Equipment management method and device utilizing partition data management

Abstract

The invention relates to an equipment management method and device utilizing partition data management, and the method comprises the steps: removing redundant features of partition data through dimension reduction processing, taking the partition data as a training data set, taking a feature mark corresponding to the partition data as a training set, and training a classification model for carrying out the mark classification of the partition data; inputting the partitioned data into a classification model, and outputting a first classification mark to determine a first mapping table; analyzing an approximation result of the adjacently sorted partition data in the first mapping table; performing partition data remodeling and mark sorting adjustment on the first mapping table according to an approximation result to obtain a second mapping table; and performing spatial distribution optimization on the partition data in the second mapping table to obtain a third mapping table for partition data management. On the basis, the partition data of the laboratory equipment originally recorded by the system is subjected to multiple times of data processing, so that the data reference value of the partition data is improved, the accuracy is improved for further large model training, and the data noise is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of laboratory equipment data management, and in particular to an equipment management method and device utilizing partition data management. Background Art

[0002] Laboratory equipment is an indispensable tool in research work, and it is necessary to ensure that the equipment is regularly inspected, calibrated and maintained. Laboratory equipment status management refers to the process of effectively managing and maintaining various scientific instruments, equipment and facilities in the laboratory. Its purpose is to ensure that the equipment is always in good operating condition, improve laboratory work efficiency, ensure the accuracy and reliability of experimental results, and extend the service life of the equipment.

[0003] In actual laboratory equipment data management, due to the large types and space occupied by various laboratory equipment, the workload of laboratory space planning and data management is huge. Therefore, in the laboratory equipment data management platform, laboratory equipment is generally managed by partitions, and partition data is determined in advance. However, the partition management of laboratory equipment is greatly interfered by human factors, resulting in low reference value and inaccuracy of such data applied to data training and data management in large models. Summary of the invention

[0004] Based on this, it is necessary to provide an equipment management method and device using partition data management to address the problem that the partition management of laboratory equipment is greatly interfered by human factors, resulting in low reference and inaccuracy of data training and data management of such data applied to large models.

[0005] A device management method using partition data management, comprising the steps of: Removing redundant features of the partition data by dimensionality reduction processing, taking the partition data as a training data set, taking feature labels corresponding to the partition data as a training set, and training a classification model for labeling and classifying the partition data; Input the partition data into the classification model and output a first classification label to determine a first mapping table; wherein the first mapping table includes a mapping relationship between the partition data and the first classification label and a label sorting; Analyze the approximate results of the adjacently sorted partition data in the first mapping table; According to the approximate result, partition data is reshaped and mark order is adjusted on the first mapping table to obtain a second mapping table; The partition data in the second mapping table is spatially optimized to obtain a third mapping table for partition data management.

[0006] The above-mentioned equipment management method using partition data management removes redundant features of partition data through dimensionality reduction processing, uses partition data as a training data set, and uses feature labels corresponding to partition data as a training set to train a classification model for labeling and classifying partition data; inputs partition data into the classification model, outputs a first classification label to determine a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of laboratory equipment originally recorded in the system, the data reference value of the partition data is improved, and the accuracy is improved and data noise is reduced for further large model training.

[0007] In one embodiment, the partition data includes device partitions, each device partition having a plurality of characteristic tags, as follows: ; in, Indicates Device partition number The data value of the feature marker.

[0008] In one embodiment, the process of removing redundant features of partition data by dimensionality reduction processing includes the steps of: Performing standardization processing on the partition data to obtain standardized data; Extracting the principal components of the standardized data and constructing a principal component matrix; Perform dimensionality reduction processing on the principal component matrix to obtain a dimensionality reduction matrix to remove redundant data.

[0009] In one embodiment, the partition data is used as a training data set, and the feature labels corresponding to the partition data are used as a training set to train a classification model for labeling and classifying the partition data, including the steps of: A training data set is constructed using a dimension reduction matrix, a training set is constructed using partitioned data, and a mapping relationship is trained based on a decision tree model to obtain the classification model.

[0010] In one embodiment, the process of analyzing the approximate results of adjacently sorted partition data in the first mapping table comprises the steps of: An approximate analysis matrix is ​​constructed according to the first mapping table, and the approximation of the adjacent partition data is calculated according to the approximate analysis matrix as the approximation result.

[0011] In one embodiment, the process of analyzing the approximate results of adjacently sorted partition data in the first mapping table comprises the steps of: The transition smoothness and data boundaries of the partitioned data are optimized through data reshaping and data optimization.

[0012] In one embodiment, the steps are also included: Merge some partition data to reduce data fragmentation.

[0013] A device management device using partition data management, comprising: A model training module is used to remove redundant features of partition data by dimensionality reduction processing, use the partition data as a training data set, and use feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; A mapping establishment module, used for inputting the partition data into the classification model and outputting a first classification label to determine a first mapping table; wherein the first mapping table includes a mapping relationship between the partition data and the first classification label and a label sorting; An approximate analysis module, used for analyzing the approximate results of adjacently sorted partition data in the first mapping table; A mapping update module, configured to reshape partition data and adjust the mark order of the first mapping table according to the approximate result to obtain a second mapping table; The mapping optimization module is used to optimize the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management.

[0014] The above-mentioned equipment management device using partition data management removes redundant features of partition data through dimensionality reduction processing, uses partition data as a training data set, and uses feature labels corresponding to partition data as a training set to train a classification model for labeling and classifying partition data; inputs partition data into the classification model, outputs a first classification label to determine a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of laboratory equipment originally recorded in the system, the data reference value of the partition data is improved, and the accuracy is improved and data noise is reduced for further large model training.

[0015] A computer storage medium stores computer instructions, which, when executed by a processor, implement the device management method using partition data management of any of the above embodiments.

[0016] The above-mentioned computer storage medium removes redundant features of partition data through dimensionality reduction processing, uses the partition data as a training data set, and uses the feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; inputs the partition data into the classification model, outputs a first classification label, and determines a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of the laboratory equipment originally recorded by the system, the data reference value of the partition data is improved, and the accuracy is improved and the data noise is reduced for further large model training.

[0017] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the device management method utilizing partition data management of any of the above embodiments is implemented.

[0018] The above-mentioned computer device removes redundant features of partition data through dimensionality reduction processing, uses the partition data as a training data set, and uses the feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; inputs the partition data into the classification model, outputs a first classification label, and determines a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of the laboratory equipment originally recorded by the system, the data reference value of the partition data is improved, and the accuracy is improved and the data noise is reduced for further large model training. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a device management method using partition data management according to an embodiment of an application; Figure 2 A module structure diagram of a device management device utilizing partition data management according to an embodiment of an application; Figure 3 A schematic diagram of a computer structure of an implementation method. DETAILED DESCRIPTION

[0020] In order to better understand the purpose, technical solution and technical effect of the present invention, the present invention is further explained below in conjunction with the accompanying drawings and embodiments. At the same time, it is stated that the embodiments described below are only used to explain the present invention and are not used to limit the present invention.

[0021] The embodiment of the present invention provides a device management method using partition data management.

[0022] Figure 1 FIG. 1 is a flow chart of a device management method using partition data management according to an embodiment of an application. Figure 1 As shown, a device management method using partition data management in an embodiment of the application includes the following steps: S100, removing redundant features of the partition data by dimensionality reduction processing, taking the partition data as a training data set, taking feature labels corresponding to the partition data as a training set, and training a classification model for labeling and classifying the partition data; S101, inputting the partition data into the classification model, outputting a first classification label, to determine a first mapping table; wherein the first mapping table includes a mapping relationship between the partition data and the first classification label and a label sorting; S102, analyzing the approximate result of adjacently sorted partition data in the first mapping table; S103, reshaping partition data and adjusting mark order of the first mapping table according to the approximate result to obtain a second mapping table; S104: Optimize the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management.

[0023] In an embodiment of the present application, the data source of the partition data is the management system of the laboratory equipment. The management system determines the partition data and feature tags by partitioning the laboratory equipment. Among them, the feature tags are label data pre-marked according to the partition data, which are used to characterize the management characteristics of the partition data. The method execution subject of this embodiment, as an intermediary to connect the management system and the AI ​​big model, performs secondary processing on the partition data of the management system as training data for the AI ​​big model. The AI ​​big model is used to further enhance the accuracy of laboratory equipment partition management and improve the generalization of partition data application.

[0024] The partition data includes the device partitions recorded by the management system. Each device partition has a unified partition standard and projection method, and is marked with a corresponding feature tag. The feature tag serves as the data record of the device partition.

[0025] Preferably, the partition data is as follows: ;in, Indicates Device partition number The data value of the feature marker.

[0026] Based on this, the process of removing redundant features of the partition data by dimensionality reduction processing in step S100 includes the steps of: performing standardization processing on the partition data to obtain standardized data; Extracting the principal components of the standardized data and constructing a principal component matrix; Perform dimensionality reduction processing on the principal component matrix to obtain a dimensionality reduction matrix to remove redundant data.

[0027] Referring to the above steps, partition data in the matrix The standardized data is obtained in the form of matrix. .

[0028] The principal components of the standardized data are extracted and the principal component matrix is ​​constructed as follows: , the principal component matrix is ​​reduced in dimension to obtain the reduced dimension matrix , as follows: .

[0029] Among them, the principal component matrix The column vector is The principal component directions are retained after dimensionality reduction. principal components.

[0030] Preferably, in step S100, the partition data is used as a training data set, and the feature labels corresponding to the partition data are used as a training set to train a classification model for labeling and classifying the partition data, comprising the steps of: A training data set is constructed using a dimension reduction matrix, a training set is constructed using partitioned data, and a mapping relationship is trained based on a decision tree model to obtain the classification model.

[0031] As in the above steps, the dimension reduction matrix Constructing training data set . Partitioning the Device The reduced dimension feature vector of , For the feature markers, A collection of feature tags.

[0032] The mapping relationship is as follows, and its model function is characterized as follows: The classification model is trained based on the model function. The classification model outputs a first mapping table based on the model function according to the re-input of the device partition in the partition data.

[0033] Preferably, the process of analyzing the approximate results of adjacently sorted partition data in the first mapping table in step S102 comprises the steps of: An approximate analysis matrix is ​​constructed according to the first mapping table, and the approximation of the adjacent partition data is calculated according to the approximate analysis matrix as the approximation result.

[0034] The adjacent sorting of the first mapping table, that is, the adjacent sorting of the device partitions, constructs the approximate analysis matrix ,in and Both represent device partitions.

[0035] ;in, Indicates device partition and Whether to sort adjacently to construct a feature similarity matrix , as follows: in: Indicates device partition According to the dimension reduction matrix The eigenvectors calculated are, Indicates device partition According to the dimension reduction matrix Compute the eigenvector space of the single; For vector The Euclidean norm of , For vector The Euclidean norm of , As an approximate result.

[0036] Preferably, the process of analyzing the approximate results of adjacently sorted partition data in the first mapping table in step S102 comprises the steps of: The transition smoothness and data boundaries of the partitioned data are optimized through data reshaping and data optimization.

[0037] Based on approximate results , identify adjacent device partitions whose approximate results are greater than the set approximate threshold, adjust the device partitions and unify the feature labels, improve the spatial transition smoothness of the device partitions with approximate feature labels, optimize the device partition data boundaries reflected in the second mapping table, and facilitate improving the data smoothness of the third mapping table, which is conducive to enhancing the training efficiency of subsequent AI large models.

[0038] Preferably, the method further comprises the steps of: Merge some partition data to reduce data fragmentation.

[0039] Among them, the partition data can be merged through cluster analysis algorithm.

[0040] In one of the embodiments, by setting the core parameters of the clustering analysis algorithm, some device partitions are screened and isolated, and the clustered device partitions are merged to reduce the fragmentation of data subsequently input into the AI ​​large model.

[0041] In the third mapping table after completing the above steps, the device partition and feature tag mapping with the above data reprocessing are used as training data for the AI ​​big model.

[0042] The equipment management method using partition data management in any of the above embodiments removes redundant features of the partition data through dimensionality reduction processing, uses the partition data as a training data set, and uses the feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; inputs the partition data into the classification model, outputs a first classification label, and determines a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of the laboratory equipment originally recorded by the system, the data reference value of the partition data is improved, and the accuracy is improved and the data noise is reduced for further large model training.

[0043] The embodiment of the present invention also provides a device management apparatus using partition data management.

[0044] Figure 2 FIG. 1 is a module structure diagram of a device management device utilizing partition data management according to an implementation method. Figure 2 As shown, a device management device using partition data management in one embodiment includes: The model training module 100 is used to remove redundant features of the partition data by dimensionality reduction processing, use the partition data as a training data set, and use the feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; A mapping establishment module 101 is used to input the partition data into the classification model and output a first classification label to determine a first mapping table; wherein the first mapping table includes a mapping relationship between the partition data and the first classification label and a label sorting; An approximate analysis module 102, configured to analyze approximate results of adjacently sorted partition data in the first mapping table; A mapping update module 103, configured to reshape partition data and adjust the mark order of the first mapping table according to the approximate result to obtain a second mapping table; The mapping optimization module 104 is used to optimize the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management.

[0045] The above-mentioned equipment management device using partition data management removes redundant features of partition data through dimensionality reduction processing, uses partition data as a training data set, and uses feature labels corresponding to partition data as a training set to train a classification model for labeling and classifying partition data; inputs partition data into the classification model, outputs a first classification label to determine a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of laboratory equipment originally recorded in the system, the data reference value of the partition data is improved, and the accuracy is improved and data noise is reduced for further large model training.

[0046] An embodiment of the present invention further provides a computer storage medium on which computer instructions are stored. When the instructions are executed by a processor, the device management method using partition data management of any of the above embodiments is implemented.

[0047] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0048] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, RAM, ROM, magnetic disks or optical disks.

[0049] Corresponding to the above-mentioned computer storage medium, in one embodiment, a computer device is also provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a device management method utilizing partition data management as described in any one of the above-mentioned embodiments is implemented.

[0050] The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a device management method using partition data management is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0051] The above-mentioned computer device removes redundant features of partition data through dimensionality reduction processing, uses the partition data as a training data set, and uses the feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; inputs the partition data into the classification model, outputs a first classification label, and determines a first mapping table; analyzes the approximate results of adjacently sorted partition data in the first mapping table; reshapes the partition data and adjusts the label sorting of the first mapping table according to the approximate results to obtain a second mapping table; optimizes the spatial distribution of the partition data in the second mapping table to obtain a third mapping table for partition data management. Based on this, by performing multiple data processing on the partition data of the laboratory equipment originally recorded by the system, the data reference value of the partition data is improved, and the accuracy is improved and the data noise is reduced for further large model training.

[0052] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0053] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A device management method using partition data management, characterized in that: Includes steps: Removing redundant features of the partition data by dimensionality reduction processing, taking the partition data as a training data set, taking feature labels corresponding to the partition data as a training set, and training a classification model for labeling and classifying the partition data; Input the partition data into the classification model and output a first classification label to determine a first mapping table; wherein the first mapping table includes a mapping relationship between the partition data and the first classification label and a label sorting; Analyze the approximate results of the adjacently sorted partition data in the first mapping table; According to the approximate result, partition data is reshaped and mark order is adjusted on the first mapping table to obtain a second mapping table; The partition data in the second mapping table is spatially optimized to obtain a third mapping table for partition data management.

2. The device management method using partition data management according to claim 1, characterized in that: The partition data includes device partitions, each of which has multiple characteristic tags, as follows: ;in, Indicates Device partition number The data value of the feature marker.

3. The device management method using partition data management according to claim 1, characterized in that: The process of removing redundant features of partition data by dimensionality reduction processing comprises the steps of: Performing standardization processing on the partition data to obtain standardized data; Extracting the principal components of the standardized data and constructing a principal component matrix; Perform dimensionality reduction processing on the principal component matrix to obtain a dimensionality reduction matrix to remove redundant data.

4. The device management method using partition data management according to claim 1, characterized in that: The process of using the partition data as a training data set and the feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data includes the steps of: A training data set is constructed using a dimension reduction matrix, a training set is constructed using partitioned data, and a mapping relationship is trained based on a decision tree model to obtain the classification model.

5. The device management method using partition data management according to claim 1, characterized in that: The process of analyzing the approximate results of adjacently sorted partition data in the first mapping table comprises the steps of: An approximate analysis matrix is ​​constructed according to the first mapping table, and the approximation of the adjacent partition data is calculated according to the approximate analysis matrix as the approximation result.

6. The device management method using partition data management according to claim 5, characterized in that: The process of analyzing the approximate results of adjacently sorted partition data in the first mapping table comprises the steps of: The transition smoothness and data boundaries of the partitioned data are optimized through data reshaping and data optimization.

7. The device management method using partition data management according to claim 1, characterized in that: Also includes the steps: Merge some partition data to reduce data fragmentation.

8. A device management apparatus utilizing partition data management, characterized in that: include: A model training module is used to remove redundant features of partition data by dimensionality reduction processing, use the partition data as a training data set, and use feature labels corresponding to the partition data as a training set to train a classification model for labeling and classifying the partition data; Input the partition data into the classification model and output a first classification label to determine a first mapping table; wherein the first mapping table includes a mapping relationship between the partition data and the first classification label and a label sorting; Analyze the approximate results of the adjacently sorted partition data in the first mapping table; According to the approximate result, partition data is reshaped and mark order is adjusted on the first mapping table to obtain a second mapping table; The partition data in the second mapping table is spatially optimized to obtain a third mapping table for partition data management.

9. A computer storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the device management method using partition data management as described in any one of claims 1 to 7 is implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the device management method using partition data management as described in any one of claims 1 to 7 is implemented.

Images