-th abnormal temperature in the main temperature sequence, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs.

6. The data processing method for the intelligent manufacturing production line of footwear according to claim 5 is characterized in that: The expression of mutation is: In the formula, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs, It represents the mean of the difference between all adjacent data points in the main subsequence to which the nth abnormal temperature in the main temperature data sequence belongs. It represents the number of data points with a second-order derivative of 0 in the subsequence of the nth abnormal temperature in the main temperature sequence. Represents the normalization function.

7. The data processing method for the intelligent manufacturing production line of footwear according to claim 5 is characterized in that: The expression of mutation is: In the formula, Indicates the mutation of the main subsequence to which the nth abnormal temperature in the main temperature sequence belongs, It represents the mean of the difference between all adjacent data points in the main subsequence to which the nth abnormal temperature in the main temperature data sequence belongs. It represents the number of data points with a second-order derivative of 0 in the subsequence of the nth abnormal temperature in the main temperature sequence. Represents the normalization function.

8. The data processing method for a footwear intelligent manufacturing production line according to claim 1 is characterized in that: The method of intercepting multiple data and constructing the main subsequence with abnormal temperature as the center is: In the main temperature sequence, multiple data are intercepted along the left side of the abnormal temperature, and multiple data are intercepted along the right side of the abnormal temperature, and the number of data intercepted on the left side is the same as the number of data intercepted on the right side. The intercepted data and the abnormal temperature are used as the main subsequence.

Images