Application classification method and configuration software service system

Abstract

The application discloses an application program classification method and a configuration software service system, and the method comprises the following steps: obtaining input end data and output end data of various application programs to form a training data set; performing classification training on the training data set to obtain a classification algorithm of the application programs; and classifying the application programs to be classified according to the classification algorithm to determine the categories of the application programs to be classified. The application solves the problem that the application program classification mode of the configuration software in the prior art is not objective and accurate, and improves the accuracy, objectivity and reliability of the application program classification mode.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of software, in particular to an application program classification method and a configuration software service system. BACKGROUND

[0002] In the field of industrial control, containers, virtualization and cloud configuration can be combined to form a configuration development system for sharing application programs. In the system, there are developers and users (the developers develop application programs for the users using configuration). The application programs developed by the developers can be used by different users (similar to a configuration application program market).

[0003] The above system faces the problem of application program classification, because users need to select the application programs to be used according to the classification. The classification method is either manual or based on the description or label after use by the user. The above classification method may not match the actual application scene or may be maliciously brushed, resulting in inaccurate classification results.

[0004] In view of the problem that the application program classification method of the configuration software is not objective and accurate in the related art, no effective solution has been proposed so far. SUMMARY

[0005] The present application provides an application program classification method and a configuration software service system to at least solve the problems in the prior art.

[0006] To solve the above technical problems, according to an aspect of an embodiment of the present application, an application program classification method is provided, including: obtaining input end data and output end data of various application programs to form a training data set; performing classification training on the training data set to obtain a classification algorithm of the application program; and classifying the application program to be classified according to the classification algorithm to determine the category of the application program to be classified.

[0007] Further, obtaining the input end data and the output end data of various application programs to form a training data set includes: randomly extracting a first preset number of samples from each category of application programs; for each sample, collecting input end data and output end data of a second preset number of scanning periods to form a training data set.

[0008] Further, performing classification training on the training data set to obtain a classification algorithm of the application program includes: dividing the training data set into a training set and a test set; training the training set and the test set according to a machine learning algorithm to obtain a classification algorithm.

[0009] Further, the application program to be classified is classified according to the classification algorithm, and the category of the application program to be classified is determined, including: collecting input end data and output end data of a third preset number of scanning periods of the application program to be classified; and classifying the input end data and the output end data of the third preset number of scanning periods according to the classification algorithm to determine the category of the application program to be classified.

[0010] Further, the input end data and the output end data of the third preset number of scanning periods of the application program to be classified are collected, including: collecting the input end data and the output end data of the third preset number of scanning periods of the application program to be classified when the application program to be classified runs in the simulation environment, and / or collecting the input end data and the output end data of the third preset number of scanning periods of the application program to be classified when the application program to be classified runs in the production environment.

[0011] According to another aspect of the embodiment of the application, a configuration software service system is provided, including: a cloud computing node configured to provide a classification service of an application program for a user, including: a first digital twin module configured to obtain input end data and output end data of various application programs; a classification algorithm training module configured to form a training data set from the input end data and the output end data, and to perform classification training on the training data set to obtain a classification algorithm of the application program; and a classification module configured to classify an application program to be classified according to the classification algorithm, and to determine a category of the application program to be classified.

[0012] Further, the system further includes: an end node configured to run the application program, and to collect the input end data and the output end data and upload the input end data and the output end data to the cloud computing node.

[0013] Further, the end node includes: an application program running module configured to run the application program; an input port sampling module configured to collect input end data of each scanning period when the application program runs; an output port sampling module configured to collect output end data of each scanning period when the application program runs; and a second digital twin module configured to communicate with the first digital twin module and to send the input end data and the output end data to the first digital twin module.

[0014] Further, the cloud computing node further includes: an application configuration module configured to provide a configuration design service of the application program for the user; and an application simulation module configured to provide a simulation service for the user.

[0015] Further, the cloud computing node is further configured to provide a recommendation and selection service of an application instance for the user.

[0016] Further, the system further includes: an edge computing node; and the edge computing node includes: a simulation edge computing node configured to perform the simulation service; and a production edge computing node configured to deploy the recommendation and selection service of the application instance.

[0017] Furthermore, the edge computing node includes a runtime scheduling module, which schedules the execution of applications based on the resources of the end node.

[0018] Furthermore, the edge computing node also includes a third digital twin module, which communicates with the first and second digital twin modules to synchronize data from the cloud computing node, the edge computing node, and the end node.

[0019] According to another aspect of the present invention, a storage medium comprising computer-executable instructions is provided, which, when executed by a computer processor, are used to perform the application classification method described above.

[0020] This invention provides a method for classifying applications. Based on the signal data corresponding to the inputs at the application's input ports and the outputs at its output ports, a classification algorithm is trained using machine learning. The trained algorithm is then used to classify the application. Classification based on measurable signal data (i.e., input and output data) offers higher accuracy, objectivity, and reliability, avoids the influence of malicious data manipulation, and prevents discrepancies between stated claims and actual usage, making it more suitable for industrial control applications. Attached Figure Description

[0021] Figure 1 This is an optional flowchart of an application classification method according to an embodiment of the present invention;

[0022] Figure 2 This is an optional structural block diagram of a configuration software service system according to an embodiment of the present invention;

[0023] Figure 3 This is an optional structural block diagram of a cloud computing node according to an embodiment of the present invention;

[0024] Figure 4 This is an optional structural block diagram of an end node according to an embodiment of the present invention;

[0025] Figure 5 This is an optional structural block diagram of an edge computing node according to an embodiment of the present invention. Detailed Implementation

[0026] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0027] Example 1

[0028] In a preferred embodiment 1 of the present invention, an application classification method is provided, specifically... Figure 1 This illustrates one possible flowchart of the application's classification method, such as... Figure 1 As shown, the method includes the following steps S102-S106:

[0029] S102: Acquire input and output data from various applications to form a training dataset;

[0030] S104: Perform classification training on the training dataset to obtain the classification algorithm for the application;

[0031] S106: Classify the application to be classified according to the classification algorithm to determine the category of the application to be classified.

[0032] Industrial configuration software applications typically operate with a fixed scan cycle. Each scan cycle reads input from the input port, performs logic processing, and then sends the output to the output port. This process is repeated cyclically. These applications have well-defined boundaries and interfaces, where the boundaries are measurable.

[0033] Based on this, the above embodiments provide a method for classifying applications. A classification algorithm is trained using machine learning based on the signal data corresponding to the inputs at the application's input ports and the outputs at its output ports. The trained algorithm is then used to classify the application to be classified. Classification based on measurable signal data (i.e., input and output data) offers higher accuracy, objectivity, and reliability, avoids the influence of malicious data manipulation, and prevents discrepancies between stated claims and actual usage, making it more suitable for industrial control applications.

[0034] The training dataset is formed by acquiring input and output data from various applications, including: randomly selecting a first preset number of samples from each application category; and for each sample, collecting input and output data for a second preset number of scan cycles. Optionally, N samples are randomly selected from each known application category, and for each sample, input signal features at the input port and output signal features at the output port are collected for multiple scan cycles, forming M feature data records. Finally, a training dataset of Cn (number of categories) * N (number of samples) * M (number of feature data records) is obtained.

[0035] After determining the training dataset, classification training is performed on the training dataset to obtain the application's classification algorithm. This includes: dividing the training dataset into training and test sets; and training the training and test sets according to a machine learning algorithm to obtain the classification algorithm. Classification algorithms, such as K-nearest neighbors, support vector machines, decision trees, or deep learning, can all be used for classification training.

[0036] After completing the classification training, the application to be classified is classified according to the classification algorithm to determine the category of the application to be classified. This includes: collecting input and output data of the application to be classified for a third preset number of scanning cycles; classifying the input and output data of the application to be classified according to the classification algorithm for the third preset number of scanning cycles to determine the category of the application to be classified.

[0037] Specifically, the system collects input signals from input ports and output data from output ports during multiple scan cycles of the newly submitted application running in the simulation / production environment. That is, it collects input and output data of the application to be classified during a third preset number of scan cycles, including: collecting input and output data of the application to be classified during a third preset number of scan cycles running in the simulation environment, and / or collecting input and output data of the application to be classified during a third preset number of scan cycles running in the production environment.

[0038] Furthermore, the classification algorithm can be further trained and improved based on more training data, providing the algorithm with the ability to evolve.

[0039] Example 2

[0040] In a preferred embodiment 2 of the present invention, a configuration software service system is provided. Specifically, Figure 2 This diagram illustrates one possible structural design of the system, such as... Figure 2 As shown, the entire system consists of three layers: the first layer is the cloud computing node L1; the second layer is the edge computing node L2, which includes simulation edge computing nodes and production edge computing nodes (used in the production environment); and the third layer is the end node L3, which includes industrial control units such as PLC, HMI, and APC.

[0041] L1: Provides application configuration services (coding, testing, simulation, etc.), application instance browsing, recommendation, and selection services for developers and users.

[0042] L2: Execute service functions. Simulation edge nodes execute simulation functions, while production edge nodes are used to deploy instances in the production environment.

[0043] L3: Run application instances, execute the functions of the application instances, and perform functions such as virtual simulation and actual hardware simulation.

[0044] Figure 3 This diagram illustrates one possible structure of the cloud computing node in the system, such as... Figure 3 As shown, cloud computing nodes include:

[0045] The first digital twin module is used to acquire input and output data from various applications;

[0046] The classification algorithm training module is used to form a training dataset from the input and output data, perform classification training on the training dataset, and obtain the classification algorithm for the application. It executes the classification algorithm training function based on the training data prepared by the first digital twin module. Furthermore, as the classification algorithm evolves, updated training data is incorporated into the training dataset.

[0047] The classification module is used to classify the applications to be classified according to the classification algorithm and determine the category of the applications to be classified.

[0048] The above embodiments provide a method for classifying applications. Based on the signal data corresponding to the inputs at the application's input ports and the outputs at its output ports, a classification algorithm is trained using machine learning. The trained algorithm is then used to classify the application. Classification based on measurable signal data (i.e., input and output data) offers higher accuracy, objectivity, and reliability, avoids the influence of malicious data manipulation, and prevents discrepancies between stated claims and actual usage, making it more suitable for industrial control applications.

[0049] In addition, cloud computing nodes also include:

[0050] The application configuration module provides users with application configuration design services, including coding and testing.

[0051] The application simulation module provides simulation services to users and developers, with simulation functionality running directly on the simulation edge nodes. The simulation module includes virtual simulation and actual hardware simulation. The simulation edge nodes respectively assume the roles of virtual simulation nodes and actual hardware simulation nodes.

[0052] like Figure 2 As shown, the system also includes: end nodes, which are used to run applications and collect input and output data, and upload them to cloud computing nodes. Figure 4 This diagram illustrates one possible structure of the system's end node, such as... Figure 4 As shown, the end nodes include:

[0053] The application runtime module is used to run the application;

[0054] The input port sampling module is used to collect input data for each scan cycle during application runtime;

[0055] The output port sampling module is used to collect output data for each scan cycle during application runtime;

[0056] The second digital twin module is used to communicate with the first digital twin module and send input and output data to the first digital twin module.

[0057] like Figure 2 As shown, the system also includes: edge computing nodes, which include:

[0058] Simulated edge computing nodes are used to execute simulation services;

[0059] Production edge computing nodes are used for the recommendation and selection of application instances.

[0060] Figure 5 This diagram illustrates one possible structure of the edge computing node in the system, such as... Figure 5 As shown, the edge computing node includes a runtime scheduling module, used to schedule the execution of applications based on the resources of the edge node. Application runtime scheduling includes at least resource startup, shutdown, and migration.

[0061] The edge computing node also includes a third digital twin module, used to communicate with the first and second digital twin modules to synchronize data from the cloud computing node, edge computing node, and endpoint node. The first, second, and third digital twin modules are used to synchronize the state (resource configuration, input port data, and output port data) of cloud-edge-end application instance containers.

[0062] Classification based on measurable signal data offers higher accuracy and objectivity, avoids the influence of malicious data manipulation, and prevents discrepancies between claims and actual use, making it more suitable for industrial control applications.

[0063] Example 3

[0064] Based on the application classification method provided in Embodiment 1 above, in a preferred embodiment 3 of the present invention, a storage medium containing computer-executable instructions is also provided, which, when executed by a computer processor, are used to execute the application classification method as described above.

[0065] The above embodiments provide a method for classifying applications. Based on the signal data corresponding to the inputs at the application's input ports and the outputs at its output ports, a classification algorithm is trained using machine learning. The trained algorithm is then used to classify the application. Classification based on measurable signal data (i.e., input and output data) offers higher accuracy, objectivity, and reliability, avoids the influence of malicious data manipulation, and prevents discrepancies between stated claims and actual usage, making it more suitable for industrial control applications.

[0066] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not invented by the invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0067] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An application classification method characterized by, The application comprises the following steps: acquiring input data and output data of various application programs to form a training data set, wherein the input data and output data are input signal data and output signal data of the industrial configuration software application program within a fixed scanning period; performing classification training on the training data set to obtain a classification algorithm of the application program; classifying a to-be-classified application program according to the classification algorithm to determine the category of the to-be-classified application program; the step of classifying the to-be-classified application program according to the classification algorithm to determine the category of the to-be-classified application program comprises the following steps: collecting input data and output data of a third preset number of scanning periods of the to-be-classified application program; classifying the input data and output data of the third preset number of scanning periods according to the classification algorithm to determine the category of the to-be-classified application program; the step of collecting the input data and output data of the third preset number of scanning periods of the to-be-classified application program comprises the following steps: collecting the input data and output data of the third preset number of scanning periods when the to-be-classified application program runs in a simulation environment, and / or collecting the input data and output data of the third preset number of scanning periods when the to-be-classified application program runs in a production environment.

2. The method of claim 1, wherein, The step of acquiring input data and output data of various application programs to form a training data set comprises the following steps: randomly selecting a first preset number of samples from each type of application program; for each sample, collecting input data and output data of a second preset number of scanning periods to form a training data set.

3. The method of claim 1, wherein, The step of performing classification training on the training data set to obtain a classification algorithm of the application program comprises the following steps: dividing the training data set into a training set and a test set; training the training set and the test set according to a machine learning algorithm to obtain the classification algorithm.

4. A configuration software service system, characterized by comprising: The application comprises the following steps: a cloud computing node for providing a classification service of an application program for a user, comprising: a first digital twin module for acquiring input data and output data of various application programs; a classification algorithm training module for forming a training data set from the input data and the output data, performing classification training on the training data set, and obtaining a classification algorithm of the application program; a classification module for classifying a to-be-classified application program according to the classification algorithm to determine the category of the to-be-classified application program; wherein the classification module comprises: collecting input data and output data of a third preset number of scanning periods of the to-be-classified application program; classifying the input data and output data of the third preset number of scanning periods according to the classification algorithm to determine the category of the to-be-classified application program; the step of collecting the input data and output data of the third preset number of scanning periods of the to-be-classified application program comprises the following steps: collecting the input data and output data of the third preset number of scanning periods when the to-be-classified application program runs in a simulation environment, and / or collecting the input data and output data of the third preset number of scanning periods when the to-be-classified application program runs in a production environment. Collecting the input data and the output data of the application to be classified when the application runs in a production environment for the third preset number of scanning periods.

5. The system of claim 4, wherein, The system further comprises: An end node for running the application and collecting the input data and the output data and uploading to the cloud computing node.

6. The system of claim 5, wherein, The end node comprises: An application running module for running the application; An input port sampling module for collecting the input data of each scanning period when the application runs; An output port sampling module for collecting the output data of each scanning period when the application runs; A second digital twin module for communicating with the first digital twin module and sending the input data and the output data to the first digital twin module.

7. The system of claim 4, wherein, The cloud computing node further comprises: An application configuration module for providing the user with configuration design services of the application; An application simulation module for providing the user with simulation services.

8. The system of claim 6, wherein, The cloud computing node is further configured to provide the user with recommendation and selection services of the application instance.

9. The system of claim 8, wherein, The system further comprises an edge computing node; the edge computing node comprises: A simulation edge computing node for performing simulation services; A production edge computing node for deploying the recommendation and selection services of the application instance.

10. The system of claim 9, wherein, The edge computing node comprises: A running scheduling module for scheduling the running of the application according to the resources of the end node.

11. The system of claim 9, wherein, The edge computing node further comprises: A third digital twin module for communicating with the first digital twin module and the second digital twin module and synchronizing the data of the cloud computing node, the edge computing node and the end node.

12. A storage medium containing computer-executable instructions, wherein: The computer executable instructions, when executed by a computer processor, are configured to perform the application classification method according to any one of claims 1 to 3. The computer executable instructions, when executed by a computer processor, are configured to perform the application classification method according to any one of claims 1 to 3.

Images