A product data anomaly prediction method, system, device and storage medium

Abstract

The application discloses a product data anomaly prediction method, system and device and a storage medium, and applies to the field of data anomaly prediction. The average value and the dispersion degree index corresponding to the current sampling data are obtained, the preset relationship corresponding to each historical sampling data is obtained according to the average value, the dispersion degree index and the defect probability of the next output product corresponding to the historical sampling data, the probability that the next output product corresponding to the current sampling data has a defect is obtained according to the average value and the dispersion degree index corresponding to the current sampling data and the preset relationship, the two change trends of the average value and the dispersion degree index are combined, the abnormal trend is more clearly monitored, the detection result is obtained according to the change trend, the possibility that the next product has a defect is obtained, prediction is performed, and the detection is more accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data anomaly prediction, in particular to a product data anomaly prediction method, system and device, and a storage medium. BACKGROUND

[0002] With the development of industrial production, in the process of industrial production, it is necessary to scientifically distinguish the random fluctuation and abnormal fluctuation of product quality in the production process, so as to obtain the abnormal trend in the production process, so that the production management personnel can take timely measures to eliminate the abnormality and restore the stability of the process, so as to achieve the purpose of improving and controlling the quality.

[0003] The commonly used method for judging abnormal trend is average-range (X-R) control chart and average-standard deviation (X-S) control chart, etc. By observing the change trend of X and the change trend of R / S, the abnormal trend can be judged, but this method can only observe one-dimensional change based on time sequence, and can only monitor the average value or the dispersion degree of data, and cannot combine the two change trends to more clearly monitor the abnormal trend, nor can it obtain the good or bad of the detection result through the change trend.

[0004] In view of the above-mentioned technology, it is a problem for those skilled in the art to find a method to solve the above-mentioned technical problems. SUMMARY

[0005] The purpose of the present application is to provide a product data anomaly prediction method, system, device and storage medium, which is used to obtain the average value and dispersion degree index corresponding to the current sampling data, obtain the preset relationship corresponding to the average value, dispersion degree index and corresponding defect probability of the next product output of each historical sampling data, and obtain the probability of the next product output of the current sampling data. By the above method, the average value and dispersion degree index can be combined to more clearly monitor the abnormal trend, and the good or bad of the detection result can be obtained through the change trend, the possibility of the next product defect can be obtained, the prediction can be made, and the detection can be more accurate.

[0006] To solve the above technical problems, the present application provides a product data anomaly prediction method, comprising:

[0007] Obtaining the target average value and target dispersion degree index corresponding to the current sampling data;

[0008] Calling the preset relationship obtained by the historical sampling data; wherein the historical sampling data includes historical average value, historical dispersion degree index, and the preset relationship is the defect probability of the next product corresponding to the historical average value and historical dispersion degree index.

[0009] The target defect probability corresponding to the target average value and the target dispersion degree index is determined through the preset relationship.

[0010] Preferably, the method for obtaining the preset relationship through historical sampling data comprises:

[0011] Obtaining the corresponding historical average value and the historical dispersion degree index in the historical sampling data;

[0012] Dividing the average value and the dispersion degree index corresponding to the historical sampling data according to the preset region respectively;

[0013] The value of the effective sampling sample quantity determined in each preset region is a first preset value.

[0014] Preferably, the method for obtaining the preset relationship through historical sampling data further comprises:

[0015] Obtaining the quantity of all products and the quantity of defective products distributed in each preset region respectively;

[0016] Obtaining the probability that the next product output in each preset region is a defective product according to the corresponding relationship between the quantity of all products and the quantity of defective products.

[0017] Preferably, the value of the effective sampling sample quantity determined in each preset region is a first preset value, which comprises:

[0018] If the quantity of historical sampling samples is greater than the first preset quantity, the first preset value adopts a fixed value;

[0019] If the quantity of historical sampling samples is not greater than the first preset quantity, the first preset value is calculated according to a preset parameter.

[0020] Preferably, the preset parameter comprises:

[0021] The confidence, the confidence interval and the total sampling sample quantity are preset.

[0022] Preferably, the preset relationship comprises:

[0023] If the data quantity in any region in the preset region is greater than four times of the first preset value, the average value and the dispersion degree index corresponding to any one data are obtained as a target point;

[0024] According to the target point, any region is divided into four regions;

[0025] The value of the effective sampling sample quantity in each of the four regions is greater than the first preset value.

[0026] Preferably, after obtaining the probability of the next product output in each preset area being a defective product according to the correspondence between the number of all products and the number of defective products, the method further comprises:

[0027] A defective probability table is formed according to the average value and the dispersion degree index in each preset area and the probability of defective products corresponding thereto.

[0028] The application also provides a product data anomaly prediction system, comprising:

[0029] The first obtaining module is configured to obtain a target average value and a target dispersion degree index corresponding to the current sampling data;

[0030] The calling module is configured to call a preset relationship obtained through historical sampling data, wherein the historical sampling data comprises a historical average value and a historical dispersion degree index, and the preset relationship is a defective probability of the next product corresponding to the historical average value and the historical dispersion degree index;

[0031] The determining module is configured to determine a target defective probability corresponding to the target average value and the target dispersion degree index through the preset relationship.

[0032] The application also provides a product data anomaly prediction device, comprising a memory for storing a computer program;

[0033] The processor is configured to implement the steps of the product data anomaly prediction method as described above when executing the computer program.

[0034] The application also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the product data anomaly prediction method as described above.

[0035] The product data anomaly prediction method provided by the application can combine the two change trends of the average value and the dispersion degree index, more clearly monitor the abnormal trend, and obtain the detection result, the possibility of the next product being defective, and the prediction, so that the detection is more accurate.

[0036] The product data anomaly prediction system, device and medium provided by the application have the same beneficial effects as the product data anomaly prediction method. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings described below are only some of the embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0038] Figure 1 A flowchart of a product data anomaly prediction method provided by the present application;

[0039] Figure 2 A structural diagram of a product data anomaly prediction system provided by the present application;

[0040] Figure 3 A structural diagram of a product data anomaly prediction device provided by another embodiment of the present application. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present application.

[0042] The core of the present application is to provide a product data anomaly prediction method, system, device and storage medium, which is used for obtaining the average value and the dispersion degree index corresponding to the current sampling data, obtaining the preset relationship according to the average value, the dispersion degree index and the defect probability of the next product corresponding to each historical sampling data, and obtaining the probability of the next product corresponding to the current sampling data from the average value and the dispersion degree index corresponding to the current sampling data and the preset relationship. Through the above method, the average value and the dispersion degree index can be combined together, the abnormal trend can be more clearly monitored, and the detection result can be obtained according to the change trend, the possibility of the next product from the defect can be obtained, the prediction can be made, and the detection can be more accurate.

[0043] In order to make the person skilled in the art better understand the present application, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0044] The present application provides a product data anomaly prediction method, as shown in Figure 1 Figure 1 A flowchart of a product data anomaly prediction method provided by the present application, the method comprises:

[0045] ​S10: obtaining a target average value and a target dispersion degree index corresponding to the current sampling data;

[0046] It should be noted that the application does not limit the type of monitored product, and the data can be but is not limited to the data of the product in industrial production. When monitoring the quality of the product, a variety of data indexes can be used for monitoring, and the average value of the sampling data, the standard deviation of the sampling data, and the range can be obtained, but the application embodiments are not limited to this. As long as the index can be used to represent the dispersion degree of the data, it can be used. The application embodiments do not specifically limit the number of sampling samples when obtaining the average value and the dispersion degree index corresponding to the current sampling data.

[0047] S11: calling a preset relationship obtained through historical sampling data; wherein the historical sampling data includes a historical average value, a historical dispersion degree index, and the preset relationship is a defect probability of the next product corresponding to the historical average value and the historical dispersion degree index;

[0048] It should be noted that when obtaining the preset relationship through the historical sampling data, the number of historical sampling samples is not limited, but in order to ensure the reliability of the data, the number of sampling samples should be sufficient as much as possible. The application embodiments do not specifically limit the form of the described preset relationship, which can be in the form of a table, a curve, a statistical chart, or a formula, and the like. The specific selection can be made according to the specific situation. The historical sampling data can include but is not limited to the average value corresponding to the historical sample and the historical dispersion degree, and the preset relationship can include but is not limited to the relationship between the historical average value, the historical dispersion degree, and the defect probability of the next product. In addition, the application embodiments do not specifically limit the reference sample and the calculation method of the defect probability of the next product.

[0049] S12: determining a target defect probability corresponding to the target average value and the target dispersion degree index through the preset relationship.

[0050] It should be noted that the target average value and the dispersion degree index are input into the preset relationship to obtain the corresponding defect probability, wherein the defect probability can be but is not limited to being observed in regions, and the application embodiments do not specifically limit this.

[0051] It can be seen that the method provided in the embodiment can combine the average value and the dispersion degree index to more clearly monitor the abnormal trend, and can obtain the detection result, the possibility of the defect of the next product, and the prediction, so that the detection is more accurate.

[0052] On the basis of the above-mentioned embodiments, the application provides a preferred embodiment, and the method for obtaining the preset relationship through the historical sampling data includes:

[0053] obtaining a corresponding historical average value and a historical dispersion degree index in the historical sampling data;

[0054] respectively dividing the average value and the dispersion degree index corresponding to the historical sampling data according to preset regions;

[0055] The effective sampling sample quantity determined in each preset region is a first preset value.

[0056] It should be noted that the historical average value and the historical dispersion degree index of the historical sampling sample are historical sampling data, and the historical sampling data can also include other indexes. The present embodiment does not make specific limitations on the historical average value and the historical dispersion degree index in the historical sampling data. The historical average value and the historical dispersion degree index in the historical sampling data are divided according to regions. The present embodiment does not limit the size of the region division, as long as the sample quantity in each region is effective. The present embodiment does not make specific limitations on the size of the sample quantity and the calculation method. The sample quantity reaching the preset value can make the data in each region effective and as accurate as possible. In addition, each preset region can be fixed or can change in real time with data. It can also be updated once after a fixed time period. The present embodiment does not make specific limitations here.

[0057] It can be seen that the method provided in the present embodiment can combine the two change trends of the average value and the dispersion degree index, more clearly monitor the abnormal trend, and make the detection more accurate.

[0058] On the basis of the above-mentioned embodiments, the present application provides a preferred embodiment. The method for obtaining a preset relationship through historical sampling data further comprises:

[0059] respectively obtaining the quantity of all products and the quantity of defective products distributed in each preset region;

[0060] obtaining the probability that the next output product in each preset region is a defective product according to the corresponding relationship between the quantity of all products and the quantity of defective products.

[0061] It should be noted that when the historical average value and the historical dispersion degree index corresponding to the historical sampling data are divided into the forbidden area, the probability of the next product having a defect can be obtained according to each region, and the relationship between the average value and the dispersion degree index and the probability of the next product having a defect is obtained as a preset relationship for monitoring the probability of the subsequent product having a defect, wherein the probability of the defect can be obtained by, but not limited to, obtaining the number of all products falling in each region, the number of good products falling in the region, and the number of defective products falling in the region, and the probability of the product having a defect in the region is obtained by obtaining the ratio or other corresponding relationship between the number of defective products in the region and the number of all products, wherein the probability of the application embodiment is not limited to the change, and the probability can be updated with the change of the data when new data is updated.

[0062] It can be seen that the method provided in the embodiment can combine the two change trends of the average value and the dispersion degree index, more clearly monitor the abnormal trend, and obtain the detection result by the change trend, obtain the probability of the next product having a defect, and form the corresponding preset relationship, so as to obtain the possibility of the next product having a defect, and make prediction, so that the detection is more accurate.

[0063] On the basis of the above-mentioned embodiments, the application provides a preferred embodiment, and the effective sampling sample quantity determined in each preset region is a first preset value, which includes:

[0064] If the number of historical sampling samples is greater than the first preset number, the first preset value adopts a fixed value; if the number of historical sampling samples is not greater than the first preset number, the first preset value is calculated according to a preset parameter.

[0065] It should be noted that in order to ensure that the data in each preset region has high accuracy and referenceability as much as possible, the effective sample quantity in the region should reach the first preset value, wherein the effective sample quantity can be determined by calculation according to the characteristics of the historical sample data, or can be set according to the empirical value, and the application embodiment does not make specific limitation here. The application only provides a preferred embodiment, wherein the effective sample quantity can be determined according to the preset parameters such as preset confidence, confidence interval, and the number of total sampling samples, and the application embodiment does not make specific limitation here, and the effective sample quantity can also be determined according to other parameters. When the number of historical effective samples is large enough, for example, when the number of historical samples reaches more than ten, the effective sample quantity can be valued as 1070 or 1850 for rapid calculation according to the prior empirical value, and when the number of historical samples reaches other values, the effective sample quantity can be determined according to the actual situation, and the application embodiment does not make specific limitation here.

[0066] It can be seen that the method provided in the embodiment of the application can better improve the accuracy and effectiveness of the preset relationship in the region by obtaining the effective sample quantity, and when the historical effective sample quantity can reach a certain value, the effective sample quantity can be quickly calculated by using the prior experience value, the region updating speed is increased, and the product data anomaly monitoring process is more accurate.

[0067] On the basis of the above embodiment, the application provides a preferred embodiment, and the preset relationship includes:

[0068] If the data quantity in any region in the preset region is greater than four times the first preset value, the average value and the dispersion degree index corresponding to any data are obtained as the target point;

[0069] According to the target point, any region is divided into four regions;

[0070] The effective sample quantity in each of the four regions is greater than the first preset value.

[0071] It should be noted that after the current sampling data is input into the preset relationship to obtain the defect probability of the next product, the sampling data is updated to the preset relationship, the data in the preset relationship is refined, and the data quantity in each region is also refined. When the data quantity in a certain region or a certain part of the region has reached four times the effective sample quantity, the region can be divided. When the region is divided, the average value and the dispersion degree index corresponding to a certain data can be found, and the region is divided by the point. The embodiment of the application does not limit the specific division point, and only needs to ensure that the data quantity in each of the four regions after division can reach the value of the effective sample quantity.

[0072] It can be seen that the method provided in the embodiment can make the monitoring of product data anomaly more accurate and more clearly monitor the abnormal trend, and when the region becomes smaller, the probability of defect is also more refined, so that the detection is more accurate and more convincing.

[0073] On the basis of the above embodiment, the application further provides a specific embodiment, after obtaining the probability that the next product output in each preset region is a defective product according to the corresponding relationship between the number of all products and the number of defective products, the method further includes:

[0074] According to the average value and the dispersion degree index in each preset region and the probability of defective products corresponding thereto, a defect probability table is formed, as shown in Table 1, which is a defect probability prediction quick reference table provided by the application.

[0075] Table 1

[0076]

[0077] Wherein, the defect probability prediction table can also be expressed in the form of a statistical chart, it should be noted that the present application provides only a possible situation, the average value and the dispersion degree of the region is not limited to be divided as shown in table 1, can be divided according to the actual situation, and the defect probability can also change accordingly, the average value and the dispersion degree of the region can also be refined, the present application embodiment does not make specific limitation here.

[0078] Based on the functional module angle, the present application also provides a product data anomaly prediction system, as shown in Figure 2 The structure diagram of a product data anomaly prediction system provided by the present application is shown in Figure 2 The system comprises:

[0079] The first acquisition module 30 is configured to acquire the target average value and the target dispersion degree index corresponding to the current sampling data.

[0080] The calling module 31 is configured to call the preset relationship obtained through the historical sampling data, wherein the historical sampling data comprises the historical average value and the historical dispersion degree index, and the preset relationship is the defect probability of the next product corresponding to the historical average value and the historical dispersion degree index.

[0081] The determining module 32 is configured to determine the target defect probability corresponding to the target average value and the target dispersion degree index through the preset relationship.

[0082] Since the embodiments of the system part correspond to the embodiments of the method part, the embodiments of the device part are described in the description of the embodiments of the method part, which will not be described here.

[0083] The product data anomaly prediction system provided by the present embodiment corresponds to the above method, and has the same beneficial effects as the above method.

[0084] Figure 3 The structure diagram of a product data anomaly prediction device provided by another embodiment of the present application is shown in Figure 3 The product data anomaly prediction device comprises a memory 20 configured to store a computer program.

[0085] The processor 21 is configured to execute the computer program to realize the steps of the product data anomaly prediction method mentioned in the above embodiments.

[0086] The product data anomaly prediction device provided by the present embodiment can include but is not limited to smart phones, tablet computers, notebook computers or desktop computers, etc.

[0087] The processor 21 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 can be implemented in at least one of a hardware form of a digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic array (PLA), etc. The processor 21 can also include a main processor and a coprocessor. The main processor is a processor for processing data in a wake-up state, also referred to as a central processing unit (CPU). The coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 21 can be integrated with a graphics processor (GPU) for rendering and drawing content required to be displayed by a display screen. In some embodiments, the processor 21 can further include an artificial intelligence (AI) processor for processing computing operations related to machine learning.

[0088] The memory 20 can include one or more computer-readable storage media, which can be non-transitory. The memory 20 can further include a high-speed random access memory, and a nonvolatile memory such as one or more disk storage devices, flash storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201, wherein the computer program is loaded and executed by the processor 21, and can implement the related steps of the product data anomaly prediction method disclosed in any of the preceding embodiments. In addition, the resources stored by the memory 20 can further include an operating system 202 and data 203, etc., and the storage manner can be temporary storage or permanent storage. The operating system 202 can include Windows, Unix, Linux, etc. The data 203 can include, but is not limited to, data in the product data anomaly prediction method, etc.

[0089] In some embodiments, the product data anomaly prediction apparatus can further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.

[0090] Those skilled in the art can understand that the structure shown in the above embodiments does not constitute a limitation on the product data anomaly prediction apparatus, and can include more or fewer components than those shown in the above embodiments. Figure 3 The product data anomaly prediction apparatus can include more or fewer components than those shown in the above embodiments. Figure 3 The product data anomaly prediction apparatus can include more or fewer components than those shown in the above embodiments.

[0091] The product data anomaly prediction device provided by the embodiment of the present application comprises a memory and a processor. When the processor executes the program stored in the memory, the product data anomaly prediction method can be realized.

[0092] Finally, the present application also provides an embodiment corresponding to a computer readable storage medium. The computer readable storage medium stores a computer program. When the computer program is executed by a processor, the steps recorded in the above method embodiments are realized.

[0093] It can be understood that if the method in the above embodiment is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and executes all or part of the steps of the methods of the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0094] The product data anomaly prediction, system, device and storage medium provided by the present application are described in detail above. The embodiments in the specification are described in a progressive manner. Each embodiment mainly describes the differences from other embodiments. The same or similar parts of each embodiment can be referred to. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part. It should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, the present application can be improved and modified. These improvements and modifications also fall within the protection scope of the claims of the present application.

[0095] It also needs to be explained that in the present specification, the relational terms such as first and second and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

Claims

1. A method for predicting product data anomalies, characterized in that, include: Obtain the target average value and target dispersion index corresponding to the current sampled data; The system invokes a preset relationship obtained from historical sampling data; wherein the historical sampling data includes historical average values ​​and historical dispersion indicators, and the preset relationship is the defect probability of the next product corresponding to the historical average values ​​and the historical dispersion indicators. The target defect probability corresponding to the target average value and the target dispersion index is determined by the preset relationship; The method for obtaining the preset relationship through historical sampling data includes: Obtain the historical average value and the historical dispersion index corresponding to the historical sampling data; The average value and the dispersion index corresponding to the historical sampling data are respectively divided into preset regions; The number of valid samples determined in each preset region is a first preset value; The preset relationships include: If the amount of data in any region within the preset area is greater than four times the first preset value, then the average value and the dispersion index corresponding to any data point are obtained as the target point. Based on the target point, any corresponding region is divided into four regions; Among them, the number of valid samples in each of the four regions is greater than the first preset value; The method for obtaining a preset relationship through historical sampling data further includes: The quantity of all products distributed within each of the preset areas and the quantity of defective products are obtained respectively; Based on the correspondence between the quantity of all products and the quantity of defective products, the probability that the next product produced in each preset area is a defective product is obtained. A defect probability table is formed based on the average value and the dispersion index of each preset region and the probability of the defective product corresponding to them.

2. The product data anomaly prediction method according to claim 1, characterized in that, The number of valid samples determined within each of the preset regions is a first preset value, including: If the number of historical samples is greater than the first preset number, then the first preset value is a fixed value; If the number of historical samples is not greater than the first preset number, then the first preset value is calculated according to the preset parameters.

3. The product data anomaly prediction method according to claim 2, characterized in that, The preset parameters include: Preset the reliability, confidence interval, and total number of samples.

4. A product data anomaly prediction system, characterized in that, include: The first acquisition module is used to acquire the target average value and the target dispersion index in the current sampled data; The calling module is used to call a preset relationship obtained through historical sampling data; wherein, the historical sampling data includes historical average value and historical dispersion index, and the preset relationship is the defect probability of the next product corresponding to the historical average value and the historical dispersion index; The determination module is used to determine the target defect probability corresponding to the target average value and the target dispersion index through the preset relationship; Specifically, the calling module is used to obtain the historical average value and the historical dispersion index corresponding to the historical sampling data; and to divide the average value and the dispersion index corresponding to the historical sampling data into preset regions; wherein the value of the number of valid sampling samples determined in each preset region is a first preset value. The preset relationship includes: if the amount of data in any region within the preset region is greater than four times the first preset value, then the average value and the dispersion index corresponding to any data are obtained as a target point; and the corresponding region is divided into four regions according to the target point. Among them, the number of valid samples in each of the four regions is greater than the first preset value; The calling module is further configured to obtain the quantity of all products distributed in each preset area and the quantity of defective products; obtain the probability that the next produced product in each preset area is a defective product based on the correspondence between the quantity of all products and the quantity of defective products; and form a defect probability table based on the average value and the dispersion index in each preset area and the probability of the defective products corresponding to them.

5. A product data anomaly prediction device, characterized in that, Includes memory used to store computer programs; A processor, configured to implement the steps of the product data anomaly prediction method as described in any one of claims 1 to 3 when executing the computer program.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the product data anomaly prediction method as described in any one of claims 1 to 3.

Images