Knowledge graph-based algorithm model running driving engine construction method

Abstract

The application discloses a kind of algorithm model running drive engine construction method based on knowledge graph, data source, algorithm and model are respectively configured, form data source library, algorithm library and model library, select the data source, algorithm and model needed from library, and arrange according to model running rule, obtain algorithm model running rule graph atlas.Algorithm model running rule graph atlas is parsed as triple text data, triple text data is converted into formal language text, using multi-model interactive behavior verification, if rule is unqualified, then return error prompt information through interactive interface.If it is qualified, then rule text is input into task engine generator, and algorithm model drive engine is generated by generator according to rule text.The application solves the problems of difficult arrangement of algorithm and model running mode in manufacturing environment, complicated management of running rules, low execution efficiency and the like, so as to improve production efficiency and product quality.

Description

TECHNICAL FIELD

[0001] The application relates to an algorithm model running driving engine construction method based on a knowledge graph, and belongs to the technical field of artificial intelligence application. BACKGROUND

[0002] With the popularization and in-depth development of artificial intelligence technology, intelligent applications formed by artificial intelligence algorithms and models have become a key force to promote the transformation and upgrading of various industries. Under this background, manufacturing enterprises are increasingly attaching importance to and widely applying intelligent algorithms and models to empower their production and management. Algorithms and models have gradually become an important part of the core competitiveness of enterprises as a kind of enterprise soft assets. At the same time, the application scenarios of algorithms and models are gradually complex, and a large number of algorithms and models are scattered in various scenarios in the enterprise digital environment. These application scenarios not only cover various operation processes, but also involve strict data processing standards and complex model running rules. This makes it difficult to efficiently manage and maintain algorithms, models and application scenarios, and it requires high technical and management costs to manage and maintain them. Therefore, how to scientifically manage and effectively organize algorithms, models and their application scenarios has become an important challenge for enterprises. However, most enterprises currently have many limitations in managing and organizing these complex algorithms, models and their application rule systems.

[0003] Algorithm and model management is complex: with the complexity of intelligent application scenarios, enterprises use a variety of algorithms and models in different application scenarios. These algorithms and models not only have a wide variety, but also differ in implementation mechanisms, input and output requirements, parameter settings, etc., leading to management complexity.

[0004] Algorithm and model running mode arrangement is difficult: existing technologies mostly use text or table forms to display the running mode of algorithms and models, lacking intuitive visual interfaces. This makes it difficult for technical personnel to quickly arrange the running rule system of algorithms and models, and without a corresponding running mode verification mechanism, the application has a risk of error.

[0005] The implementation cycle of scene application is long and the technical cost is high: the traditional scene application implementation method requires a high technical threshold and cannot quickly adapt to changes in the production environment. When new algorithms or model running modes need to be modified or added, it is usually necessary to rewrite the code or perform complex system configuration, resulting in slow response speed and affecting production efficiency.

[0006] In summary, in the intelligent environment, various functional theme applications such as fault diagnosis and prediction of various devices, process parameter optimization, quality prediction, dynamic production scheduling, machine vision and the like need effective integration of data, algorithms and models to meet the application needs of actual algorithms and models. However, different application themes have different integration methods, which brings complexity to integration. At the same time, multiple algorithms and models running separately consume a large amount of computing resources and may cause program deadlock. In addition, frequent manual intervention in the traditional way also increases the difficulty of application implementation. SUMMARY

[0007] In view of the complicated execution mode of a large number of algorithms and models in the intelligent manufacturing environment and the difficulty in flexibly responding to the production environment which is limited in resources and full of changes, the application provides a method for constructing an algorithm model running rule driven engine based on a knowledge graph.

[0008] The technical scheme of the application is as follows:

[0009] The method for constructing an algorithm model running driven engine based on a knowledge graph comprises the following steps:

[0010] 1) configuring data, algorithms and models; the data source, algorithms and models are managed and configured respectively by a configuration management module to form a data source library, an algorithm library and a model library;

[0011] 2) constructing an algorithm model running rule graph; the required data source, algorithms and models are selected from the data source library, the algorithm library and the model library, and are arranged according to a model running rule to obtain an algorithm model running rule graph; the model running rule is set through a graphical visual operation interface;

[0012] 3) analyzing the running rule of the algorithms and models; the algorithm model running rule graph constructed in step 2) is analyzed into triple text data;

[0013] 4) consistency test of the rule; the triple text data is converted into a formal language text of description logic, and the formal language text constitutes a rule text; a multi-model interactive behavior verification based on a timed automaton is used to test the algorithm model running logic represented by the rule text, if there is a logical error or a deadlock problem in the rule in the rule text, the rule text is unqualified, and an error prompt information is returned through an interactive interface; otherwise, the rule text is determined to be qualified and enters the next step;

[0014] 5) task engine generation; the qualified rule text is input into a task engine generator, and an algorithm model driven engine is generated by the task engine generator according to the rule text.

[0015] Further, in step 1), when configuring the data source, the data interface is first configured, and the configuration content includes the use protocol and the protocol related parameters; after completing the data interface configuration, the data source of the data interface is configured, including the variable name, the data type and the physical quantity formula; after completing the configuration, the corresponding data source data is accessed, and the data value is converted by the configured physical quantity formula to form the data source library.

[0016] Further, in step 1), when configuring the algorithm, the algorithm of the algorithm library is added by uploading the algorithm file, and the algorithm name, the input parameter, the output parameter and the hyperparameter information need to be configured when uploading;

[0017] When configuring the model, the model of the model library is added by uploading the model file, and the model name, the input parameter, the output parameter and the hyperparameter information need to be configured when uploading.

[0018] Further, in step 2), the algorithm model running rule graph is constructed as follows,

[0019] 2.1) First, the trigger condition is set;

[0020] 2.2) Next, the model running after triggering and the data source input by the model are configured; the model input dimension is determined by the input parameter attribute of the model, and the configured data source and the model are displayed as nodes in the algorithm model running rule graph and connected by the "input" edge relationship;

[0021] 2.3) According to the output parameter attribute of the model, the output source node is automatically generated and connected with the model by the "output" edge relationship;

[0022] 2.4) After the configuration is completed, the end node is set, and when the running task executes to the end node, the driving task of this time is ended, and the driving task is triggered again.

[0023] Further, the trigger condition is set as time trigger or event trigger; the time trigger condition is set as interval setting time trigger, and the event trigger condition is set as triggering when a variable exceeds a set threshold.

[0024] Further, in step 5), the task engine generator generates the algorithm model driving engine as follows,

[0025] 5.1) The qualified rule text is stored in the relational database as the set algorithm and model running rule record;

[0026] 5.2) The algorithm and model running rule graph configured this time is stored in the graph database for subsequent reference and modification;

[0027] 5.3) The task engine generator will construct the corresponding table in the relational database based on the input and output of the algorithm and model in the rule text, form a task space, and store the data generated after the algorithm and model run;

[0028] 5.4) The task engine generator generates algorithm model driven engine according to the rule text, monitors the trigger condition in real time, and drives the algorithm and model in a sequential structure when the trigger response is executed.

[0029] The present application is based on the combination of algorithm library and model library with knowledge graph, which automatically arranges the running mode of algorithms and models in the system through parsing graph relationship data and executes, realizes the effective organization and optimized operation of a large number of algorithms and models based on theme rules in the intelligent manufacturing environment, solves the problems of difficult algorithm and model running mode arrangement, complicated management of running rules, low execution efficiency and other problems in the manufacturing environment, thereby improving production efficiency and product quality.

[0030] Compared with the prior art, the present application has the following beneficial effects:

[0031] 1) Configurable management of algorithms and models: Provide algorithm management and model management modules to realize the configurable management of algorithms and models. Through configuration information, the input parameters, output parameters and setting parameters of algorithms and models are clear. This modularizes the management process of algorithms and models, so that each algorithm and model has clear and unified management specifications, thereby enhancing the integration and collaborative work between different algorithms and models, improving the overall management efficiency and reducing the management complexity. At the same time, through the clear configuration information, the user can quickly trace to the specific algorithm or model, reducing errors and confusion caused by management complexity.

[0032] 2) Visual arrangement of algorithm and model running mode: Provide a visual interface for algorithm and model running mode, allowing users to add, delete or modify rules through simple operations such as drag and drop, click, etc. to improve the convenience of algorithm and model running mode arrangement. At the same time, the visualization function of knowledge graph can provide a global perspective to help technical personnel quickly understand and handle the running logic of intelligent models and algorithms, improve the scientificity and accuracy of rule making.

[0033] 3) Automatic algorithm and model running mode verification: When the algorithm and model running mode of the application scenario is arranged, the system will automatically call the built-in rule verifier to verify whether there are logical errors or deadlocks in the running mode. If problems are found, the corresponding error information will be returned to prompt the user to modify. Only when the arranged application scenario passes the verification, can it enter the next step of application generation, thereby reducing the risk of application errors.

[0034] 4) Scene application function generation: When building a new scene application or modifying an existing scene application, only configuration through the front-end interaction page is required, without the need to write back-end code. When the scene application passes the test, the system will call the task engine generator to build the task space of the application and generate the corresponding task engine, automatically driving the algorithms and models in the application to run in accordance with the arranged manner. This ensures the quick configuration of the user on the running manner of the algorithms and models, thereby making the function application adapt to changes in the production environment.

[0035] The present application introduces knowledge graph technology, and realizes the automatic management and operation of algorithms and models in a digital environment based on the combination of an algorithm library and a model library with a knowledge graph. The knowledge graph is a structured knowledge representation form that can effectively organize and manage complex data and information, and supports efficient querying and reasoning. By building a knowledge graph of a specific topic, the automatic management and execution of the running manner of various algorithms and models in the functional application process can be realized, thereby improving the overall production efficiency and product quality. The present application can provide more intelligent, efficient and flexible algorithm and model running driving methods for intelligent applications in various fields. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 - Schematic diagram of the driving engine construction process of the present application.

[0037] Figure 2 - Example scene setting flowchart of the present application.

[0038] Figure 3 - Data source configuration flowchart of the present application.

[0039] Figure 4 - Algorithm entry flowchart of the present application.

[0040] Figure 5 - Model entry flowchart of the present application.

[0041] Figure 6 - Algorithm and model running manner configuration flowchart of the present application.

[0042] Figure 7 - Algorithm and model running driving engine generation flowchart of the present application.

[0043] Figure 8 - Scene application flowchart of the present application. DETAILED DESCRIPTION

[0044] The specific embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0045] The present application is based on a knowledge graph-based algorithm model running driving engine construction method, which is described in detail in the accompanying drawings. Figure 1The driving engine construction process diagram shown,

[0046] 1) Configurable management of data, algorithms and models

[0047] The elements such as data sources, algorithms and models are managed and configured by the configuration management module, and then the corresponding data source library, algorithm library and model library are formed, and these data source library, algorithm library and model library constitute the resource library required by the driving engine of the application. When configuring the data source, the data interface needs to be configured first, and the configuration content includes the use protocol and the protocol related parameters. After completing the data interface configuration, the data source of the data interface can be configured, including variable name, data type and physical quantity formula and other information. After the configuration is completed, the corresponding data source data is accessed, and the data values collected are converted by the configured physical quantity formula to form the data source library. When configuring the algorithm, the algorithm of the algorithm library can be added by uploading the algorithm file, and the algorithm name, input parameter, output parameter, hyperparameter and other information need to be configured when uploading. The configuration of the model also adopts a similar way as the algorithm configuration, and the model of the model library is added by uploading the model file, and the model name, input parameter, output parameter, hyperparameter and other information need to be configured when uploading.

[0048] 2) Constructing algorithm model running rule atlas

[0049] The required data source, algorithm and model are selected from the data source library, algorithm library and model library, and arranged according to the model running rule to obtain the algorithm model running rule atlas; the model running rule is set through a graphical operation interface. The specific method is,

[0050] 2.1) First, the trigger condition is set, which can be set as time trigger or event trigger. For example, the time trigger condition can be set as 5-minute interval trigger, or the event trigger condition can be set as triggering when a variable exceeds the set threshold.

[0051] 2.2) Next, the model running after triggering and the data source input of the model are configured. The model input dimension is determined by the input parameter attribute of the model, and the configured input source and the model are displayed as nodes in the atlas and connected by "input" edge relationship.

[0052] 2.3) After the configuration is completed, the output source node is automatically generated according to the output parameter attribute of the model configuration, and is connected with the model by "output" edge relationship. At this time, the configuration of the model is initially completed. Subsequently, different configurations can be performed based on the model output, such as configuring the output of the model as the input of other models, or analyzing and judging the output source to perform different task operations.

[0053] 2.4) When the configuration is completed, the end node needs to be set, and when the running task executes to the end node, the driving task of this time is ended, and the driving task is triggered again.

[0054] 3) Analysis of algorithm and model running rules

[0055] The running rule knowledge graph of the algorithm and the model is obtained through graphical configuration, which belongs to graph structure data. In order to perform subsequent algorithm and model execution arrangement, the graph structure data needs to be parsed into sequential structure text data. Therefore, in this method, the running rule graph structure is parsed into triple text data, which is used for subsequent operation.

[0056] 4) Consistency test of rules

[0057] In this method, a rule tester is set to verify the running mode of the configured algorithm and model. Different model interaction mechanisms are the basis for realizing information sharing between models, and such interaction can be effectively described through logical behavior. Therefore, the triple text data is converted into a formal language text describing logic, which is used for model testing. The multi-model interaction behavior verification based on time automata is used to verify the algorithm model running logic. If there is a logical error or deadlock problem in the rule, the tester will judge it as unqualified, and return an error prompt information through the interaction interface; otherwise, the rule is judged as qualified.

[0058] 5) Task engine generation

[0059] The qualified rule text will be input into the task engine generator, which will generate an algorithm model driving engine according to the rule text. Four main operations will be performed in the generator. First, the rule text is stored in a relational database as a record of the set algorithm and model running rules. Second, the algorithm and model running graph of this configuration is stored in a graph database for subsequent consultation and modification. In the third step, the generator will construct the corresponding table in the relational database based on the input and output of the algorithm and the model in the rule text, form a task space, and store the data generated after the algorithm and model running. Finally, the generator generates an algorithm model driving engine according to the rule text, monitors the trigger condition in real time, and drives the algorithm and the model in a sequential structure when the trigger response is received, to execute the corresponding task.

[0060] The following gives specific embodiments to further understand the implementation process and logic of the present application.

[0061] This embodiment adopts B / S (Browser / Server) architecture. The front-end uses the layui framework to implement the user interface, and the back-end uses the Spring Boot framework to process business logic. In terms of data storage, MySQL is used as the relational database and Neo4j is used as the graph database to support the management and execution of algorithm model operation rules. This embodiment takes the configuration of the fault diagnosis theme function as an example to demonstrate. The flow chart is as follows Figure 2 shown.

[0062] 1. Configure the data source

[0063] In the data interface management function, users can add data interfaces by entering information such as IP address, port, user name, password and topic to access MQTT data. After the configuration is completed, a data interface record will be formed. Then, users can use the data source management function to further configure the data source information of the interface. According to the data source name, data type and physical quantity conversion formula, each data source information is set in detail. After the configuration is completed, each data source in the data interface will be connected to the system with the data converted by the physical quantity conversion formula. The data source configuration process is as follows: Figure 3 shown.

[0064] 2. Algorithm input

[0065] In the algorithm management module, users can add and configure data preprocessing algorithms and feature value extraction algorithms. Through the "Add" function in the algorithm management, users can upload algorithm files and configure the algorithm name, input parameters, output parameters, and setting parameters. After the configuration is completed, the system will generate a new algorithm record in the algorithm library. Based on the configuration information, the input form, output form, and adjustable parameters of the algorithm will be clearly determined. The algorithm entry process is as follows: Figure 4 shown.

[0066] 3. Model input

[0067] In the model management module, users can add and configure fault diagnosis models. Through the "Add" function in the model management, users can upload model files and configure the model name, input parameters, output parameters, and model setting parameters. After the configuration is completed, the system will generate a new model record in the model library. Based on the configuration information, the model's input form, output form, and adjustable parameters will be clearly determined. The model entry process is as follows: Figure 5 shown.

[0068] 4. Operation mode configuration

[0069] In the intelligent application scenario configuration function, the algorithm and model operation mode of the fault diagnosis application theme are configured.

[0070] First, set the trigger conditions for the operation mode in the application scenario map, setting a 5-minute interval as the trigger condition. Once added successfully, a "Trigger Condition" node will be generated in the map, and its "Condition Property" will be set to "Trigger every five minutes."

[0071] Next, add the data preprocessing algorithm to the scene. Once the algorithm is added, a "Data Preprocessing Algorithm" node is generated in the graph. Based on the algorithm's configuration information, the required input nodes and "Output Source 1" node are automatically generated. Algorithm parameters are set by filling in the attribute values ​​of the data preprocessing algorithm node. Configure the input source node and select the "Vibration Signal 1" data source from the data source library as the node's content. Once configured, the input node becomes the "Vibration Signal 1" data source node.

[0072] After signal preprocessing, a feature extraction algorithm is needed to extract features from the data. The next step is to add the feature extraction algorithm from the algorithm library to the graph. Successful addition will generate a "Feature Extraction Algorithm" node in the graph, along with corresponding input nodes and an "Output Source 2" node. To use the preprocessed data as a feature extraction target, configure the input source node generated by the feature extraction algorithm as an "Output Source 1" node.

[0073] Next, input the features extracted by the feature extraction algorithm (i.e., "Output Source 2") into the fault diagnosis model. Select the fault diagnosis model and add it to the graph. A "Fault Diagnosis Model" node will be generated in the graph, along with corresponding input and output nodes. Configure the input node as an "Output Source 2" node to input the feature extraction results into the fault diagnosis model. Change the output node's name property to "Diagnosis Result."

[0074] Because fault diagnosis scenarios require processing of different diagnostic results, a judgment node named "Judgment Condition" is added to the graph. The "Diagnosis Result" node is configured as the input for this judgment condition, and the judgment expression is set as a node attribute. Based on the judgment expression, corresponding branch-edge relationships are generated, including "faulty" and "no fault" branches.

[0075] Next, configure the corresponding task execution nodes. Under the "Faulty" branch, add "Task Execution Node A" and connect it to the "Judgment Condition" node via the "Faulty" edge relationship. This will serve as the task to be executed after a fault is determined. For execution content, select "Store Records" and "Push Alerts" from the execution library and configure them as two node attributes within the node. Under the "No Fault" branch, add "Task Execution Node B" and connect it to the "Judgment Condition" node via the "No Fault" edge relationship. This will serve as the task to be executed after a fault is determined. Select "Store Records" from the execution library and configure them as node attributes within the node.

[0076] So far, the logic setup of fault diagnosis is basically completed, and an end node needs to be added to the graph to end the running of the application function. After adding the end node to the graph, connect "task execution node A" and "task execution node B" to it, and automatically form the "end" edge relationship. When the application executes the task content of "task execution node A" or "task execution node B", it is considered to complete one application of the scene function. The scene application model, algorithm running mode configuration process is as shown in Figure 6 .

[0077] 5. Engine generation

[0078] When the running mode configuration is completed, the configuration content will be transmitted to the rule verifier for verification. If there is a logical contradiction or deadlock between the configured algorithm and the model running mode, it will not pass the verification, and the system will return the error reason on the front-end page, prompting the user that the theme function application setup fails and needs to be modified. If it passes the verification, the task space will be generated according to the content of the task execution node.

[0079] In this fault diagnosis scene, the fault diagnosis results will be stored and an alarm will be triggered when a fault occurs. Therefore, the task generator will create two tables in the MySQL database to store fault diagnosis results and alarm records respectively. After generating the task space, the task generator will build the algorithm model running drive engine of the scene function. The engine content is based on the parsed formal language text to generate the trigger condition monitoring method and the sequential structure task execution method. At this point, the scene application configuration is completed. The algorithm, model running drive engine generation process is as shown in Figure 7 .

[0080] 6. Scene application

[0081] When the scene application configuration is successful, the algorithm model running drive engine of the scene function will start running in the background. Before the function task is triggered, the engine only executes the trigger condition monitoring method. The trigger condition of this fault diagnosis scene configuration is triggered once every 5 minutes, so the trigger condition monitoring method of the engine will implement the timer function. When it counts to 5 minutes, the timer will be reset to zero and enter the sequential structure task execution method.

[0082] In this method, the engine will collect the data of the data source "vibration signal 1", and the collection length is consistent with the input parameter length of "data preprocessing algorithm". The processing result will be input into "feature extraction algorithm" for feature extraction. The extracted features are then input into "fault diagnosis model", and the output content will be conditionally judged: if there is a fault, the record of this diagnosis will be stored, and an alarm message will be pushed in the system platform and the alarm information will be recorded; if there is no fault, only the diagnosis record will be stored.

[0083] When any branch is executed, the engine sequential structure task execution method ends, and re-enters the trigger condition monitoring method to wait for the triggering of the next task. The user can view the records generated by each scene task space in the scene application management. The scene application flow is as shown in Figure 8

[0084] The present application aims to flexibly manage and effectively organize and run a large number of algorithms and models in a digital environment such as an intelligent manufacturing workshop and a digital twin system, so as to realize configurable automatic execution of algorithms and models in an intelligent manufacturing process. By applying knowledge graph technology to a complex digital environment, the present application can effectively integrate various data sources, construct theme-oriented algorithm and model running rules, and automatically verify and execute the running mode, so as to realize empowerment of artificial intelligence algorithms and models to various fields.

[0085] Finally, it should be noted that the above examples of the present application are merely examples for illustrating the present application, and are not intended to limit the embodiments of the present application. Although the applicant has described the present application in detail with reference to the preferred embodiments, those skilled in the art can make other different forms of changes and variations on the basis of the above description. It is impossible to enumerate all the embodiments here. Any obvious changes or variations derived from the technical solutions of the present application are still within the protection scope of the present application.​

Claims

1. A method for constructing an algorithm model-driven engine based on a knowledge graph, characterized in that: The following steps are involved: 1) Configure data, algorithms, and models: manage and configure data sources, algorithms, and models separately through the configuration management module to form a data source library, algorithm library, and model library; 2) Constructing an algorithm model operation rule map; selecting the required data sources, algorithms, and models from the data source library, algorithm library, and model library, and arranging them according to the proposed model operation rules to obtain an algorithm model operation rule map; the model operation rules are set through a graphical visual operation interface; 3) Analyze the operation rules of the algorithm and model; parse the algorithm model operation rule map constructed in step 2) into triple text data; 4) Rule consistency verification: convert triple text data into a formal language text describing logic, which constitutes the rule text. Use multi-model interactive behavior verification based on timed automata to verify the operation logic of the algorithm model represented by the rule text. If the rules in the rule text contain logical errors or deadlocks, the rule fails and an error message is returned through the interactive interface. Otherwise, the rule text is judged as qualified and proceed to the next step; 5) Task engine generation: input qualified rule text into the task engine generator, which generates an algorithm model to drive the engine based on the rule text; In step 1), when configuring the algorithm, add the algorithm to the algorithm library by uploading the algorithm file. When uploading, you need to configure the algorithm name, input parameters, output parameters, and hyperparameter information; When configuring a model, you can add a new model to the model library by uploading the model file. When uploading, you need to configure the model name, input parameters, output parameters, and hyperparameter information; In step 2), the algorithm model operation rule map is constructed as follows: 2.1) First, set the trigger conditions; 2.2) Next, configure the model to run after triggering and the data source for the model input. The model input dimension is determined by the model's input parameter properties. The configured data source and model are displayed as nodes in the algorithm model operation rule map and connected by an "input" edge relationship. 2.3) Based on the output parameter attributes of the model configuration, automatically generate output source nodes and connect them to the model with an "output" edge relationship; 2.4) After the configuration is completed, set the end node. When the running task reaches the end node, the current driving task ends and waits for the driving task to be triggered again; In step 5), the task engine generator generates the algorithm model driving engine as follows: 5.1) Storing the qualified rule text in a relational database as a record of the set algorithm and model operation rules; 5.2) Store the configured algorithm and model operation rule graph in the graph database for subsequent review and modification; 5.3) The task engine generator constructs corresponding tables in the relational database based on the inputs and outputs of the algorithms and models in the rule text, forming a task space for storing the data generated after the algorithms and models are run; 5.4) The task engine generator generates an algorithm model driving engine based on the rule text, monitors the trigger conditions in real time, and drives the algorithm and model in a sequential structure when the trigger response is triggered to execute the corresponding task.

2. The method for constructing an algorithm model operation driving engine based on a knowledge graph according to claim 1, characterized in that: In step 1), when configuring the data source, first configure the data interface, including the protocol used and protocol-related parameters. After completing the data interface configuration, configure the data source of the data interface, including the variable name, data type, and physical quantity formula. After the configuration is completed, the corresponding data source data is connected, and the collected data values ​​are converted by the configured physical quantity formula to form a data source library.

3. The method for constructing an algorithm model operation driving engine based on a knowledge graph according to claim 1, characterized in that: The trigger condition is set as time trigger or event trigger; the time trigger condition is set as interval set time trigger, and the event trigger condition is set as a variable exceeding a set threshold to trigger.

Images