A method and system for data classification in manufacturing industry

Abstract

The present invention discloses a method and system for data grading and classification for the manufacturing industry, and belongs to the field of electronic digital data processing. The present invention imports the calculated estimated production product abnormal value and document safety of the document into the classification and grading strategy to obtain the danger level of the document, and outputs the documents whose danger level is greater than or equal to the set danger level threshold in descending order of the danger level. The management personnel manage the components corresponding to the documents according to the output danger level corresponding to the documents. The present invention derives the relationship between the scene data and the product quality data based on the evolution of the historical manufacturing scene data and the changes in the product quality data, quickly discovers the abnormal data according to the development of the scene data, handles the faults in advance, and improves the production quality of the product.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic digital data processing, and in particular to a method and system for data grading and classification in the manufacturing industry. Background Art

[0002] At present, data governance stakeholders need to do a lot of work to achieve the goal of data labeling, and lack sufficient data governance efficiency. It is impossible to judge how to classify and categorize data more effectively in the early stage of data processing, and the growing demand for big data has not been fully considered; existing data management solutions are mostly based on their own rule systems, without fully considering the newly issued data governance guidelines or industry specifications, and the scope of adaptability and scalability are limited. The above problems exist in existing technologies;

[0003] For example, a Chinese patent with publication number CN113268548A discloses a data grading and classification method based on a data middle platform, which specifically includes the following steps: A. Establish data classification standards; B. Establish data grading standards; C. Combine according to the data middle platform. The present invention classifies individual sub-item combinations into module sets, and assigns the module sets a name etymology based on word vectors, which is used to associate all sub-items in the module set with a name etymology, and at the same time sets a grading mode based on demand, sets a total decreasing table, and uses the total decreasing table as the update basis. It can be used as the basis for use by multiple different user permissions, and realizes the branch table downward according to the total decreasing table. Finally, according to the data matrix, the purpose of rapid retrieval and rapid classification is achieved, so as to reduce the calculation process and complexity of the overall database. The grading and classification effects are mainly formed based on the database in one step, and the authenticity of the data conclusions is higher;

[0004] At the same time, for example, a Chinese patent with publication number CN115438243A provides a data classification and grading system based on blockchain, including a data classification system and a data grading system. The data grading system can grade each type of data in the data classification system. The data classification system includes an important data module, a personal and enterprise information module and a business data module. The important data module can be divided into first-level important data, second-level important data, third-level important data, fourth-level important data and fifth-level important data. The personal and enterprise information module can be divided into low-level personal and enterprise information, intermediate-level personal and enterprise information and high-level personal and enterprise information. The present invention can refine the data, form a data asset directory, and conduct effective data management, which is convenient for upstream and downstream enterprises and regulatory departments to provide a standard and unified resource classification system and resource statistical caliber, and is convenient for staff to search data assets and obtain required data.

[0005] The above patents all have the problems raised by this background technology: in the process of product production, the existing manufacturing industry is unable to quickly deduce the relationship between scene data and product quality data based on the evolution of historical manufacturing scene data and changes in product quality data. The speed of discovering abnormal data is slow and faults cannot be handled in advance. In order to solve these problems, this application designs a method and system for data grading and classification for the manufacturing industry. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention proposes a method and system for data grading and classification in the manufacturing industry. The present invention obtains the operating parameter data of each device and the quality parameter data of the produced product in the manufacturing production scenario process, screens the obtained data to obtain the abnormal product and abnormal equipment operating parameter data therein, imports the abnormal product and abnormal equipment operating parameter data in the manufacturing production scenario process into the business data recognition model construction strategy to construct a business data recognition model, based on the constructed business data recognition model, imports the document data to be determined into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document, based on the constructed business data recognition model, imports the document data to be determined into the document security calculation strategy to calculate the document security, imports the calculated estimated production product abnormal value and document security into the classification and grading strategy to obtain the danger level of the document, outputs the document whose danger level is greater than or equal to the set danger level threshold in descending order according to the size of the danger level, and the management personnel manage the components corresponding to the document according to the output danger level corresponding to the document. The present invention derives the relationship between the scene data and the product quality data based on the evolution of the historical manufacturing scene data and the change of the product quality data, quickly discovers the abnormal data according to the development of the scene data, handles the fault in advance, and improves the production quality of the product.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for data classification and categorization in the manufacturing industry comprises the following specific steps:

[0009] S1. Obtain the operating parameter data of each device and the quality parameter data of the products produced in the manufacturing production scenario, and screen the obtained data to obtain the operating parameter data of abnormal products and abnormal equipment;

[0010] S2. Importing abnormal product and abnormal equipment operating parameter data in the manufacturing production scenario into the business data recognition model construction strategy to build a business data recognition model;

[0011] S3. Based on the constructed business data recognition model, the document data to be judged is imported into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document;

[0012] S4. Based on the constructed business data identification model, the document data to be judged is imported into the document security calculation strategy to calculate the security of the document;

[0013] S5. Import the calculated estimated production product abnormal value and document security into the classification and grading strategy to obtain the danger level of the document;

[0014] S6. Output documents whose danger levels are greater than or equal to a set danger level threshold in descending order of danger level, and the management personnel manage components corresponding to the documents according to the output documents corresponding to the danger level.

[0015] Specifically, S1 includes the following specific steps:

[0016] S11. Collecting the operation parameter data of each device in the manufacturing production scenario through the device operation data collection terminal, wherein the device operation parameter data includes the voltage, current, vibration amplitude and vibration frequency data of the device operation;

[0017] S12. Collect the production product image data through the production product image acquisition terminal, extract the production product image, and take the image data of the normal production sample at the same time, obtain the pixel value of each corresponding pixel point of the production product image and the normal production sample image, and substitute it into the production product abnormal value calculation formula to calculate the abnormal value of the production product. The production product abnormal value calculation formula is: Among them, n1 is the number of pixels on the product, x j is the pixel value of the jth pixel on the production product image, z j is the pixel value of the pixel point on the normal production sample corresponding to the j-th pixel point on the production product image, where j is any one of 1 to n1;

[0018] S13, extracting corresponding production products whose calculated production product abnormality values ​​are greater than or equal to the production product abnormality threshold and setting them as abnormal products, and extracting corresponding equipment operating parameter data of the abnormal production products and setting them as abnormal equipment operating parameter data;

[0019] It should be noted that the abnormal threshold of the production product here is to substitute the pixel values ​​of each pixel point of 500 groups of defective products and the pixel values ​​of each pixel point of the normal production sample image into the production product abnormal value calculation formula to calculate the production product abnormal value, and take the average value of the 500 groups of production product abnormal values ​​to obtain;

[0020] Specifically, the specific content of the business data identification model construction strategy in S2 is as follows:

[0021] The business data recognition model equation is constructed by using the historical abnormal equipment operation parameter data and the corresponding production product abnormal values, and these data are divided into a 70% weighted training set and a 30% weighted test set; the business data recognition model is weighted trained to construct a deep learning neural network model with the historical abnormal equipment operation parameter data as input and the production product abnormal values ​​as output, and the 70% weighted training set is input into the deep learning neural network model for training to obtain an initial neural network model, and the initial neural network model is tested using the 30% weighted test set, and the optimal initial neural network model that meets the accuracy of the production product abnormal values ​​is output as the business data recognition model equation, wherein the output strategy formula of the business data recognition model equation is: in is the n+1 layer p-term neuron output, is the connection weight from the ith neuron in the nth layer to the pth neuron in the n+1th layer, represents the ith neuron in the nth layer, Represents the bias from the ith neuron in the nth layer to the pth neuron in the n+1th layer, sig() represents the Sigmoid activation function, and m is the number of neurons in the nth layer.

[0022] Specifically, the specific content of the production product abnormal value estimation calculation strategy in S3 is:

[0023] The real-time equipment operation parameter data of the equipment in the document is obtained, including the real-time voltage, real-time current, real-time vibration amplitude and real-time vibration frequency data of the equipment, and is imported into the business data recognition model to output the estimated production product abnormal value NK.

[0024] Specifically, the document security calculation strategy in S4 includes the following specific steps:

[0025] S41, obtaining historical equipment operation parameter data corresponding to the scenario where the ratio of the difference between the historical production product abnormal value and the estimated production product abnormal value NK is less than 5%, and setting it as the preliminary selected equipment parameter set;

[0026] S42, separate various parameters of the preliminary device parameter set, obtain average values ​​of various parameters in the preliminary device parameter set, import the obtained average values ​​of various parameters and the device operation parameter data of the device in real time in the document into the document security calculation formula to calculate the document security. The document security calculation formula is: Where exp() is the exponential power of e, C is the category of the parameters in the preliminary device parameter set, and v c is the cth item of data in the average value of the parameters in the preliminary selected equipment parameter set, The cth item of data in the equipment operation parameters of the equipment in real time operation, a c is the proportion coefficient of the cth item data in the parameter, where Wherein, c is any one of 1 to C.

[0027] Specifically, the classification and grading strategy in S5 is as follows:

[0028] The document security of the real-time operating parameter data of the equipment and the estimated abnormal values ​​of the production products in the obtained document are imported into the classification and grading calculation formula to calculate the danger level of the document. The classification and grading calculation formula is: W=M×(1-NK).

[0029] Specifically, a system for data grading and classification in the manufacturing industry is implemented based on the above-mentioned method for data grading and classification in the manufacturing industry, and specifically includes: a data acquisition module, a screening module, a business data identification model construction module, a production product outlier value calculation module, a safety calculation module, a danger level acquisition module, a threat output module, a management module and a control module. The data acquisition module is used to obtain the operating parameter data of each device in the manufacturing production scenario and the quality parameter data of the produced products. The screening module is used to screen the data to obtain the operating parameter data of abnormal products and abnormal equipment.

[0030] Specifically, the business data identification model construction module is used to construct a business data identification model based on the abnormal products and abnormal equipment operating parameter data in the manufacturing production scenario process imported into the business data identification model construction strategy; the production product abnormal value calculation module is used to import the document data that needs to be judged into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document based on the constructed business data identification model; the safety calculation module is used to import the document data that needs to be judged into the document safety calculation strategy to calculate the safety of the document based on the constructed business data identification model; the danger level acquisition module is used to import the calculated estimated production product abnormal value and document safety of the document into the classification and grading strategy to obtain the danger level of the document.

[0031] Specifically, the threat output module is used to output documents whose danger levels are greater than or equal to a set danger level threshold in descending order of danger level; the management module is used for management personnel to manage components corresponding to documents according to the output danger level corresponding to the documents; the control module is used to control the operation of the data acquisition module, the screening module, the business data identification model construction module, the production product outlier calculation module, the safety calculation module, the danger level acquisition module, the threat output module and the management module.

[0032] Specifically, an electronic device includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0033] The processor executes the above-mentioned method for data grading and classification in the manufacturing industry by calling the computer program stored in the memory.

[0034] Specifically, a computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the above-mentioned method for data grading and classification in the manufacturing industry.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention obtains the operation parameter data of each device and the quality parameter data of the produced products in the manufacturing production scenario process, screens the obtained data to obtain the abnormal products and abnormal equipment operation parameter data therein, imports the abnormal products and abnormal equipment operation parameter data in the manufacturing production scenario process into the business data recognition model construction strategy to construct a business data recognition model, based on the constructed business data recognition model, imports the document data that needs to be determined into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document, based on the constructed business data recognition model, imports the document data that needs to be determined into the document security calculation strategy to calculate the document security, imports the calculated estimated production product abnormal value and document security into the classification and grading strategy to obtain the danger level of the document, outputs the document whose danger level is greater than or equal to the set danger level threshold in descending order of the danger level, and the management personnel manage the components corresponding to the document according to the output danger level corresponding to the document. The present invention derives the relationship between the scene data and the product quality data based on the evolution of the historical manufacturing scene data and the changes in the product quality data, quickly discovers the abnormal data according to the development of the scene data, handles the fault in advance, and improves the production quality of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a method for data grading and classification in the manufacturing industry according to the present invention;

[0038] Figure 2 This is a specific flow chart of step S1 of a method for data grading and classification in the manufacturing industry according to the present invention;

[0039] Figure 3 This is a specific flow chart of step S4 of a method for data grading and classification in the manufacturing industry of the present invention;

[0040] Figure 4The present invention is a schematic diagram of a system architecture for data grading and classification in the manufacturing industry. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0042] Example 1

[0043] See also Figure 1-Figure 3 , an embodiment of the present invention provides: a method for data classification and classification in the manufacturing industry, which comprises the following specific steps:

[0044] S1. Obtain the operating parameter data of each device and the quality parameter data of the products produced in the manufacturing production scenario, and screen the obtained data to obtain the operating parameter data of abnormal products and abnormal equipment;

[0045] In this embodiment, S1 includes the following specific steps:

[0046] S11. Collecting the operation parameter data of each device in the manufacturing production scenario through the device operation data collection terminal, wherein the device operation parameter data includes the voltage, current, vibration amplitude and vibration frequency data of the device operation;

[0047] The following is the code for the equipment operation data collection terminal written in C language, which is used to collect the operating parameter data of each device in the manufacturing production scenario.

[0048]

[0049]

[0050]

[0051] S12. Collect the production product image data through the production product image acquisition terminal, extract the production product image, and take the image data of the normal production sample at the same time, obtain the pixel value of each corresponding pixel point of the production product image and the normal production sample image, and substitute it into the production product abnormal value calculation formula to calculate the abnormal value of the production product. The production product abnormal value calculation formula is: Among them, n1 is the number of pixels on the product, x j is the pixel value of the jth pixel on the production product image, z j is the pixel value of the pixel point on the normal production sample corresponding to the j-th pixel point on the production product image, where j is any one of 1 to n1;

[0052] S13, extracting corresponding production products whose calculated production product abnormality values ​​are greater than or equal to the production product abnormality threshold and setting them as abnormal products, and extracting corresponding equipment operating parameter data of the abnormal production products and setting them as abnormal equipment operating parameter data;

[0053] It should be noted that the abnormal threshold of the production product here is to substitute the pixel values ​​of each pixel point of 500 groups of defective products and the pixel values ​​of each pixel point of the normal production sample image into the production product abnormal value calculation formula to calculate the production product abnormal value, and take the average value of the 500 groups of production product abnormal values ​​to obtain;

[0054] S2. Importing abnormal product and abnormal equipment operating parameter data in the manufacturing production scenario into the business data recognition model construction strategy to build a business data recognition model;

[0055] In this embodiment, the specific content of the business data identification model construction strategy in S2 is as follows:

[0056] The business data recognition model equation is constructed through the historical abnormal equipment operation parameter data and the corresponding production product abnormal values, and these data are divided into a 70% weighted training set and a 30% weighted test set; the business data recognition model is weighted trained to construct a deep learning neural network model with the historical abnormal equipment operation parameter data as input and the production product abnormal values ​​as output, and the 70% weighted training set is input into the deep learning neural network model for training to obtain the initial neural network model, and the initial neural network model is tested with the 30% weighted test set, and the optimal initial neural network model that meets the accuracy of the production product abnormal values ​​is output as the business data recognition model equation, wherein the output strategy formula of the business data recognition model equation is: in is the output of n+1 layers of p-term neurons, is the connection weight from the ith neuron in the nth layer to the pth neuron in the n+1th layer, represents the i-th neuron in the n-th layer, represents the bias from the i-th neuron in the n-th layer to the p-th neuron in the n+1-th layer, sig() represents the Sigmoid activation function, and m is the number of neurons in the n-th layer;

[0057] The following is a sample code for building a business data recognition model using Python and the deep learning framework TensorFlow. Assume that the data is stored in X and y, where X is the historical abnormal equipment operation parameter data and y is the corresponding production product abnormal value.

[0058]

[0059]

[0060] Please note that a simple neural network model is used here, and the model structure can be adjusted according to actual conditions;

[0061] S3. Based on the constructed business data recognition model, the document data to be judged is imported into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document;

[0062] In this embodiment,

[0063] S4. Based on the constructed business data identification model, the document data to be judged is imported into the document security calculation strategy to calculate the security of the document;

[0064] In this embodiment, the specific content of the production product abnormal value estimation calculation strategy in S3 is:

[0065] Obtain the real-time equipment operation parameter data of the equipment in the document, including the real-time voltage, real-time current, real-time vibration amplitude and real-time vibration frequency data of the equipment, and import them into the business data recognition model to output the estimated production product abnormal value NK;

[0066] S5. Import the calculated estimated production product abnormal value and document security into the classification and grading strategy to obtain the danger level of the document;

[0067] In this embodiment, the document security calculation strategy in S4 includes the following specific steps:

[0068] S41, obtaining historical equipment operation parameter data corresponding to the scenario where the ratio of the difference between the historical production product abnormal value and the estimated production product abnormal value NK is less than 5%, and setting it as the preliminary selected equipment parameter set;

[0069] S42, separate various parameters of the preliminary device parameter set, obtain average values ​​of various parameters in the preliminary device parameter set, import the obtained average values ​​of various parameters and the device operation parameter data of the device in real time in the document into the document security calculation formula to calculate the document security. The document security calculation formula is: Where exp() is the exponential power of e, C is the category of the parameters in the preliminary device parameter set, and v c is the cth item of data in the average value of the parameters in the preliminary selected equipment parameter set, The cth item of data in the equipment operation parameters of the equipment in real time operation, a c is the proportion coefficient of the cth item data in the parameter, where Wherein, c is any one of 1 to C;

[0070] S6. Output documents whose danger levels are greater than or equal to a set danger level threshold in descending order of danger level, and the management personnel manage components corresponding to the documents according to the output documents corresponding to the danger level.

[0071] Through this embodiment, it is possible to achieve: obtaining the operating parameter data of each device and the quality parameter data of the produced products in the manufacturing production scenario process, screening the obtained data to obtain the abnormal products and abnormal equipment operating parameter data therein, importing the abnormal products and abnormal equipment operating parameter data in the manufacturing production scenario process into the business data recognition model construction strategy to build a business data recognition model, based on the constructed business data recognition model, importing the document data that needs to be determined into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document, based on the constructed business data recognition model, importing the document data that needs to be determined into the document security calculation strategy to calculate the document security, importing the calculated estimated production product abnormal value and document security into the classification and grading strategy to obtain the danger level of the document, outputting the documents whose danger level is greater than or equal to the set danger level threshold in descending order of the danger level, and the management personnel manage the components corresponding to the documents according to the documents corresponding to the output danger level. The present invention derives the relationship between the scene data and the product quality data based on the evolution of the historical manufacturing scene data and the changes in the product quality data, quickly discovers the abnormal data according to the development of the scene data, handles the fault in advance, and improves the production quality of the product.

[0072] Example 2

[0073] like Figure 4 As shown, a system for data grading and classification in the manufacturing industry is implemented based on the above-mentioned method for data grading and classification in the manufacturing industry, and specifically includes: a data acquisition module, a screening module, a business data identification model construction module, a production product outlier value calculation module, a safety calculation module, a danger level acquisition module, a threat output module, a management module and a control module. The data acquisition module is used to obtain the operating parameter data of each device in the manufacturing production scenario and the quality parameter data of the produced products. The screening module is used to screen the data to obtain the operating parameter data of abnormal products and abnormal equipment.

[0074] In this embodiment, the business data identification model construction module is used to construct a business data identification model based on the abnormal products and abnormal equipment operating parameter data in the manufacturing production scenario process imported into the business data identification model construction strategy; the production product abnormal value calculation module is used to import the document data that needs to be judged into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document based on the constructed business data identification model; the safety calculation module is used to import the document data that needs to be judged into the document safety calculation strategy to calculate the safety of the document based on the constructed business data identification model; the danger level acquisition module is used to import the calculated estimated production product abnormal value and document safety of the document into the classification and grading strategy to obtain the danger level of the document.

[0075] In this embodiment, the threat output module is used to output documents whose danger level is greater than or equal to the set danger level threshold in descending order of the danger level. The management module is used for management personnel to manage the components corresponding to the documents according to the output danger level corresponding to the documents. The control module is used to control the operation of the data acquisition module, the screening module, the business data identification model construction module, the production product outlier calculation module, the safety calculation module, the danger level acquisition module, the threat output module and the management module.

[0076] Example 3

[0077] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0078] The processor executes the above-mentioned method for data grading and classification in the manufacturing industry by calling the computer program stored in the memory.

[0079] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, and the computer program is loaded and executed by the processor to implement a method for data grading and classification for the manufacturing industry provided by the above method embodiment. The electronic device may also include other components for implementing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0080] Example 4

[0081] This embodiment provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0082] When the computer program runs on a computer device, the computer device is caused to execute the above-mentioned method for data grading and classification in the manufacturing industry.

[0083] For example, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0084] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0085] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0086] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or / and a wireless network. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0087] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0088] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0089] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0090] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0091] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0092] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0093] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for data classification in the manufacturing industry. It is characterized in that It includes the following specific steps: S1. Obtain the operating parameter data of each device and the quality parameter data of the products produced in the manufacturing production scenario, and screen the obtained data to obtain the operating parameter data of abnormal products and abnormal equipment; S2. Importing abnormal product and abnormal equipment operating parameter data in the manufacturing production scenario into the business data recognition model construction strategy to build a business data recognition model; S3. Based on the constructed business data recognition model, the document data to be judged is imported into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document; S4. Based on the constructed business data identification model, the document data to be judged is imported into the document security calculation strategy to calculate the security of the document; S5. Import the calculated estimated production product abnormal value and document security into the classification and grading strategy to obtain the danger level of the document; S6, outputting documents whose danger levels are greater than or equal to the set danger level threshold in descending order of danger level, and the management personnel manage the components corresponding to the documents according to the output documents corresponding to the danger level; S1 includes the following specific steps: S11. Collecting the operation parameter data of each device in the manufacturing production scenario through the device operation data collection terminal, wherein the device operation parameter data includes the voltage, current, vibration amplitude and vibration frequency data of the device operation; S12. Collect the production product image data through the production product image acquisition terminal, extract the production product image, and take the image data of the normal production sample at the same time, obtain the pixel value of each corresponding pixel point of the production product image and the normal production sample image, and substitute it into the production product abnormal value calculation formula to calculate the abnormal value of the production product. The production product abnormal value calculation formula is: Among them, n1 is the number of pixels on the product, x j is the pixel value of the jth pixel on the product image, z j is the pixel value of the pixel point on the normal production sample corresponding to the j-th pixel point on the production product image, where j is any one of 1 to n1; S13, extracting the corresponding production products whose calculated production product abnormality values ​​are greater than or equal to the production product abnormality threshold and setting them as abnormal products, extracting the corresponding equipment operation parameter data of the production abnormal products and setting them as abnormal equipment operation parameter data; the specific content of the business data identification model construction strategy in S2 is as follows: The business data recognition model equation is constructed by using the historical abnormal equipment operation parameter data and the corresponding production product abnormal values, and these data are divided into a 70% weighted training set and a 30% weighted test set; the business data recognition model is weighted trained to construct a deep learning neural network model with the historical abnormal equipment operation parameter data as input and the production product abnormal values ​​as output, and the 70% weighted training set is input into the deep learning neural network model for training to obtain an initial neural network model, and the initial neural network model is tested using the 30% weighted test set, and the optimal initial neural network model that meets the accuracy of the production product abnormal values ​​is output as the business data recognition model equation, wherein the output strategy formula of the business data recognition model equation is: in is the n+1 layer p-term neuron output, is the connection weight from the ith neuron in the nth layer to the pth neuron in the n+1th layer, represents the ith neuron in the nth layer, represents the bias from the i-th neuron in the n-th layer to the p-th neuron in the n+1-th layer, sig() represents the Sigmoid activation function, and m is the number of neurons in the n-th layer; the specific content of the production product abnormal value estimation calculation strategy in S3 is: The equipment operation parameter data of the equipment in the document is obtained in real time, including the real-time voltage, real-time current, real-time vibration amplitude and real-time vibration frequency data of the equipment, and is imported into the business data recognition model to output the estimated production product abnormal value NK; the document security calculation strategy in S4 includes the following specific steps: S41, obtaining historical equipment operation parameter data corresponding to the scenario where the ratio of the difference between the historical production product abnormal value and the estimated production product abnormal value NK is less than 5%, and setting it as the preliminary selected equipment parameter set; S42, separate various parameters of the preliminary device parameter set, obtain average values ​​of various parameters in the preliminary device parameter set, import the obtained average values ​​of various parameters and the device operation parameter data of the device in real time in the document into the document security calculation formula to calculate the document security. The document security calculation formula is: Where exp() is the exponential power of e, C is the category of the parameters in the preliminary device parameter set, and v c is the cth item of data in the average value of the parameters in the preliminary selected equipment parameter set, is the cth item of data in the equipment operation parameters of the equipment in real time operation, ac is the proportion coefficient of the cth item of data in the parameters, where, Wherein, c is any one of 1 to C; the specific content of the classification and grading strategy in S5 is as follows: The document security of the real-time operating parameter data of the equipment and the estimated abnormal values ​​of the production products in the obtained document are imported into the classification and grading calculation formula to calculate the danger level of the document. The classification and grading calculation formula is: W=M×(1-NK).

2. A system for data classification and categorization in the manufacturing industry, which is implemented based on the method for data classification and categorization in the manufacturing industry as claimed in claim 1. It is characterized in that It specifically includes: a data acquisition module, a screening module, a business data identification model building module, a production product outlier value calculation module, a safety calculation module, a danger level acquisition module, a threat output module, a management module and a control module. The data acquisition module is used to obtain the operating parameter data of each device in the manufacturing production scenario and the product quality parameter data produced. The screening module is used to screen the data to obtain the operating parameter data of abnormal products and abnormal equipment.

3. A system for data classification and categorization in the manufacturing industry as claimed in claim 2, It is characterized in that The business data identification model construction module is used to construct a business data identification model based on the abnormal products and abnormal equipment operating parameter data in the manufacturing production scenario process imported into the business data identification model construction strategy; the production product abnormal value calculation module is used to import the document data that needs to be judged into the production product abnormal value estimation calculation strategy to calculate the estimated production product abnormal value of the document based on the constructed business data identification model; the safety degree calculation module is used to import the document data that needs to be judged into the document safety degree calculation strategy to calculate the safety degree of the document based on the constructed business data identification model; the danger level acquisition module is used to import the calculated estimated production product abnormal value and document safety degree of the document into the classification and grading strategy to obtain the danger level of the document.

4. A system for data classification and categorization in the manufacturing industry as claimed in claim 3, It is characterized in that The threat output module is used to output documents whose danger level is greater than or equal to the set danger level threshold in descending order of danger level. The management module is used for management personnel to manage components corresponding to documents according to the output danger level corresponding to the documents. The control module is used to control the operation of the data acquisition module, the screening module, the business data identification model construction module, the production product outlier calculation module, the safety calculation module, the danger level acquisition module, the threat output module and the management module.

5. An electronic device, It is characterized in that include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the method for data grading and classification for the manufacturing industry as claimed in claim 1 by calling the computer program stored in the memory.

Images