Real-time online decision execution SCADA system based on artificial intelligence
Through the real-time online decision-making and execution SCADA system based on artificial intelligence, the problem that traditional SCADA systems find it difficult to implement global optimization strategies has been solved, complex logic editing and efficient real-time decision execution at the system level have been realized, and the real-time performance and reliability of the SCADA system have been improved.
Patent Information
- Application Number
- CN202510790549.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-13
- Publication Date
- 2025-09-16
AI Technical Summary
Traditional SCADA systems have difficulty implementing global optimization strategies, control logic is difficult to unify, programming and networking are difficult, and it is difficult to achieve complex logic editing and customized workflows at the system level. They also lack efficient real-time decision-making and execution capabilities.
A real-time online decision-making and execution SCADA system based on artificial intelligence is adopted. The real-time data is received and logically processed through the expert system module. The module and the execution module are combined with the real-time data image to convert the reasoning results into control instructions. The web interface and distributed database are used to realize online editing and efficient reasoning of rules. The CLIPS reasoning engine and Websocket server are combined to realize low-latency interaction.
It achieves global strategy optimization at the system level, simplifies the editing and execution of complex control logic, improves the real-time and versatility of the SCADA system, supports efficient system strategy implementation and abnormal self-healing capabilities, and enhances the reliability and scalability of the system.
Smart Images
Figure CN120654831A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of industrial SCADA systems, and in particular to a real-time online decision-making and execution SCADA system based on artificial intelligence. Background Art
[0002] In the mid-20th century, the expansion of industrial scale led to the inefficiency and error-proneness of traditional manual control methods. Although relays and timers provided basic automation, they had problems such as complex configuration and difficult troubleshooting. After the introduction of computer technology, combined with microprocessor and PLC technology, it promoted the birth of SCADA systems.
[0003] Existing SCADA systems implement control logic through distributed logic controllers (PLCs). The implementation is relatively distributed, making it difficult to implement overall control operation strategies at the system level. Many overall process logics are difficult to implement, and there is no way to edit and inject complex logic. They often remain in the data monitoring link, used for data monitoring and warning issuance, and cannot customize workflows. Most systems remain in a single scenario, and programming and networking are difficult. They are difficult to interconnect with each other, making it difficult to implement unified overall control strategies.
[0004] Therefore, in response to the above problems, the present invention proposes a real-time online decision-making and execution SCADA system based on artificial intelligence. By integrating the expert system, a real-time and efficient inference engine is provided to the SCADA system, and then a WEB-based expert system knowledge editing system is constructed to provide various rule editing interfaces through the WEB method to solve various control logic implementation problems in the system implementation. Summary of the Invention
[0005] In order to overcome the problem that the decision logic of the existing SCADA system is difficult to implement, the present invention proposes a real-time online decision-making execution SCADA system based on artificial intelligence.
[0006] The technical solution of the present invention is: a real-time online decision-making and execution SCADA system based on artificial intelligence, comprising:
[0007] Expert system module, used to receive real-time data images from the SCADA system and perform logical processing through the inference engine;
[0008] Real-time data image building module, used to organize the real-time data acquired by the SCADA system into a system operation image and inject it into the expert system module;
[0009] The execution module is used to convert the processing results of the expert system module into control instructions and send them to the field equipment for execution through the SCADA system.
[0010] Preferably, the expert system module includes:
[0011] Knowledge base, used to store device operation rules and policies;
[0012] The inference engine is used to perform logical reasoning on real-time data images based on the rules in the knowledge base;
[0013] Web interface, providing online editing rules and real-time interaction functions.
[0014] Preferably, the knowledge base is implemented through a relational database or a NoSQL database, supports the addition, deletion, modification and query operations of rules, and allows users to dynamically edit and load rules in the knowledge base through a web interface. At the same time, the database adopts a distributed architecture to improve data access efficiency and system reliability.
[0015] Preferably, the real-time data image construction module parses the sensor data, equipment status information and process parameters collected by the SCADA system, organizes them into structured data images according to predefined equipment topology relationships, and associates them with logical tags in the equipment management module to provide complete factual input for the expert system.
[0016] Preferably, the system further comprises a device model building module, which defines static and dynamic properties of the device in the expert system and references these properties as variables in the rule engine, so that the control strategy can be dynamically adjusted based on the specific characteristics of the device.
[0017] Preferably, the expert system module adopts the CLIPS reasoning engine as the core algorithm, integrates an HTTP server to provide a RESTful API for the knowledge management platform to call, and deploys a Websocket server to achieve low-latency real-time data interaction.
[0018] Preferably, the execution module parses the control instructions output by the expert system, maps them into predefined command codes in the SCADA device model, and sends the instructions to the PLC or field actuator using the OPC UA or Modbus industrial protocol.
[0019] Preferably, the expert system module provides a Web front-end interface based on Vue.js or React, supports users to customize complex control strategies through a drag-and-drop rule editor or script input, and provides a version management function to record the rule change history.
[0020] Preferably, the real-time data image construction module establishes a long connection with the data acquisition service of the SCADA system through the Websocket protocol, continuously pushes structured data images to the expert system with millisecond delay, and uses lightweight JSON format to encapsulate data to reduce network load.
[0021] Preferably, the system deeply couples the symbolic reasoning capability of the CLIPS expert system with the real-time control function of the SCADA system, thereby adding global strategy optimization, abnormal self-healing and energy efficiency management functions while retaining the advantages of traditional PLC distributed control.
[0022] Beneficial effects of the present invention:
[0023] 1. By building a system-level real-time data image and injecting it into the expert system inference engine, the problem of traditional SCADA systems being difficult to implement global optimization strategies is solved.
[0024] 2. By editing and executing the SCADA system based on complex control logic reasoning at the system level, the control logic originally scattered in various control devices can be partially transplanted at the system level, solving the problem of difficulty in unifying various systems.
[0025] 3. The system simplifies the editing and injection of expert system knowledge through online editing of complex rules. Implementing the expert system background through C language greatly improves the real-time performance of the background reasoning part. Combining the output results of the expert system with the SCADA execution system solves the real-time judgment execution and customized workflow problems of the SCADA system. The entire system greatly improves the versatility of the SCADA system in handling system policy issues, fully integrates artificial intelligence and SCADA, and provides an efficient way to implement system policies. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 Shown is a schematic diagram of a system implementation of the present invention;
[0027] Figure 2 What is shown is a schematic diagram of the workflow of the present invention. DETAILED DESCRIPTION
[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.
[0029] Referring to FIG. 1 , the present invention provides an embodiment of a real-time online decision-making and execution SCADA system based on artificial intelligence, comprising:
[0030] Expert system module, used to receive real-time data images from the SCADA system and perform logical processing through the inference engine;
[0031] Real-time data image building module, used to organize the real-time data acquired by the SCADA system into a system operation image and inject it into the expert system module;
[0032] The execution module is used to convert the processing results of the expert system module into control instructions and send them to the field equipment for execution through the SCADA system.
[0033] The expert system module includes:
[0034] Knowledge base, used to store device operation rules and policies;
[0035] The inference engine is used to perform logical reasoning on real-time data images based on the rules in the knowledge base;
[0036] Web interface, providing online editing rules and real-time interaction functions.
[0037] The knowledge base is implemented through a relational database or a NoSQL database, supports the addition, deletion, modification and query of rules, and allows users to dynamically edit and load rules in the knowledge base through a web interface, ensuring that the expert system can update strategies in real time to adapt to the needs of different industrial scenarios. At the same time, the database adopts a distributed architecture to improve data access efficiency and system reliability.
[0038] The real-time data image construction module parses the sensor data, equipment status information and process parameters collected by the SCADA system, organizes them into structured data images according to predefined equipment topology relationships, and associates them with logical tags in the equipment management module to ensure that the data image can accurately reflect the system operation status and provide complete factual input for the expert system.
[0039] The system also includes an equipment model building module, which defines the static and dynamic properties of the equipment in the expert system and uses these properties as variable references in the rule engine, so that the control strategy can be dynamically adjusted based on the specific characteristics of the equipment, thereby achieving more refined industrial process optimization.
[0040] The expert system module uses the CLIPS inference engine as its core algorithm, and integrates an HTTP server to provide a RESTful API for the knowledge management platform to call. At the same time, it deploys a Websocket server to achieve low-latency real-time data interaction, ensuring that the inference engine can quickly respond to dynamic changes in the SCADA system and meet the stringent real-time requirements of industrial scenarios.
[0041] The execution module parses the control instructions output by the expert system, maps them to predefined command codes in the SCADA device model, and uses industrial protocols such as OPC UA or Modbus to send the instructions to the PLC or field actuators, ensuring that the control actions strictly match the logic of the device model, avoiding misoperation and improving system reliability.
[0042] The expert system module provides a web front-end interface based on Vue.js or React, supporting users to customize complex control strategies through a drag-and-drop rule editor or script input. It also provides a version management function to record the history of rule changes, allowing non-programmers to easily participate in the design and optimization of industrial intelligent strategies.
[0043] The real-time data image construction module establishes a persistent connection with the data acquisition service of the SCADA system through the Websocket protocol, continuously pushing structured data images to the expert system with millisecond-level delays. At the same time, it uses the lightweight JSON format to encapsulate data to reduce network load, ensuring efficient transmission and real-time processing of large-scale industrial data.
[0044] The system deeply couples the symbolic reasoning capabilities of the CLIPS expert system with the real-time control functions of the SCADA system. While retaining the advantages of traditional PLC distributed control, it adds global strategy optimization, abnormal self-healing and energy efficiency management functions, thus forming a new generation of industrial intelligent control system architecture that supports complex decision-making, high real-time response and is easy to expand.
[0045] Furthermore, the complete technical solution of the present invention is described in detail:
[0046] Build a WEB-based expert system that provides an http access interface for online editing of expert system rules, a Websocket interface for real-time online injection of real-time fact set images into the expert system inference engine, and HMI interaction.
[0047] The expert system knowledge base is managed through a database and needs to be connected to the expert system's knowledge management method to achieve knowledge storage, editing, and loading and injection.
[0048] Build a real-time data image of the SCADA system, organize the real-time data obtained by the SCADA system and associate it with the device management structure to build a real-time operation image of the system. This image is then injected into the web-based expert system as facts.
[0049] In the WEB-based expert system, the operating rules of various devices are formulated and edited online according to the device organization logic definition of the SCADA system, such as various energy-saving strategies, and saved in the knowledge database.
[0050] Build a device model in the SCADA system and combine the device control instructions with the model. The control information output by the expert system is injected into the SCADA system, and then the control information is transmitted to the on-site execution equipment through the SCADA system.
[0051] A device model is constructed in the expert system, which includes various attribute information of the device. The attribute information is used for the fact definition of various rules of the expert system.
[0052] This will enable the entire SCADA system to acquire real-time data, inject it into the expert system, process it logically, and send the processing results to execute the entire process.
[0053] Further, the specific embodiments of the present invention are described in detail:
[0054] Build a web-based expert system knowledge management background and inference engine implementation, port the CLIPS expert system, build HTTP server and Websocket server, and design the knowledge injection and interaction methods of the CLIPS expert system.
[0055] Build a knowledge management platform based on VUE architecture to manage the knowledge facts, rules and templates of the expert system, and provide online knowledge management through the WEB.
[0056] The real-time data in the SCADA system is injected into the expert system background through Websocket and the reasoning results are transmitted to the field equipment through the execution process.
[0057] By combining with the SCADA system, the system obtains the real-time data image of the system equipment, injects the data image into the expert system inference engine, makes real-time logical inferences based on the rules in the inference engine, and then transmits the logical inference results to the field equipment through the SCADA execution module. The entire system greatly facilitates the formulation of control strategies in various digital system scenarios and solves the high real-time problem, making up for the defects of the existing SCADA system. Secondly, the HMI format adopts the WEB format to facilitate its implementation in various scenarios.
[0058] Further, the workflow of the present invention is described in detail:
[0059] The SCADA system collects the operating data of field equipment (such as temperature, pressure, and current) in real time through distributed sensor networks and industrial communication protocols, and performs pre-processing such as filtering, normalization, and outlier detection on the raw data to ensure the accuracy and consistency of the input data, providing a high-quality data source for the subsequent construction of the system operation image.
[0060] The data preprocessing module organizes the collected discrete data points into a structured real-time operation image based on the device topology and logical tags defined in the device management system. The image is encapsulated in JSON or Protocol Buffers format, fully reflecting the interconnection status and operating parameters of all devices in the current system, forming a standardized fact set that can be processed by the expert system.
[0061] Through the Websocket long connection channel, the built system running image is pushed to the expert system module with a millisecond delay. The device status parameters in the image are automatically mapped to dynamic variables in the expert system fact library, and the real-time rule matching process of the inference engine is triggered at the same time.
[0062] The CLIPS inference engine loads predefined process rules (such as "if the temperature exceeds the threshold, start cooling") and equipment policies (such as the energy efficiency optimization rule set) from the knowledge base, performs pattern matching on the injected fact set, generates candidate control instruction sets through forward chain reasoning, and selects the optimal decision solution based on confidence evaluation.
[0063] The inference results are converted into standardized control instructions that can be recognized by the SCADA system and undergo redundancy verification by the security module to ensure that the instructions comply with industrial safety regulations and have no logical contradictions with the current system status.
[0064] The verified control instructions are issued through the SCADA execution module, and the instructions are transmitted to the target PLC or actuator using industrial real-time Ethernet or fieldbus, driving the field equipment to complete the action, and at the same time updating the equipment status and feeding it back to the data acquisition layer.
[0065] The new status data of field equipment is re-collected by the SCADA system, updated by the image construction module, and then injected into the expert system again, forming a closed-loop control flow of "perception-decision-execution-verification", ensuring that the system can dynamically modify strategies to adapt to changes in production conditions.
[0066] Through the Web management platform, engineers can monitor the reasoning process in real time and edit knowledge base rules online based on historical execution results. The updated rules take effect immediately after version control, achieving continuous evolution of the system's intelligence level.
[0067] When anomalies such as communication interruption or rule conflict are detected, the system automatically switches to the preset redundancy strategy, triggers an alarm to notify the operation and maintenance personnel, and analyzes the root cause of the fault through the meta-rules of the expert system, generating repair suggestions and pushing them to the HMI interface.
[0068] For multi-workshop / factory-level applications, the expert decision-making modules of each subsystem share data and coordinate strategies through the Industrial Internet platform, forming a hierarchical intelligent decision-making network, realizing full-link intelligence from device-level control to enterprise-level optimization.
[0069] Through the above steps, by integrating the expert system into the SCADA system to provide a real-time and efficient inference engine, and then building a WEB-based expert system knowledge editing system to provide various rule editing interfaces through the WEB, the problem that the traditional SCADA system is difficult to implement global optimization strategies is solved, so as to solve the problem that the existing SCADA system decision logic is difficult to implement.
Claims
1. A real-time online decision-making and execution SCADA system based on artificial intelligence, characterized by: Includes: Expert system module, used to receive real-time data images from the SCADA system and perform logical processing through the inference engine; Real-time data image building module, used to organize the real-time data acquired by the SCADA system into a system operation image and inject it into the expert system module; The execution module is used to convert the processing results of the expert system module into control instructions and send them to the field equipment for execution through the SCADA system.
2. A real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The expert system module includes: Knowledge base, used to store device operation rules and policies; The inference engine is used to perform logical reasoning on real-time data images based on the rules in the knowledge base; Web interface, providing online editing rules and real-time interaction functions.
3. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 2, characterized in that: The knowledge base is implemented through a relational database or a NoSQL database, supports the addition, deletion, modification and query operations of rules, and allows users to dynamically edit and load rules in the knowledge base through a web interface. At the same time, the database adopts a distributed architecture to improve data access efficiency and system reliability.
4. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The real-time data image construction module parses the sensor data, equipment status information and process parameters collected by the SCADA system, organizes them into structured data images according to predefined equipment topology relationships, and associates them with logical tags in the equipment management module to provide complete factual input for the expert system.
5. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1 is characterized in that: The system also includes an equipment model building module, which defines static and dynamic properties of the equipment in the expert system and uses these properties as variable references in the rule engine, so that the control strategy can be dynamically adjusted based on the specific characteristics of the equipment.
6. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The expert system module adopts the CLIPS reasoning engine as the core algorithm, integrates the HTTP server to provide RESTful API for the knowledge management platform to call, and deploys the Websocket server to achieve low-latency real-time data interaction.
7. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The execution module parses the control instructions output by the expert system, maps them into predefined command codes in the SCADA device model, and uses the OPC UA or Modbus industrial protocol to send the instructions to the PLC or field actuator.
8. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The expert system module provides a web front-end interface based on Vue.js or React, supports users to customize complex control strategies through a drag-and-drop rule editor or script input, and provides version management function to record the rule change history.
9. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The real-time data image construction module establishes a long connection with the data acquisition service of the SCADA system through the Websocket protocol, continuously pushes structured data images to the expert system with millisecond delay, and uses the lightweight JSON format to encapsulate data to reduce network load.
10. The real-time online decision-making and execution SCADA system based on artificial intelligence according to claim 1, characterized in that: The system deeply couples the symbolic reasoning capability of the CLIPS expert system with the real-time control function of the SCADA system, thereby adding global strategy optimization, abnormal self-healing and energy efficiency management functions while retaining the advantages of traditional PLC distributed control.