Quality inspection data treatment method and device

Abstract

The invention relates to the technical field of big data processing, and discloses a quality inspection data treatment method and device, and the method comprises the steps: determining a current quality inspection index based on current quality inspection data; utilizing a machine learning model to predict a current mean value and a current standard deviation of normal distribution obeyed by the current quality inspection index according to the historical quality inspection index sequence of the current quality inspection index; determining a current difference value between the current quality inspection index and the current mean value; and performing data management operation on the current quality inspection data based on the difference between the current difference value and the current standard deviation. Therefore, the accuracy of quality inspection data detection is effectively improved, and subsequent data treatment operation on the current quality inspection data is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and in particular to a method and device for managing quality inspection data. Background Art

[0002] At present, in the construction of the data middle platform, data governance, as a core component, bears the key responsibility of ensuring data quality, integrity and availability. Data quality management is particularly important. It detects and fixes data problems through multi-dimensional rules and strategies to ensure that data can support business decisions and operations. Data quality inspection rules are usually verified according to specific thresholds to ensure that the data meets the preset quality standards. However, with the complexity of business and the diversification of data, traditional quality inspection rules based on fixed thresholds have gradually exposed some shortcomings and cannot cope with complex business scenarios with periodic or dynamic changes. Therefore, how to govern quality inspection data is a technical problem that needs to be solved urgently. Summary of the invention

[0003] In view of this, the present invention provides a quality inspection data management method, apparatus, computer equipment, storage medium, and computer program product to solve the problem of how to manage quality inspection data.

[0004] In a first aspect, the present invention provides a method for managing quality inspection data, the method comprising:

[0005] Determine the current quality inspection indicators based on the current quality inspection data;

[0006] Using the machine learning model, based on the historical quality inspection indicator sequence of the current quality inspection indicator, the current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator are predicted;

[0007] Determine the current difference between the current quality inspection index and the current mean;

[0008] Based on the difference between the current difference and the current standard deviation, data governance operations are performed on the current quality inspection data.

[0009] The quality inspection data management method provided by the present invention has the following advantages:

[0010] The current mean and standard deviation of the prediction are generated by the machine learning model based on the historical quality inspection indicator sequence of the current quality inspection indicator. According to the difference between the current difference between the current quality inspection indicator and the current mean and the current standard deviation, data governance operations are performed on the current quality inspection data. In this way, the current mean and standard deviation of the prediction can be dynamically generated based on the historical quality inspection indicator sequence of the current quality inspection indicator, and the current quality inspection data can be governed, so as to adapt to complex business scenarios with periodic or dynamic changes, and the normal data range can be defined more accurately, so as to effectively improve the accuracy of quality inspection data detection, so as to facilitate the subsequent data governance operations on the current quality inspection data.

[0011] In an optional implementation, the machine learning model is trained according to the following steps:

[0012] Using the machine learning model, perform the following steps:

[0013] Generate a predicted mean and a predicted standard deviation of a normal distribution to which the predicted quality inspection indicator obeys according to the quality inspection indicator sample, wherein the quality inspection indicator sample includes a historical quality inspection indicator sequence of the real quality inspection indicator;

[0014] Based on the predicted mean and predicted standard deviation of the normal distribution obeyed by the predicted quality inspection indicator, determine the probability density function and cumulative distribution function of the normal distribution obeyed by the predicted quality inspection indicator;

[0015] Based on the predicted mean and predicted standard deviation, probability density function and cumulative distribution function of the normal distribution obeyed by the predicted quality inspection indicators, the loss is generated and used to update the values ​​of the parameters of the machine learning model.

[0016] In an optional implementation, the data governance operation includes: an alarm operation and a fuse operation;

[0017] Based on the difference between the current difference and the current standard deviation, data governance operations are performed on the current quality inspection data, including:

[0018] When the current difference is greater than or equal to the current standard deviation or less than twice the current standard deviation, an alarm operation is performed on the current quality inspection data;

[0019] When the current difference is greater than twice the current standard deviation, the current quality inspection data is fused.

[0020] In an optional implementation, performing an alarm operation on the current quality inspection data includes:

[0021] Send an alarm message to the monitoring end, indicating that the current quality inspection data is abnormal data;

[0022] Perform fuse operations on the current quality inspection data, including:

[0023] A fuse instruction is sent to the control end, and the fuse instruction instructs the control end to fuse the data node corresponding to the current quality inspection data according to the fuse instruction.

[0024] In an optional embodiment, the method further includes:

[0025] Obtaining a preset threshold range, the preset threshold range including a preset maximum value and a preset minimum value;

[0026] When the current quality inspection index is greater than the preset maximum value or less than the preset minimum value, a fuse operation is performed on the current quality inspection data.

[0027] In an optional implementation, the loss is calculated as follows:

[0028]

[0029] Among them, y is the real quality inspection index in the quality inspection index sample, μ is the predicted mean of each iterative training output, and σ is the predicted standard deviation of each iterative training output;

[0030] The probability density function of the normal distribution is φ:

[0031]

[0032] The cumulative distribution function of the normal distribution is Φ:

[0033]

[0034] t is the time corresponding to y.

[0035] In a second aspect, the present invention provides a device for managing quality inspection data, the device comprising:

[0036] A first determination module, used to determine the current quality inspection index based on the current quality inspection data;

[0037] A prediction module is used to use a machine learning model to predict the current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator based on the historical quality inspection indicator sequence of the current quality inspection indicator;

[0038] The second determination module is used to determine the current difference between the current quality inspection index and the current mean;

[0039] The governance module is used to perform data governance operations on the current quality inspection data based on the difference between the current difference and the current standard deviation.

[0040] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the quality inspection data management method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0041] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the quality inspection data management method of the first aspect or any corresponding embodiment thereof.

[0042] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the quality inspection data management method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 is a flow chart of a method for managing quality inspection data according to an embodiment of the present invention;

[0045] Figure 2 is a schematic diagram of the training process of a machine learning model according to an embodiment of the present invention;

[0046] Figure 3 is a schematic diagram of a device for managing quality inspection data according to an embodiment of the present invention;

[0047] Figure 4 It is a structural schematic diagram of a computer device provided by an optional embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0049] According to an embodiment of the present invention, an embodiment of a method for managing quality inspection data is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0050] In this embodiment, a method for managing quality inspection data is provided, which can be used in the above-mentioned computer. Figure 1 is a flow chart of a method for managing quality inspection data according to an embodiment of the present invention. Figure 1 As shown, the above method includes:

[0051] Step S101, determining the current quality inspection index based on the current quality inspection data.

[0052] In this embodiment, the current quality inspection data is data used in the data management system to monitor data quality. The current quality inspection index can be a quality inspection result obtained by performing quality inspection on the current quality inspection data according to a preset verification rule.

[0053] Step S102, using a machine learning model, based on the historical quality inspection indicator sequence of the above-mentioned current quality inspection indicator, predict the current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator.

[0054] In this embodiment, the key to this step is to use historical data for prediction. The machine learning model will use historical quality inspection data to predict the current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator. Specifically, it can be several historical quality inspection indicators that obey the normal distribution selected from the historical quality inspection indicator sequence. The current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator can be the mean and standard deviation corresponding to the data set obeying the normal distribution formed by the current quality inspection indicator and several historical quality inspection indicators.

[0055] Step S103, determining the current difference between the current quality inspection index and the current mean.

[0056] In this embodiment, the current difference between the current quality inspection index and the current mean value may be the absolute value of the difference between the current quality inspection index and the current mean value.

[0057] Step S104: performing data governance operations on the current quality inspection data based on the difference between the current difference and the current standard deviation.

[0058] In this embodiment, data governance operations are performed based on the calculated difference and standard deviation. The standard deviation reflects the fluctuation range of the data, so by comparing the difference with the standard deviation, it can be determined whether the current data quality is within the normal range. If the difference between the difference and the standard deviation is very large, it may mean that there is a large deviation in the current data (such as data quality problems or outliers). If the difference is small and the difference is within the tolerance range of the standard deviation, it indicates that the current data quality is within the normal fluctuation range. Different data governance measures can be taken according to the comparison results of the difference and the standard deviation. For example: if the difference exceeds the set threshold of the standard deviation, data cleaning, correction or alarm measures can be triggered. If the difference is within the standard deviation range, it can be ignored or continue to be monitored.

[0059] The quality inspection data management method provided by the present invention has the following advantages:

[0060] The predicted current mean and current standard deviation are generated by the machine learning model based on the historical quality inspection indicator sequence of the above current quality inspection indicator, and data governance operations are performed on the current quality inspection data based on the difference between the current difference between the current quality inspection indicator and the current mean and the above current standard deviation. In this way, the predicted current mean and current standard deviation can be dynamically generated based on the historical quality inspection indicator sequence of the current quality inspection indicator, and the current quality inspection data can be governed, so as to adapt to complex business scenarios with periodic or dynamic changes, and the normal data range can be more accurately defined, so as to effectively improve the accuracy of quality inspection data detection, so as to facilitate subsequent data governance operations on the current quality inspection data.

[0061] In some optional implementations, the above machine learning model is trained according to the following steps:

[0062] Figure 2 Schematic diagram of the training process of the machine learning model according to an embodiment of the present invention. Figure 2 As shown, using the above machine learning model, perform the following steps:

[0063] Step S201: Generate a predicted mean and a predicted standard deviation of a normal distribution to which the quality inspection index is subject based on the quality inspection index sample.

[0064] In the present embodiment, the above-mentioned quality inspection indicator sample includes a historical quality inspection indicator sequence of a real quality inspection indicator. The quality inspection indicator sample can be a set of historical quality inspection indicators, such as quality inspection indicators of the past few months. This sample sequence can be used to describe the changing trend and fluctuation of the quality inspection indicators in history. The quality inspection indicator sample can be a set of quality inspection indicators for the historical quality inspection indicator sequence, using a sliding window method, when the sliding step is 1 and the sequence length is 30, a quality inspection indicator set is constructed as a quality inspection indicator sample, and each quality inspection indicator set includes 30 continuous historical quality inspection indicators. If the amount of historical quality inspection indicator data is large, the sliding step can be increased, or a random method can be used to construct a quality inspection indicator sample. The purpose of generating the mean and standard deviation is to predict the normal fluctuation range of the quality inspection indicator. These two parameters are the core of describing the normal distribution, and they will help us understand the center of the data distribution and its fluctuations.

[0065] Step S202, based on the predicted mean and predicted standard deviation of the normal distribution obeyed by the above-mentioned predicted quality inspection indicator, determine the probability density function and cumulative distribution function of the normal distribution obeyed by the above-mentioned predicted quality inspection indicator.

[0066] In this embodiment, the probability density function describes the probability of a data value occurring. In a normal distribution, the probability density function gives the probability of a particular value occurring. The cumulative distribution function describes the probability that a quality inspection indicator is less than or equal to a certain value. In a normal distribution, the cumulative distribution function represents the integral of the probability density function. The cumulative distribution function can characterize the relative position of the quality inspection indicator.

[0067] Step S203, generating a loss based on the predicted mean and predicted standard deviation of the normal distribution obeyed by the above-mentioned predicted quality inspection index, the above-mentioned probability density function and the cumulative distribution function.

[0068] In the present embodiment, the above-mentioned loss is used to update the value of the parameter of the above-mentioned machine learning model. The way of generating loss can be to calculate the continuous ranking probability score (CRPS, Continuous Ranked Probability Score) based on the predicted mean and predicted standard deviation, the above-mentioned probability density function and the cumulative distribution function. The lower the above-mentioned loss, the closer the predicted mean and predicted standard deviation are to the mean and standard deviation of the normal distribution obeyed by the real quality inspection index. Specifically, a gradient-based learning machine algorithm can be adopted, for example, linear regression, SVM regression, decision tree regression and random forest regression, as well as a gradient boosting decision tree (GradientBoosting Decision Tree, GBDT) and XGBoost (Extreme Gradient Boosting, extreme gradient boosting) in the AdaBoost (Adaptive Boosting) algorithm, based on the gradient descent algorithm to optimize the model.

[0069] The quality inspection data governance method in this embodiment establishes a quality prediction model based on normal distribution through historical quality inspection data. By predicting the mean and standard deviation, the distribution range of future quality inspection indicators can be reasonably inferred, and the description of quality inspection data can be further optimized through probability density function and cumulative distribution function. By modeling the normal distribution of quality inspection data, the fluctuation of data can be accurately described, avoiding the limitations of single value prediction. The training method based on the CRPS loss function can comprehensively consider the difference between the entire distribution and the actual observed value, optimize the model parameters, and make the prediction more accurate. It can cope with complex fluctuations in quality inspection data, has strong adaptability, and is suitable for different types of quality inspection data.

[0070] In some optional implementations, the above data governance operations include: alarm operations, fuse operations;

[0071] The above-mentioned data governance operation is performed on the above-mentioned current quality inspection data based on the difference between the above-mentioned current difference and the above-mentioned current standard deviation, including: when the above-mentioned current difference is greater than or equal to the above-mentioned current standard deviation or less than twice the above-mentioned current standard deviation, an alarm operation is performed on the above-mentioned current quality inspection data; when the above-mentioned current difference is greater than twice the above-mentioned current standard deviation, a fuse operation is performed on the above-mentioned current quality inspection data.

[0072] In this embodiment, based on the difference between the current quality inspection data and the predicted value (i.e., the current difference) and the relationship between the current standard deviation, it is determined whether an alarm operation or a fuse operation needs to be triggered. Specifically, when the current difference is greater than or equal to the current standard deviation or less than twice the current standard deviation, it means that the current quality inspection data deviates from the expected range. Although the difference has not reached an extreme level, an alarm prompt is still required, and an alarm operation is performed on the current quality inspection data; when the current difference is greater than twice the current standard deviation, it means that the quality inspection data is seriously abnormal, and a fuse operation is performed on the current quality inspection data.

[0073] The quality inspection data governance method in this embodiment effectively improves the recognition ability and response speed of abnormal quality inspection indicators in the quality inspection process by combining alarm operation and fuse operation based on the relationship between the current difference and the current standard deviation. In the case of large fluctuations in quality inspection indicators, the alarm operation can promptly remind operators to pay attention to potential problems, and the fuse operation can quickly take measures to stop production when the abnormal quality inspection indicators exceed the acceptable range to prevent abnormal quality inspection data from spreading to other network nodes.

[0074] In some optional implementations, the above-mentioned alarm operation on the above-mentioned current quality inspection data includes:

[0075] Sending an alarm message to the monitoring end, the alarm message indicating that the current quality inspection data is abnormal data;

[0076] The above-mentioned fusing operation on the above-mentioned current quality inspection data includes: sending a fusing instruction to the control end, and the above-mentioned fusing instruction instructs the above-mentioned control end to fuse the data node corresponding to the above-mentioned current quality inspection data according to the above-mentioned fusing instruction.

[0077] In this embodiment, the alarm operation may be that if the difference of the current quality inspection data (i.e., the difference between the current value and the predicted value) is greater than or equal to the current standard deviation, or less than twice the current standard deviation, the alarm operation is triggered. The purpose of the alarm operation is to promptly notify the monitoring end where the quality control personnel or system administrator is located, indicating that the current quality inspection data deviates from the expected range. Although the difference is not extreme, it is still enough to attract attention and may indicate potential quality problems or data anomalies. The fuse operation may be triggered when the difference of the current quality inspection data is greater than twice the current standard deviation. The function of the fuse operation is to force the node in the data processing to stop or pause to prevent further spread of unqualified or abnormal data. The control end may be a device that controls the connection relationship between network nodes. The fuse mechanism is usually triggered when a serious anomaly occurs, with the aim of reducing possible losses and ensuring that quality problems can be resolved in a timely manner.

[0078] The quality inspection data governance method in this embodiment can quickly respond to minor anomalies through alarm operations. The problem of abnormal quality inspection data diffusion can be effectively avoided through fuse operations. The response strategy can be adjusted according to the fluctuation of data to adapt to different quality inspection environments. The occurrence of false positives and false negatives can be reduced, and the data governance process can be optimized.

[0079] In some optional embodiments, the above method also includes: obtaining a preset threshold range, the above preset threshold range includes a preset maximum value and a preset minimum value; when the above current quality inspection index is greater than the above preset maximum value or less than the above preset minimum value, performing a fuse operation on the above current quality inspection data.

[0080] In this embodiment, the quality inspection index may be an index with a fixed value range, that is, the preset threshold range. When the quality inspection data exceeds the fixed value range, that is, greater than the preset maximum value or less than the preset minimum value, the current quality inspection data can be directly fused.

[0081] The quality inspection data governance method in this embodiment can effectively improve the accuracy and response speed of quality inspection indicators by introducing preset threshold ranges and fuse operations in quality inspection data governance. The setting of preset thresholds can ensure that the quality inspection data is within a reasonable range and avoid quality fluctuations caused by abnormal data; while the fuse operation can quickly cut off potential risks and prevent abnormal quality inspection data from spreading to other nodes.

[0082] In some optional implementations, the above loss is calculated as follows:

[0083]

[0084] Among them, y is the real quality inspection index in the quality inspection index sample, μ is the predicted mean output of each iterative training, σ is the predicted standard deviation of each iterative training output; the probability density function of the normal distribution is φ:

[0085]

[0086] The cumulative distribution function of the normal distribution is Φ:

[0087]

[0088] t is the time corresponding to y.

[0089] In this embodiment, the real quality inspection indicator in the quality inspection indicator sample is y, the predicted mean of each iterative training output is μ, and the predicted standard deviation of each iterative training output is σ; Φ is the cumulative distribution function of the normal distribution; φ is the probability density function of the normal distribution, and the time corresponding to the real quality inspection indicator is t.

[0090] The quality inspection data governance method in this embodiment uses the CRPS loss function to quantify the deviation between the predicted mean and predicted standard deviation output by the prediction model and the actual quality inspection index, further optimize the model parameters, and improve the accuracy and stability of the prediction results. Effectively deal with the uncertainty and fluctuations that may exist in the data.

[0091] The present invention provides a device for managing quality inspection data. Figure 3 Schematic diagram of a device for managing quality inspection data according to an embodiment of the present invention. Figure 3 As shown, including:

[0092] The first determination module 301 is used to determine the current quality inspection index based on the current quality inspection data;

[0093] The prediction module 302 is used to use a machine learning model to predict the current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator according to the historical quality inspection indicator sequence of the current quality inspection indicator;

[0094] The second determination module 303 is used to determine the current difference between the current quality inspection index and the current mean;

[0095] The governance module 304 is used to perform data governance operations on the current quality inspection data based on the difference between the current difference and the current standard deviation.

[0096] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0097] The quality inspection data management device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0098] The embodiment of the present invention also provides a computer device having the above Figure 3 The quality inspection data management device shown.

[0099] See also Figure 4 , Figure 4 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 4 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 4 A processor 10 is taken as an example.

[0100] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include an integrated circuit. The integrated circuit may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0101] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0102] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0103] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0104] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0105] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0106] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0107] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for managing quality inspection data, characterized in that: The method comprises: Determine the current quality inspection indicators based on the current quality inspection data; Using a machine learning model, based on a historical quality inspection indicator sequence of the current quality inspection indicator, predict a current mean and a current standard deviation of a normal distribution obeyed by the current quality inspection indicator; Determine a current difference between the current quality inspection indicator and the current mean; Based on the difference between the current difference value and the current standard deviation, a data governance operation is performed on the current quality inspection data.

2. The method according to claim 1, characterized in that The machine learning model is trained according to the following steps: Using the machine learning model, perform the following steps: Generate a predicted mean and a predicted standard deviation of a normal distribution obeyed by the predicted quality inspection indicator according to the quality inspection indicator sample, wherein the quality inspection indicator sample includes a historical quality inspection indicator sequence of the real quality inspection indicator; Determine the probability density function and cumulative distribution function of the normal distribution obeyed by the predicted quality inspection indicator based on the predicted mean and predicted standard deviation of the normal distribution obeyed by the predicted quality inspection indicator; Based on the predicted mean and predicted standard deviation of the normal distribution obeyed by the predicted quality inspection indicator, the probability density function and the cumulative distribution function, a loss is generated, and the loss is used to update the value of the parameter of the machine learning model.

3. The method according to claim 1, characterized in that The data governance operations include: alarm operations and fuse operations; The performing a data governance operation on the current quality inspection data based on the difference between the current difference value and the current standard deviation includes: When the current difference is greater than or equal to the current standard deviation or less than twice the current standard deviation, an alarm operation is performed on the current quality inspection data; When the current difference is greater than twice the current standard deviation, a fuse operation is performed on the current quality inspection data.

4. The method according to claim 3, characterized in that The performing an alarm operation on the current quality inspection data includes: Sending an alarm message to the monitoring end, wherein the alarm message indicates that the current quality inspection data is abnormal data; The performing a fuse operation on the current quality inspection data includes: A fuse instruction is sent to the control end, wherein the fuse instruction instructs the control end to fuse the data node corresponding to the current quality inspection data according to the fuse instruction.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Obtaining a preset threshold range, wherein the preset threshold range includes a preset maximum value and a preset minimum value; When the current quality inspection index is greater than the preset maximum value or less than the preset minimum value, a fuse operation is performed on the current quality inspection data.

6. The method according to claim 2, characterized in that The loss is calculated as follows: Among them, y is the real quality inspection index in the quality inspection index sample, μ is the predicted mean of each iterative training output, and σ is the predicted standard deviation of each iterative training output; The probability density function of the normal distribution is φ: The cumulative distribution function of the normal distribution is Φ: t is the time corresponding to y.

7. A quality inspection data management device, characterized in that: The device comprises: A first determination module, used to determine the current quality inspection index based on the current quality inspection data; A prediction module, used to predict the current mean and current standard deviation of the normal distribution obeyed by the current quality inspection indicator according to the historical quality inspection indicator sequence of the current quality inspection indicator by using a machine learning model; A second determination module is used to determine a current difference between the current quality inspection index and the current mean; A governance module is used to perform data governance operations on the current quality inspection data based on the difference between the current difference value and the current standard deviation.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the quality inspection data management method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the quality inspection data management method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the quality inspection data management method according to any one of claims 1 to 6.

Images