A method for accurately predicting the temperature of a shoe drying oven

Abstract

The present application relates to the field of temperature measurement technology, and specifically to a method for accurately predicting the temperature of a shoemaking oven, the method comprising: obtaining the temperature of the front section, middle section and rear section of the oven at each moment and the ambient temperature at each moment through a temperature sensor; determining the first similarity, the second similarity and the third similarity, and obtaining the temperature correlation degree after fusion; obtaining the corrected oven temperature at each moment; determining the oven temperature correlation index; determining the gamma parameter of the kernel function in the support vector regression model, and predicting the temperature of the shoemaking oven. The present application can adjust the gamma parameter according to the ambient temperature and the temperature of different areas of the oven, so that the measurement result of the oven temperature is more accurate, and the accuracy of the shoemaking oven temperature measurement is improved.

Description

Technical Field

[0001] The present application relates to the technical field of temperature measurement, and in particular to a method for accurately predicting the temperature of a shoe-making oven. Background Art

[0002] A shoemaking oven is a device used in the shoemaking industry to heat and shape materials such as uppers and soles. In the shoemaking process, the use of an oven can ensure that the upper and sole materials reach the appropriate temperature before bonding, thereby improving the bond strength and durability of the finished shoe. The working principle of a shoemaking oven is usually based on hot air circulation to ensure even heat distribution. Predicting the temperature of a shoemaking oven can help reduce overheating and uneven heating. By monitoring temperature changes, equipment failure and damage caused by abnormal temperature can also be prevented, reducing downtime and maintenance costs.

[0003] Existing temperature prediction methods include prediction through time series models, modeling the historical temperature data of the shoe-making oven through machine learning or deep learning, and then predicting the temperature of the shoe-making oven at the next moment. These methods can roughly predict the temperature of the shoe-making oven, but do not take into account the uneven temperature distribution in the shoe-making oven and the influence of ambient temperature, resulting in inaccurate prediction results of the shoe-making oven temperature. Summary of the invention

[0004] In order to solve the above technical problems, a method for accurately predicting the temperature of a shoe-making oven is provided to solve the existing problems.

[0005] The solution to the technical problem of the present application is to provide a method for accurately predicting the temperature of a shoemaking oven, comprising the following steps:

[0006] The temperature of the front section, middle section and rear section of the oven at each moment and the ambient temperature at each moment are obtained through the temperature sensor;

[0007] According to the correlation between the temperatures of the temperature sensors at all times in the front section, the middle section and the back section of the oven, a first similarity degree, a second similarity degree and a third similarity degree are determined, and the temperature correlation degree is obtained after fusion;

[0008] According to the local temperature change trend of the temperature sensor at each moment in the front section, middle section and rear section of the oven and the abnormality of the temperature change, the corrected oven temperature at each moment is obtained;

[0009] Determining an oven temperature correlation index according to the correlation between the corrected oven temperature and the ambient temperature at each moment and the temperature correlation degree;

[0010] Based on the oven temperature correlation index, the gamma parameter of the kernel function in the support vector regression model is determined to predict the temperature of the shoemaking oven.

[0011] Preferably, the determining of the first degree of similarity, the second degree of similarity and the third degree of similarity comprises:

[0012] The correlation between the temperature of the temperature sensor at all times in the front section of the oven and the temperature at all times in the middle section of the oven is recorded as the first similarity degree;

[0013] The second similarity is obtained by using the same method as the first similarity for the temperature of the temperature sensor at all times in the front and rear sections of the oven. Accordingly, the third similarity is obtained for the temperature of the middle and rear sections at all times.

[0014] Preferably, the temperature correlation degree is a forward fusion of the first similarity degree, the second similarity degree and the third similarity degree.

[0015] Preferably, obtaining the corrected oven temperature at each moment includes:

[0016] The temperatures at multiple moments in the neighborhood of each moment in the middle section of the oven are combined into a local temperature sequence;

[0017] According to the degree of change of the temperature trend at each moment in the local temperature sequence, the temperature jump point of the middle section of the oven is obtained;

[0018] The temperature jump points of the front section and the back section of the oven are obtained respectively by the same method as the temperature jump point of the middle section of the oven;

[0019] According to the temperature jump point, the corrected temperature at each moment of the front section, middle section and rear section of the oven is obtained;

[0020] The average of the corrected temperatures at the same time in the front, middle and rear sections of the oven is taken as the oven temperature at each moment.

[0021] Preferably, the method for obtaining the temperature jump point is:

[0022] Determine the oven temperature jump factor at each moment in the middle section of the oven according to the temperature change difference of the local temperature sequence at each moment in the middle section of the oven and at the next moment;

[0023] The average value of the oven temperature jump factor at all moments in the middle section of the oven is used as the temperature jump threshold;

[0024] According to the change trend between the temperature at each moment and the next moment in the local temperature sequence, the temperature change rate at each moment in the local temperature sequence is obtained;

[0025] If the oven temperature jump factor is greater than the temperature jump threshold, there is a temperature jump point in the local temperature sequence at the next moment after the corresponding moment, wherein the temperature corresponding to the moment with the largest temperature change rate in the local temperature sequence is recorded as the temperature jump point.

[0026] Preferably, the method for obtaining the corrected temperature at each moment of the front section, middle section and rear section of the oven is:

[0027] All the obtained temperature jump points are clustered using a clustering algorithm, and outliers are recorded as abnormal temperature jump points;

[0028] The abnormal temperature jump points in the local temperature sequence are eliminated, and the local temperature sequence is filled with the interpolation filling method to obtain the corrected temperature at each moment in the middle section of the oven;

[0029] For the temperature at each moment in the front section and the back section of the oven, the same method as the correction temperature at each moment in the middle section of the oven is used to obtain the correction temperature at each moment in the front section and the back section of the oven.

[0030] Preferably, the step of determining the oven temperature jump factor at each moment in the middle section of the oven comprises:

[0031] The average value of the temperature at all times in the local temperature sequence is used as the local temperature value at each time in the middle section of the oven;

[0032] The oven temperature jump factor at each moment in the middle section of the oven is determined based on the difference between the local temperature value at each moment in the middle section of the oven and the next moment.

[0033] Preferably, the determining of the oven temperature correlation index comprises:

[0034] The degree of temperature similarity is calculated based on the correlation between the corrected oven temperature in each period and the ambient temperature in the corresponding period;

[0035] Based on the temperature similarity, determining an oven temperature environment influence coefficient, wherein the oven temperature environment influence coefficient is positively correlated with all the temperature similarities;

[0036] Based on the oven temperature environmental impact coefficient and the temperature correlation degree, an oven temperature correlation index is determined, and the oven temperature correlation index is positively correlated with the oven temperature environmental impact coefficient and the temperature correlation degree, respectively.

[0037] Preferably, the method for calculating the temperature similarity is:

[0038] All the remaining temperature jump points after eliminating the abnormal temperature jump point are recorded as non-abnormal temperature jump points;

[0039] All non-abnormal temperature jump points in the middle section of the oven are numbered from 1 in chronological order;

[0040] The temperatures at all moments between the odd-numbered non-abnormal temperature jump point and the next non-abnormal temperature jump point form each middle-segment temperature sequence;

[0041] According to the temperatures at all times in each middle temperature sequence, all corrected oven temperatures and ambient temperatures at the corresponding times are extracted to form each oven temperature sequence and each ambient temperature sequence;

[0042] The same method as the first similarity degree is used to calculate the correlation between each oven temperature sequence and the ambient temperature sequence of the corresponding time period, which is recorded as the temperature similarity degree.

[0043] Preferably, the step of determining the gamma parameter of the kernel function in the support vector regression model to predict the temperature of the shoe-making oven includes:

[0044] The gamma parameter is negatively correlated with the oven temperature correlation index;

[0045] The temperature data generated by the normal operation of the shoe-making oven is used to predict the temperature of the shoe-making oven using a support vector regression model.

[0046] This application has at least the following beneficial effects:

[0047] The present application analyzes the correlation between the temperatures in different areas of the oven to determine the first similarity, the second similarity and the third similarity, and obtains the temperature correlation degree. The beneficial effect is that the temperature changes in different areas of the oven are taken into account, and the uneven temperature distribution in the oven is analyzed, which is conducive to the accurate prediction of the temperature. Secondly, according to the local temperature change trend at each moment in the front, middle and rear sections of the oven and the abnormal degree of temperature change, the temperature jump point and the abnormal temperature jump point are obtained to obtain the corrected oven temperature at each moment. The beneficial effect is that the abnormal situation in the collected temperature is taken into account to reduce the prediction error. The oven temperature environmental influence coefficient is calculated to determine the oven temperature correlation index. The beneficial effect is that the influence of the environment on the oven temperature is taken into account, and the influence of environmental factors is taken into account in the prediction process. The gamma parameter of the kernel function of the support vector regression model is adjusted by the oven temperature correlation index to predict the temperature of the shoemaking oven. The beneficial effect is that the gamma parameter can be adjusted according to the ambient temperature and the temperature of different areas, so that the prediction result of the oven temperature is more accurate, thereby improving the accuracy of the shoemaking oven temperature measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The following is a further detailed description of a method for accurately predicting the temperature of a shoe-making oven of the present application in conjunction with the accompanying drawings.

[0049] Figure 1 A flow chart of the steps of a method for accurately predicting the temperature of a shoe-making oven provided in this application;

[0050] Figure 2 A flowchart of the steps of the method for obtaining the corrected oven temperature at each moment provided in the present application;

[0051] Figure 3 A flowchart of the steps of the method for obtaining the oven temperature correlation index provided in this application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clear, the following is a further detailed description of a temperature accurate prediction method of a shoe-making oven proposed in the present application in combination with the accompanying drawings and implementation examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0054] See also Figure 1 , which shows a flow chart of the steps of a method for accurately predicting the temperature of a shoe-making oven provided by an embodiment of the present application, the method comprising the following steps:

[0055] Step 1: Obtain the temperature of the front section, middle section and rear section of the oven at each moment and the ambient temperature at each moment through a temperature sensor.

[0056] Due to the different installation positions of the heating elements in the shoemaking oven, the temperature distribution in different areas of the oven will be different as the working time increases. Although there is an air circulation system to circulate the hot air inside the oven to ensure uniform heat distribution, the area close to the heating element will still be slightly higher than other areas, and the area where the shoe material passes will have a slightly lower temperature due to the absorption of heat. The temperatures in different areas of the oven are different, but there is a large correlation between the temperature data. Although the temperature differences in different areas are small, the mutual influence between them will have a certain impact on the temperature prediction.

[0057] Based on the above analysis, temperature sensors are installed in different areas of the oven. The sensors are installed at key locations where the temperature is easily affected. Temperature sensors are also installed outside the oven to collect ambient temperature. The oven temperature is collected every t time interval for a collection period of T, and the temperatures of different areas collected are preprocessed.

[0058] Preferably, as an embodiment of the present application, temperature sensors are installed at the front section, middle section, and rear section of the oven, respectively. The collection time interval is 1s, the collection time is 0.5h, the temperatures collected by different temperature sensors are used, and the missing values ​​are filled in for all the obtained temperatures using a linear interpolation method to obtain the temperatures at each moment in the front section, middle section, and rear section of the oven and the ambient temperature at each moment, respectively. Among them, linear interpolation is a well-known technology and the specific process will not be repeated here.

[0059] It is understandable that the linear interpolation method is used in the missing value filling. As other implementation methods, the implementer can use other missing value filling methods in the prior art for processing, such as neighbor interpolation, cubic spline interpolation, etc. This embodiment does not impose any special restrictions on this.

[0060] Step 2, according to the correlation between the temperatures of the temperature sensors at all times in the front section, middle section and back section of the oven, determine the first similarity, the second similarity and the third similarity, and obtain the temperature correlation degree after fusion.

[0061] The heating element is one of the key components in the shoe oven. It is responsible for generating heat. The heating element in this embodiment is installed in the middle section of the shoe oven. The temperature near the middle section of the shoe oven is higher, while the temperature of the front section and the rear section is lower than that of the middle section due to the heat absorption of the shoe material. Although the temperatures of the three areas in the shoe oven are different, they are in the same shoe oven and are close to each other. Therefore, the temperatures of different areas in the shoe oven have a certain correlation. Finding the correlation between them is the first step to achieve accurate temperature prediction. The specific method is as follows:

[0062] The correlation between the temperature of the temperature sensor at all times in the front section of the oven and the temperature at all times in the middle section of the oven is recorded as the first similarity degree;

[0063] Preferably, as an embodiment of the present application, the Pearson correlation coefficient between the temperature at all times in the front section of the oven and the temperature at all times in the middle section of the oven is used as the first correlation coefficient; the Spearman correlation coefficient between the temperature at all times in the front section of the oven and the temperature at all times in the middle section of the oven is used as the second correlation coefficient, and the absolute value of the mean of the first correlation coefficient and the second correlation coefficient is used as the first similarity degree.

[0064] Preferably, as another embodiment of the present application, the Kendall rank correlation coefficient between the temperature at all times in the front section of the oven and the temperature at all times in the middle section of the oven is used as the second correlation coefficient.

[0065] Furthermore, the Pearson correlation coefficient is mainly applicable to the case where the two samples are linear, while the Spearman correlation coefficient is applicable to nonlinear samples or the case where pulse interference occurs in the samples. Generally, the temperature changes in different areas of the oven are linearly related, and they increase or decrease at the same time over time. However, due to a sensor failure or when drying special materials, the temperature changes are no longer linearly related. At this time, the correlation between the temperatures of different areas in the oven is more complicated, so the above two correlation coefficients are combined for quantitative analysis.

[0066] The second similarity degree is obtained by using the same method as the first similarity degree for the temperature at all times in the front and rear sections of the shoemaking oven, and accordingly, the third similarity degree is obtained for the temperature at all times in the middle and rear sections.

[0067] It should be noted that the first similarity degree represents the correlation between the temperature changes in different areas of the shoe oven, and the support vector regression model can predict the temperature value of the shoe oven through this correlation. The greater the first similarity degree, the higher the temperature correlation between the two areas, and the more important the temperature data of the two areas are for the training of the support vector regression model.

[0068] forwardly fusing the first similarity degree, the second similarity degree and the third similarity degree as the temperature correlation degree;

[0069] It is understandable that forward fusion can be expressed as a multiplication relationship, an addition relationship, or the idempotence of an exponential function, which is determined by actual applications and is not specifically limited in this application.

[0070] Preferably, as an embodiment of the present application, the temperature correlation degree is the product of the first similarity degree, the second similarity degree and the third similarity degree.

[0071] Preferably, as another embodiment of the present application, the average of the first similarity degree, the second similarity degree and the third similarity degree is used as the temperature correlation degree.

[0072] It should be noted that the greater the degree of temperature correlation, the greater the impact on the shoemaking oven temperature prediction results, while the ambient temperature has a relatively small impact.

[0073] At this point, the temperature correlation degree is obtained.

[0074] Step 3, obtaining the corrected oven temperature at each moment according to the local temperature change trend of the temperature sensor at each moment in the front section, middle section and rear section of the oven and the abnormality of the temperature change.

[0075] After the shoe oven starts working, its internal temperature will quickly rise to the working temperature under the action of the heating element. After this batch of shoe materials is dried, it stops working and the internal temperature drops to the ambient temperature. This change process reflects the seasonal trend of the shoe oven temperature data, which is important for the prediction of the shoe oven temperature. The time point when the shoe oven temperature changes rapidly is recorded as the temperature jump point. In addition, when an abnormal situation occurs, such as a temperature sensor failure, additional temperature jump points will be generated in the temperature data. The temperature data between these temperature jump points will have a negative impact on the temperature prediction of the shoe oven. These temperature jump points are recorded as abnormal temperature jump points.

[0076] Based on the above analysis, the temperature jump point is obtained through the temperature at each moment in the middle section of the shoe-making oven. The specific method is as follows:

[0077] The temperatures at multiple moments in the neighborhood of each moment in the middle section of the oven are combined into a local temperature sequence;

[0078] Preferably, as an embodiment of the present application, the temperatures of 60 moments in the neighborhood of each moment in the middle section of the oven are selected to form a local temperature sequence, wherein it should be noted that if the length of the local temperature sequence of the temperatures at each moment in the middle section of the oven is insufficient, the sequence is padded with a filling value of 0.

[0079] According to the change trend between the temperature at each moment and the next moment in the local temperature sequence, the temperature change rate at each moment in the local temperature sequence is obtained;

[0080] Preferably, as an embodiment of the present application, the difference between the temperature at each moment in the local temperature sequence and the temperature at the next moment is recorded as the local temperature difference at each moment; the time length between the temperature at each moment in the local temperature sequence and the temperature at the next moment is recorded as the time length at each moment; the temperature change rate at each moment in the local temperature sequence is the ratio of the local temperature difference to the time length.

[0081] Preferably, as another embodiment of the present application, the difference between the temperature at each moment in the local temperature sequence and the temperature at the next moment is used as the temperature change rate at each moment in the local temperature sequence.

[0082] It should be noted that the temperature change rate is no longer calculated for the temperature at the last moment in the local temperature sequence, but the temperature value at the last moment in the local temperature sequence at each moment is used as the temperature value at the first moment in the local temperature sequence at the next moment.

[0083] The average value of the temperature at all times in the local temperature sequence is used as the local temperature value at each time in the middle section of the oven;

[0084] Determine the oven temperature jump factor at each moment in the middle section of the oven according to the difference between the local temperature value at each moment in the middle section of the oven and the next moment;

[0085] Preferably, as an embodiment of the present application, the difference between the local temperature value at each moment in the middle section of the oven and the local temperature value at the next moment is used as the oven temperature jump factor at each moment in the middle section of the oven.

[0086] The average value of the oven temperature jump factor at all moments in the middle section of the oven is used as the temperature jump threshold;

[0087] If the oven temperature jump factor is greater than the temperature jump threshold, there is a temperature jump point in the local temperature sequence at the moment after the corresponding moment in the middle section of the oven, wherein the temperature corresponding to the moment with the largest temperature change rate in the local temperature sequence is recorded as the temperature jump point;

[0088] It should be noted that the length of the local temperature sequence depends on the power of the shoemaking oven heating element. The greater its power, the shorter the time it takes to heat to the working temperature, and the shorter the time neighborhood range should be, because a neighborhood range that is too long may result in multiple temperature jump points. Secondly, the existence of the temperature jump point is determined by the oven temperature jump factor. The larger the oven temperature jump factor, the faster the temperature change between the two adjacent sequences, and the more likely it is that there will be a temperature jump point in the local temperature sequence at the next moment. If the temperature change rate is the largest in a sequence, the temperature change is the fastest.

[0089] All the obtained temperature jump points are clustered using a clustering algorithm, and outliers are recorded as abnormal temperature jump points;

[0090] The abnormal temperature jump points in the local temperature sequence are eliminated, and the local temperature sequence is filled with the interpolation filling method to obtain the corrected temperature at each moment in the middle of the oven;

[0091] Preferably, as an embodiment of the present application, the DBSCAN algorithm is used for clustering, wherein the neighborhood radius of the DBSCAN algorithm is preset to 0.5, and the minimum number of sample points is 5; secondly, the cubic spline interpolation method is used for filling, wherein the DBSCAN algorithm and cubic spline interpolation are well-known technologies, and the specific process will not be repeated here.

[0092] It can be understood that this embodiment uses the DBSCAN algorithm for clustering and the cubic spline interpolation method for filling. As other implementation methods, the implementer can adopt other methods of the prior art, for example, using K-means clustering for clustering, using linear interpolation or Lagrange interpolation method for filling, etc. This embodiment does not impose any special restrictions on this.

[0093] For the temperature at each moment in the front section and the rear section of the oven, the same method as the correction temperature at each moment in the middle section of the oven is used to obtain the correction temperature at each moment in the front section and the rear section of the oven;

[0094] The average of the corrected temperatures at the same time in the front section, middle section and back section of the oven is taken as the corrected oven temperature at each time.

[0095] Further, the step flow chart of the method for obtaining the corrected oven temperature at each moment provided in the present application is as follows: Figure 2 shown.

[0096] At this point, the corrected oven temperature at each moment is obtained.

[0097] Step 4: determining an oven temperature correlation index according to the correlation between the corrected oven temperature and the ambient temperature at each moment and the temperature correlation degree.

[0098] Furthermore, the influence of ambient temperature on the oven temperature is analyzed. Since the ambient temperature is lower in winter, it takes longer for the oven to reach the working temperature and it is more likely to experience sudden temperature changes. In summer, the opposite is true. Therefore, the influence of ambient temperature on the prediction of shoemaking oven temperature needs to be considered. The specific method for determining the oven temperature correlation index is as follows:

[0099] All the remaining temperature jump points after eliminating the abnormal temperature jump point are recorded as non-abnormal temperature jump points;

[0100] All non-abnormal temperature jump points in the middle section of the oven are numbered from 1 in chronological order;

[0101] The temperatures at all moments between the odd-numbered non-abnormal temperature jump point and the next non-abnormal temperature jump point form each middle-segment temperature sequence;

[0102] According to the temperatures at all times in each middle temperature sequence, all corrected oven temperatures and ambient temperatures at the corresponding times are extracted to form each oven temperature sequence and each ambient temperature sequence;

[0103] Using the same method as the first similarity degree, calculate the correlation between each oven temperature sequence and the ambient temperature sequence of the corresponding time period, and record it as the temperature similarity degree;

[0104] The oven temperature environmental impact coefficient is positively correlated with all the temperature similarities;

[0105] It can be understood that a positive relationship means that the dependent variable increases as the independent variable increases, and decreases as the independent variable decreases. The specific relationship can be a multiplication relationship, an addition relationship, or the idempotence of an exponential function, which is determined by actual application and this application does not impose any special restrictions on this.

[0106] Preferably, as an embodiment of the present application, the average of all the temperature similarities is used as the oven temperature environment impact coefficient.

[0107] It should be noted that the larger the oven temperature environmental impact coefficient is, the greater the correlation between the oven temperature and the ambient temperature during the mutation stage, and the greater the influence of the ambient temperature is. On the contrary, the smaller the influence of the ambient temperature is.

[0108] Based on the oven temperature environmental impact coefficient and the temperature correlation degree, determining an oven temperature correlation index, wherein the oven temperature correlation index is positively correlated with the oven temperature environmental impact coefficient and the temperature correlation degree, respectively;

[0109] Preferably, as an embodiment of the present application, the sum of the oven temperature environmental impact coefficient and the temperature correlation degree is used as the oven temperature correlation index.

[0110] Preferably, as another embodiment of the present application, the product of the oven temperature environmental impact coefficient and the temperature correlation degree is used as the oven temperature correlation index.

[0111] It should be noted that the larger the value of the oven temperature correlation index is, the greater the correlation between the data related to the shoemaking oven temperature prediction is, and vice versa, the smaller the correlation is.

[0112] Further, the step flow chart of the method for obtaining the oven temperature correlation index provided in the present application is as follows: Figure 3 shown.

[0113] At this point, the oven temperature correlation index is obtained.

[0114] Step 5: Based on the oven temperature correlation index, determine the gamma parameter of the kernel function in the support vector regression model to predict the temperature of the shoemaking oven.

[0115] In order to better describe the relationship between the temperature of the shoemaking oven and the temperature of different areas in the oven and the ambient temperature, and taking into account the multidimensionality, nonlinearity and interference of abnormal conditions of the parameters, the oven temperature is predicted by a support vector regression model based on the kernel function. The support vector regression model introduces an insensitive loss function on the basis of the support vector machine. Its idea is no longer to find the optimal classification surface to separate the two types of samples, but to find the optimal classification surface so that the error of the trained samples from the optimal classification surface is minimized. Among them, the gamma parameter of the kernel function has a greater impact on the prediction effect. It is necessary to adjust the gamma parameter of the kernel function in the support vector regression model through the oven temperature association index. The specific method is:

[0116] The gamma parameter is negatively correlated with the oven temperature correlation index;

[0117] It can be understood that a negative relationship indicates that the dependent variable decreases as the independent variable increases, and decreases as the independent variable increases. The specific relationship can be a subtraction relationship, a division relationship, a negative exponential idempotence, etc., which is determined by actual application and this application does not impose any special restrictions on this.

[0118] Preferably, as an embodiment of the present application, the calculation result of an exponential function with a natural constant as the base and the inverse of the oven temperature correlation index as the exponent is used as the gamma parameter.

[0119] It should be noted that the larger the oven temperature correlation index is, the smaller the gamma value is, and the smaller the oven temperature correlation index is, the larger the gamma value is. The more synchronized the trends between different temperature data are, the smaller the gamma parameter is needed to capture this trend and avoid overfitting; on the contrary, a larger gamma parameter value is needed to avoid underfitting.

[0120] The front section temperature, middle section temperature, rear section temperature and ambient temperature of the shoe-making oven are used as inputs, and the radial basis function is used to train the support vector regression model.

[0121] The temperature data generated by the normal operation of the shoe-making oven is input into the trained model, and the temperature prediction value of the shoe-making oven is output to predict the temperature of the shoe-making oven.

[0122] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0124] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the present application. It should be pointed out that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the content of the technical solution of the present application, shall fall within the protection scope of the technical solution of the present application.

Claims

1. A method for accurately predicting the temperature of a shoe-making oven, characterized in that: The method comprises the following steps: The temperature of the front section, middle section and rear section of the oven at each moment and the ambient temperature at each moment are obtained through the temperature sensor; According to the correlation between the temperatures of the temperature sensors at all times in the front section, the middle section and the back section of the oven, a first similarity degree, a second similarity degree and a third similarity degree are determined, and the temperature correlation degree is obtained after fusion; According to the local temperature change trend of the temperature sensor at each moment in the front section, middle section and rear section of the oven and the abnormality of the temperature change, the corrected oven temperature at each moment is obtained; Determining an oven temperature correlation index according to the correlation between the corrected oven temperature and the ambient temperature at each moment and the temperature correlation degree; Based on the oven temperature correlation index, the gamma parameter of the kernel function in the support vector regression model is determined to predict the temperature of the shoemaking oven. Determining the oven temperature correlation index comprises: The degree of temperature similarity is calculated based on the correlation between the corrected oven temperature in each period and the ambient temperature in the corresponding period; Based on the temperature similarity, determining an oven temperature environment influence coefficient, wherein the oven temperature environment influence coefficient is positively correlated with all the temperature similarities; Based on the oven temperature environmental impact coefficient and the temperature correlation degree, determining an oven temperature correlation index, wherein the oven temperature correlation index is positively correlated with the oven temperature environmental impact coefficient and the temperature correlation degree, respectively; The method of determining the gamma parameter of the kernel function in the support vector regression model to predict the temperature of the shoe-making oven includes: The gamma parameter is negatively correlated with the oven temperature correlation index; The temperature data generated by the normal operation of the shoe-making oven is used to predict the temperature of the shoe-making oven using a support vector regression model.

2. A method for accurately predicting the temperature of a shoemaking oven as claimed in claim 1, characterized in that: The determining of the first similarity degree, the second similarity degree and the third similarity degree comprises: The correlation between the temperature of the temperature sensor at all times in the front section of the oven and the temperature at all times in the middle section of the oven is recorded as the first similarity degree; The second similarity is obtained by using the same method as the first similarity for the temperature of the temperature sensor at all times in the front and rear sections of the oven. Accordingly, the third similarity is obtained for the temperature of the middle and rear sections at all times.

3. The method for accurately predicting the temperature of a shoe-making oven according to claim 1, characterized in that: The temperature correlation degree is a forward fusion of the first similarity degree, the second similarity degree, and the third similarity degree.

4. A method for accurately predicting the temperature of a shoe-making oven as claimed in claim 2, characterized in that: The step of obtaining the corrected oven temperature at each moment includes: The temperatures at multiple moments in the neighborhood of each moment in the middle section of the oven are combined into a local temperature sequence; According to the degree of change of the temperature trend at each moment in the local temperature sequence, the temperature jump point of the middle section of the oven is obtained; The temperature jump points of the front section and the back section of the oven are obtained respectively by the same method as the temperature jump point of the middle section of the oven; According to the temperature jump point, the corrected temperature at each moment of the front section, middle section and rear section of the oven is obtained; The average of the corrected temperatures at the same time in the front, middle and rear sections of the oven is taken as the oven temperature at each moment.

5. A method for accurately predicting the temperature of a shoe-making oven as claimed in claim 4, characterized in that: The method for obtaining the temperature jump point is: Determine the oven temperature jump factor at each moment in the middle section of the oven according to the temperature change difference of the local temperature sequence at each moment in the middle section of the oven and at the next moment; The average value of the oven temperature jump factor at all moments in the middle section of the oven is used as the temperature jump threshold; According to the change trend between the temperature at each moment and the next moment in the local temperature sequence, the temperature change rate at each moment in the local temperature sequence is obtained; If the oven temperature jump factor is greater than the temperature jump threshold, there is a temperature jump point in the local temperature sequence at the next moment after the corresponding moment, wherein the temperature corresponding to the moment with the largest temperature change rate in the local temperature sequence is recorded as the temperature jump point.

6. A method for accurately predicting the temperature of a shoe-making oven as claimed in claim 4, characterized in that: The method for obtaining the corrected temperature at each moment of the front section, middle section and rear section of the oven is: All the obtained temperature jump points are clustered using a clustering algorithm, and outliers are recorded as abnormal temperature jump points; The abnormal temperature jump points in the local temperature sequence are eliminated, and the local temperature sequence is filled with the interpolation filling method to obtain the corrected temperature at each moment in the middle section of the oven; For the temperature at each moment in the front section and the back section of the oven, the same method as the correction temperature at each moment in the middle section of the oven is used to obtain the correction temperature at each moment in the front section and the back section of the oven.

7. A method for accurately predicting the temperature of a shoe-making oven as claimed in claim 5, characterized in that: The step of determining the oven temperature jump factor at each moment in the middle section of the oven comprises: The average value of the temperature at all times in the local temperature sequence is used as the local temperature value at each time in the middle section of the oven; The oven temperature jump factor at each moment in the middle section of the oven is determined based on the difference between the local temperature value at each moment in the middle section of the oven and the next moment.

8. A method for accurately predicting the temperature of a shoe-making oven as claimed in claim 4, characterized in that: The calculation method of the temperature similarity is: All the remaining temperature jump points after eliminating the abnormal temperature jump point are recorded as non-abnormal temperature jump points; All non-abnormal temperature jump points in the middle section of the oven are numbered from 1 in chronological order; The temperatures at all moments between the odd-numbered non-abnormal temperature jump point and the next non-abnormal temperature jump point form each middle-segment temperature sequence; According to the temperatures at all times in each middle temperature sequence, all corrected oven temperatures and ambient temperatures at the corresponding times are extracted to form each oven temperature sequence and each ambient temperature sequence; The same method as the first similarity degree is used to calculate the correlation between each oven temperature sequence and the ambient temperature sequence of the corresponding time period, which is recorded as the temperature similarity degree.

Images