Hydropower station edge control system and method
By integrating PLC control capabilities and modular design into the edge controller of a hydropower station, and combining it with an artificial intelligence model, the integrated deployment of real-time control and intelligent analysis of hydropower station equipment has been realized. This solves the shortcomings of traditional edge controllers in terms of control capabilities and data fusion, and improves the stability and scalability of the system.
Patent Information
- Application Number
- CN202610338811.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-19
- Publication Date
- 2026-05-01
AI Technical Summary
Traditional hydropower station edge controllers are insufficient in terms of control capabilities, computing power expansion capabilities, interface expansion capabilities, and multi-source data fusion capabilities, making it difficult to meet the control and intelligent analysis fusion deployment needs of systems of different scales.
An improved edge controller is provided, which integrates PLC control capabilities and adopts a modular and scalable system architecture. It supports the expansion of computing resources, communication interfaces and functional modules. It performs multi-source data analysis and real-time PLC control through artificial intelligence models, forming a distributed architecture and realizing the integrated deployment of control and intelligent analysis.
It improves the stability and maintenance efficiency of hydropower station equipment, achieves high real-time and reliable equipment control, reduces the probability of unplanned downtime, and supports functional and performance expansion in complex application scenarios.
Smart Images

Figure CN121956786A_ABST
Abstract
Description
Hydropower Station Edge Control System and Method Technical Field
[0001] This application relates to the field of hydropower station control technology, and in particular to a hydropower station edge control system and method. Background Technology
[0002] The digitalization and intelligence levels of hydropower stations are continuously improving, and edge computing technology is gradually being applied to equipment condition monitoring and data analysis in the hydropower field. Typically deployed on-site for data processing, its main functions include data acquisition, protocol conversion, data preprocessing, and some local analysis. These devices generally connect to field equipment or control systems via industrial communication protocols and upload the processed data to a host server or cloud platform for further analysis. Traditional technologies already include various industrial edge computing devices or edge gateway systems. For example, data acquisition and edge analysis solutions based on embedded processors typically include a data acquisition module, a communication module, and a processing unit, mainly used to complete data aggregation, protocol conversion, and some data analysis tasks.
[0003] However, traditional edge control methods for hydropower stations are insufficient to meet the needs of hydropower systems of different scales for the integrated deployment of control and intelligent analysis. Summary of the Invention
[0004] Therefore, it is necessary to provide a hydropower station edge control system and method that can meet the needs of hydropower station systems of different scales for integrated deployment of control and intelligent analysis, in order to address the above-mentioned technical problems.
[0005] Firstly, this application proposes a hydropower station edge control system, comprising:
[0006] The multi-source data acquisition module is used to acquire multi-source data from the hydropower station.
[0007] Multiple controller nodes are communicatively connected to the multi-source data acquisition module, used to receive the multi-source data and analyze the multi-source data in real time through an artificial intelligence model to obtain evaluation results, and to perform real-time PLC control on the equipment of the hydropower station based on the evaluation results and / or control logic; the artificial intelligence model is used for fault diagnosis and prediction of the equipment of the hydropower station.
[0008] In one embodiment, the controller nodes are distributed in different physical locations and are interconnected to exchange data and schedule tasks, forming a distributed architecture.
[0009] In one embodiment, the controller node includes:
[0010] At least one edge controller, the edge controller including a real-time control kernel and an edge intelligent analysis engine; wherein, the edge intelligent analysis engine is used to output the assessment result of the health status of the hydropower station equipment based on the multi-source data; the real-time control kernel is used to perform real-time PLC control of the hydropower station equipment based on the control logic, and is also used to perform real-time PLC control of the hydropower station equipment based on the assessment result.
[0011] In one embodiment, the multi-source data acquisition module includes a vibration acquisition module, an acoustic signature acquisition module, a visual acquisition module, and an electrical quantity acquisition module.
[0012] In one embodiment, it further includes:
[0013] A host computer is connected to the edge controller. The host computer includes a display module for displaying key information of the edge controller. The key information includes at least: device status indicators, sensor readings, current alarms, and action status.
[0014] In one embodiment, it further includes:
[0015] A power supply unit is connected to the edge controller and the multi-source data acquisition module, and the power supply unit has at least two independent power supplies.
[0016] An uninterruptible power supply is connected to the edge controller and the multi-source data acquisition module.
[0017] Secondly, this application proposes a hydropower station edge control method, including:
[0018] Acquire multi-source data from hydropower stations;
[0019] The multi-source data is analyzed in real time using an artificial intelligence model to obtain evaluation results; the artificial intelligence model is used for fault diagnosis and prediction of equipment in hydropower stations.
[0020] The equipment of the hydropower station is controlled in real time by PLC based on the evaluation results and / or control logic.
[0021] In one embodiment, the artificial intelligence model includes:
[0022] A regression model is used to predict the remaining operating time of the device using the multi-source data;
[0023] A classification model is used to classify the operating status of the device using the multi-source data and determine whether the device is malfunctioning.
[0024] A deep learning model is used to determine whether there are visual anomalies in the device by analyzing the visual data in the multi-source data;
[0025] A time-series prediction model is used to determine the potential risk of equipment failure by processing time-series data from the multi-source data.
[0026] An anomaly detection model is used to detect abnormal patterns in the operation of the device using the multi-source data.
[0027] Thirdly, this application proposes a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.
[0028] Fourthly, this application proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.
[0029] The aforementioned hydropower station edge control system and method include: a multi-source data acquisition module for acquiring multi-source data from the hydropower station; multiple controller nodes, communicatively connected to the multi-source data acquisition module, for receiving multi-source data and analyzing the multi-source data in real time through an artificial intelligence model to obtain evaluation results, and performing real-time PLC control on the hydropower station equipment based on the evaluation results and / or control logic; the artificial intelligence model is used for fault diagnosis and prediction of the hydropower station equipment, thereby meeting the needs of hydropower station systems of different scales for integrated deployment of control and intelligent analysis. Attached Figure Description
[0030] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0031] Figure 1 is a schematic diagram of the overall architecture of the edge control system of a hydropower station in one embodiment;
[0032] Figure 2 is a schematic diagram of the modular hardware composition of the hydropower station edge control system in one embodiment;
[0033] Figure 3 is a schematic diagram of the modular hardware composition of the hydropower station edge control system in another embodiment;
[0034] Figure 4 is a schematic diagram of the interface distribution of the edge controller in one embodiment;
[0035] Figure 5 is a schematic diagram of the application topology of the hydropower station edge control system in one embodiment;
[0036] Figure 6 is a flowchart illustrating the edge control method for a hydropower station in one embodiment;
[0037] Figure 7 is an internal structure diagram of a computer device in one embodiment. Detailed Implementation
[0038] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings, which illustrate embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of this application will be thorough and complete.
[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
[0040] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish one element from another.
[0041] Spatial relation terms such as “below,” “under,” “below,” “under,” “above,” “above,” etc., are used herein to describe the relationship between one element or feature shown in the figure and other elements or features. It should be understood that, in addition to the orientation shown in the figure, spatial relation terms also include different orientations of the device in use and operation. For example, if the device in the figure is flipped, the element or feature described as “below,” “under,” or “below” will be oriented “above” the other element or feature. Therefore, the exemplary terms “below” and “under” can include both above and below orientations. Furthermore, the device may also include other orientations (e.g., rotated 90 degrees or other orientations), and the spatial descriptive terms used herein will be interpreted accordingly.
[0042] It should be noted that when one element is considered to be "connected" to another element, it can be directly connected to the other element or connected to the other element through an intermediary element. Furthermore, in the following embodiments, "connection" should be understood as "electrical connection," "communication connection," etc., if there is transmission of electrical signals or data between the connected objects.
[0043] When used herein, the singular forms of “a,” “an,” and “ / the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising / including” or “having,” etc., specify the presence of the stated features, wholes, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.
[0044] In traditional technologies, the level of digitalization and intelligence in hydropower stations is constantly improving. In the hydropower field, edge computing technology is gradually being applied to equipment condition monitoring and data analysis. Typically deployed on-site for data processing, its main functions include data acquisition, protocol conversion, data preprocessing, and some local analysis. These devices generally connect to field equipment or control systems via industrial communication protocols and upload the processed data to a host server or cloud platform for further analysis.
[0045] Traditional technologies typically include data acquisition modules, communication modules, and processing units, primarily used for data aggregation, protocol conversion, and some data analysis tasks. Traditional edge controllers have improved data processing capabilities compared to traditional control systems, enabling them to perform a certain degree of local computation and analysis on acquired data. However, in practical applications at hydropower stations, traditional edge controllers still have the following limitations:
[0046] Insufficient control capabilities. Existing edge controllers typically lack deterministic real-time control mechanisms, do not support PLC-style logic control and periodic scanning scheduling, and are unable to directly undertake field I / O control and equipment interlocking tasks. In practical systems, edge controllers usually need to operate in conjunction with PLCs or other control systems, with control functions and data analysis functions deployed separately, resulting in a fragmented system structure.
[0047] Limited scalability. Existing edge controllers mostly adopt a fixed hardware architecture, where computing resources, number of interfaces, and functional modules are determined during the design phase, lacking flexible computing power expansion mechanisms and modular interface expansion capabilities. When deploying more complex data analysis models or artificial intelligence algorithms, data often needs to be uploaded to servers or cloud platforms for processing due to insufficient local computing power.
[0048] It is highly dependent on external computing resources. Due to limited computing power and functional scalability, existing edge controllers typically rely on remote servers or cloud platforms to perform core calculations when running complex models and conducting multi-source data fusion analysis. In hydropower station scenarios with insufficient network stability or many network management restrictions, this mode may introduce data transmission latency, affecting real-time performance and system reliability.
[0049] The ability to fuse multi-source heterogeneous data is insufficient. Hydropower station sites involve various types of data, including electrical quantities, mechanical vibrations, temperature, pressure, switching quantities, and other physical signals. Existing edge controllers have relatively simple interface support and data fusion architectures, making it difficult to achieve unified acquisition and collaborative analysis of multi-source heterogeneous data.
[0050] While existing edge controllers possess some data processing capabilities, they still fall short in terms of control capabilities, computing power scalability, interface expansion capabilities, and multi-source data fusion capabilities, making it difficult to meet the integrated deployment requirements of control and intelligent analysis for hydropower systems of varying scales. Therefore, there is a need for an improved edge controller device that enhances its field control capabilities while retaining the advantages of edge computing, and improves computing power and functional expansion capabilities through a scalable architecture. This reduces reliance on servers or cloud platforms, enabling integrated deployment of control and intelligent analysis.
[0051] Based on the aforementioned traditional technologies, this application provides an improved edge controller device that enhances its field control capabilities while retaining the advantages of edge computing. It also improves computing power and functional expansion capabilities through a scalable architecture, reducing reliance on servers or cloud platforms and achieving integrated deployment of control and intelligent analysis. Specifically, this application belongs to the fields of industrial automation control technology, edge computing technology, and intelligent operation and maintenance technology, and relates to an edge controller device for the complex operating environment of hydropower stations. This device integrates programmable logic controller (PLC) and edge computing capabilities, combining real-time control, data acquisition, and equipment health assessment functions. It can perform multi-source data processing and model calculations on the field side, enabling status monitoring, fault early warning, and predictive maintenance of hydropower units and auxiliary control equipment.
[0052] This application applies to the intelligent upgrading of monitoring systems, utility systems, and auxiliary control equipment for hydropower station units, and is particularly suitable for industrial scenarios with high requirements for real-time performance, reliability, and environmental adaptability. Addressing the limitations of traditional edge controllers in terms of control capabilities, data fusion, and real-time response, this application integrates PLC control capabilities into the edge controller, enabling rapid deployment and access of field devices. It also constructs a modular and scalable system architecture, supporting the expansion and on-demand combination of computing resources, communication interfaces, and functional modules, thus achieving flexible deployment of control and intelligent functions.
[0053] To address the shortcomings of existing hydropower station automation systems in terms of computing power, scalability, and intelligent analysis, this application proposes a novel edge controller device aimed at achieving an intelligent upgrade of hydropower station operation control. This device integrates edge computing capabilities with the logic control functions of a traditional PLC (Programmable Logic Controller), enhancing local data analysis and equipment status assessment capabilities while ensuring real-time control and reliability. Furthermore, its modular design allows for functional and performance expansion in complex application scenarios.
[0054] It should be noted that the beneficial effects or technical problems solved by the embodiments of this application are not limited to this one, but may also be other implicit or related problems. For details, please refer to the description of the embodiments below.
[0055] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0056] In an exemplary embodiment, as shown in FIG1, a hydropower station edge control system is provided, comprising:
[0057] The multi-source data acquisition module is used to acquire multi-source data from the hydropower station.
[0058] Multiple controller nodes are connected to a multi-source data acquisition module to receive multi-source data and analyze it in real time through an artificial intelligence model to obtain evaluation results. Based on the evaluation results and / or control logic, the system performs real-time PLC control on the hydropower station equipment. The artificial intelligence model is used for fault diagnosis and prediction of the hydropower station equipment.
[0059] The multiple controller nodes can refer to control node 1 to control node N in Figure 1.
[0060] For example, the controller node of this application incorporates an edge computing platform, with an embedded computing unit built into the controller. This provides sufficient on-site data processing capabilities to run trained artificial intelligence models and achieve local analysis of key data such as equipment vibration and operating conditions. This function can be used for predictive maintenance of equipment failures, enabling early identification and warning of abnormal symptoms, thus reducing the probability of unplanned equipment downtime. The multi-source data acquisition module can collect different data from hydropower station equipment in real time using various acquisition modules or sensors.
[0061] Specifically, the computing scalability of the edge controller in the controller node of this application enables the system to cope with the needs of large-scale equipment and complex data processing. Simultaneously, by deploying AI (Artificial Intelligence) models locally, it achieves real-time data analysis and intelligent decision-making. Through this mechanism, the system can perform intelligent fault prediction and equipment status assessment without cloud dependency, improving the stability and maintenance efficiency of hydropower station equipment.
[0062] Optionally, building upon the aforementioned AI capabilities and functional module expansion capabilities, the edge controller can support high real-time control of core hydropower station equipment, such as turbines, generators, and pump sets. The system not only supports standard PLC programming languages to meet control requirements such as equipment interlocking and sequential control, but also provides flexible computing power expansion capabilities through powerful computing resources and modular design. The calculation results of the system's diverse data can be directly integrated with PLC functions and participate in the equipment control process through the edge controller's real-time logic control interface. For example, the results generated by the AI model when analyzing equipment status and predicting faults can be fed back to the PLC control system in real time, automatically adjusting equipment operating parameters or taking necessary control measures, thereby achieving more intelligent equipment management and fault prevention.
[0063] The aforementioned hydropower station edge control system acquires multi-source data from the hydropower station through a multi-source data acquisition module. Multiple controller nodes are connected to this module to receive the multi-source data and analyze it in real-time using an artificial intelligence model. This analysis yields evaluation results, and the system performs real-time PLC control of the hydropower station's equipment based on these results and / or control logic. The artificial intelligence model is used for fault diagnosis and prediction of the hydropower station's equipment. This incorporates an edge computing platform with embedded computing units within the controller, providing sufficient field data processing capabilities to run the trained artificial intelligence model and perform local analysis of key data such as equipment vibration and operating conditions. This function can be used for predictive maintenance of equipment faults, identifying and issuing early warnings of abnormal symptoms, reducing the probability of unplanned equipment downtime. Furthermore, while retaining the traditional PLC logic control capabilities, it achieves high real-time control of core hydropower station equipment such as turbines, generators, and pump sets. The system supports joint development of standard PLC programming languages and high-level languages, meeting control requirements such as equipment interlocking and sequential control. The response time reaches the microsecond level, ensuring the determinism and stability of key control logic, thereby meeting the needs of hydropower stations of different scales for the integrated deployment of control and intelligent analysis.
[0064] In one exemplary embodiment, the controller nodes are distributed in different physical locations and are interconnected for data exchange and task scheduling, forming a distributed architecture.
[0065] Optionally, to meet the growing data processing demands in complex control scenarios, this application also proposes a mechanism for horizontally scaling computing power. By deploying multiple controller nodes in the network, these nodes collaborate and share data analysis tasks, forming a horizontally scalable edge computing cluster. This approach differs from the traditional dual-master redundant backup architecture; it is a distributed architecture that actively participates in computation, enabling linear scaling of computing power and supporting the execution of more complex algorithms and larger-scale data processing.
[0066] Specifically, the working principle of the distributed architecture in this application is as follows:
[0067] Controller node deployment: In the entire control network, multiple controller nodes can be distributed in different physical locations. Each node has sufficient computing power (supporting hardware acceleration units such as CPU, GPU, and NPU) and can run some AI models or perform real-time control tasks.
[0068] Collaborative operation: The controller nodes are connected via a high-speed network to exchange data and schedule tasks. When the computing power of a controller node reaches its bottleneck, the system dynamically allocates some computing tasks to other nodes, thereby ensuring that the computing power of the entire network can be utilized efficiently and enabling real-time data processing.
[0069] Data analysis tasks are shared: Each controller node is not only responsible for the real-time control of its local equipment, but also participates in data analysis and intelligent computing. In this way, each node can process different data streams and computing tasks in parallel, such as vibration data analysis, visual image processing, and equipment status prediction, thereby supporting the execution of more complex algorithms and large-scale data processing.
[0070] Linear scaling of computing power: Because the system supports adding more controller nodes as needed, the scaling of computing power is linear, meaning that for each additional node, the overall computing power increases by a fixed percentage. This allows the system to flexibly meet the needs of applications of different scales, achieving efficient data processing from small-scale single-device monitoring to large-scale hydropower station equipment management.
[0071] In one exemplary embodiment, the controller node includes:
[0072] At least one edge controller, comprising a real-time control kernel and an edge intelligent analysis engine; wherein, the edge intelligent analysis engine is used to output an assessment result of the health status of the hydropower station equipment based on multi-source data; the real-time control kernel is used to perform real-time PLC control of the hydropower station equipment based on control logic, and also to perform real-time PLC control of the hydropower station equipment based on the assessment result.
[0073] The main control unit refers to the edge controller itself. It is a necessary core component of the entire edge control system. This unit integrates a high-performance embedded CPU board, including a multi-core processor, memory, and a large-capacity solid-state drive for storage. The main control unit comes pre-installed with an embedded real-time operating system and an edge computing software platform, ensuring that it meets both the real-time requirements of field control and supports the operating environment of intelligent algorithms. In other words, the main control unit simultaneously undertakes two types of tasks: on the one hand, it executes the deterministic logic control functions of a traditional PLC; on the other hand, it runs edge intelligent algorithm models to perform real-time analysis and decision-making on multi-source sensor data.
[0074] Optionally, referring to Figure 2, to adapt to the access requirements of different field devices, the main control unit adopts a modular design, with multiple expandable communication interface ports integrated on the outer casing. Depending on field needs, users can plug different types of interface adapter modules (such as Ethernet, serial, fiber optic, and wireless interface modules) into these ports, enabling the controller to directly connect to various industrial communication networks and devices. For example, referring to Figure 4, the controller supports RJ45 Ethernet, fiber optic SFP, RS485 / RS232 serial, CAN bus, and common industrial and IoT interfaces such as 4G / 5G and Wi-Fi. This flexible interface configuration allows the system to be seamlessly integrated into existing power plant automation systems, compatible with the communication requirements of equipment from different manufacturers, without the need to configure separate gateways or conversion devices for each new device. The main control unit also has HDMI and USB interfaces on the side, facilitating direct connection of a monitor and keyboard / mouse during maintenance and debugging, allowing it to be used as an industrial control computer to view operating status or adjust parameters.
[0075] Furthermore, referring to Figure 4, the main control unit provides local data storage expansion capabilities: a reserved solid-state drive slot is provided for installing a large-capacity SSD, capable of storing massive amounts of historical monitoring data for extended periods. This design provides data support for subsequent trend analysis and model training. In summary, the main control unit hardware platform integrates abundant computing and communication resources, forming the foundation for realizing multi-source data fusion and intelligent control functions (a necessary technical feature for achieving the main objective of this invention).
[0076] For example, for the multi-type data acquisition expansion modules of this application, i.e., multi-source data acquisition modules, the edge control system of this application adopts a modular expansion architecture to acquire diverse field signals. In addition to the interface built into the main control unit, the system provides several pluggable functional expansion modules for data acquisition and preprocessing for different signal types. These expansion modules are connected to the main control unit via a high-speed backplane bus and have plug-and-play and hot-swappable capabilities: corresponding modules are combined and installed according to monitoring needs, and new modules are automatically connected to the data bus and start working after insertion, achieving functional expansion and upgrades without changing other parts of the system. The system does not need to be shut down when replacing or adding modules, thus ensuring uninterrupted monitoring and control functions during maintenance or expansion, meeting the high reliability requirements of hydropower stations for continuous operation.
[0077] Referring to Figure 4, all expansion modules communicate with the main control unit via an internal high-speed backplane bus. This backplane employs a high-speed serial bus architecture, ensuring sufficient bandwidth and low-latency real-time communication even when multiple modules are operating in parallel and large data volumes are transmitted. Furthermore, the backplane uniformly distributes power and provides a global clock synchronization signal, ensuring that the data acquired by each module has a unified timestamp, facilitating the main control unit's time-synchronized fusion processing of multi-source information. This modular expansion design makes hardware expansion highly flexible: when a new sensor type needs to be added, simply insert the corresponding acquisition module; when more interfaces or channels are needed, simply add a new interface module, expanding functionality without major modifications to the original system. Thus, the system can be expanded to the required monitoring range and depth according to field needs.
[0078] Optionally, regarding the software architecture and intelligent algorithm flow of this application, the main control unit is divided into two main functional parts in terms of software architecture: a real-time control kernel and an edge intelligent analysis engine. The two work together to realize the intelligent control function of this invention.
[0079] The real-time control kernel (PLC logic program) is similar to the periodic scanning task of a traditional PLC, running in an embedded real-time operating system environment. It specifically handles conventional signals such as digital and analog signals from the I / O modules. The real-time kernel executes pre-programmed control logic cyclically according to a set cycle, including logical judgments, sequential control, and closed-loop regulation, to ensure millisecond-level response and determinism in the control of hydropower station equipment. For example, it is responsible for adjusting the turbine guide vane opening and generator excitation voltage in real time based on sensor inputs, and controlling the start / stop of water pumps or the opening / closing of valves. This part of the program emphasizes high-priority real-time task scheduling, ensuring that critical control commands are completed within strict time requirements, thereby maintaining the safe and stable operation of field equipment.
[0080] The edge intelligent analysis engine utilizes the multi-core processing power and embedded computing power of the main control unit to perform parallel processing and intelligent diagnosis of large amounts of data from extended modules such as vibration, acoustic signature, and vision. The analysis engine runs on an edge computing software platform and is loaded with pre-trained offline multi-source data analysis models (such as deep learning-based device fault detection models). Based on the collected data, the intelligent engine automatically calls the corresponding algorithm model for inference calculations and outputs an assessment of the device's health status. A data recognition model combining Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTMs) can be deployed in the controller to perform pattern recognition on the features of real-time collected data. Once abnormal data is detected, the possible fault type is immediately determined and an alarm is issued. A Fast Fourier Transform (FFT) is performed on the time-domain signal to obtain the spectrum, and then a machine learning model is used to identify fault characteristics such as bearing wear and rotor imbalance. All intelligent analysis processes are completed locally and in real-time at the edge, eliminating the need to upload large amounts of data to the cloud. Diagnostic results can be obtained within milliseconds of the appearance of fault symptoms, significantly shortening response time and the decision-making process.
[0081] The aforementioned real-time control kernel and intelligent analysis engine are two relatively independent yet complementary software modules: the real-time kernel ensures hard real-time control, while the analysis engine provides intelligent decision support. They communicate via shared memory or a message mechanism, and their operation is coordinated by the software scheduling logic of the main control unit. Within a single main control unit, the system prioritizes and schedules tasks: high-priority PLC control tasks are scheduled by the real-time operating system to ensure cycle stability; analysis tasks, without affecting control real-time performance, fully utilize processor idle time or other cores for parallel execution. Furthermore, if the main control unit is equipped with a multi-core CPU, control and analysis tasks can be allocated to different CPU cores for parallel execution, thus completing their respective functions without interference. This combined hardware and software scheduling mechanism ensures that critical control functions remain timely and reliable even when processing massive amounts of data.
[0082] Optionally, this application can also achieve intelligent control closed loop: This application closely links the results of the intelligent analysis engine with the PLC control logic to form an autonomous closed-loop control system on-site. For example, when the analysis engine detects abnormal signs in the equipment (such as identifying a potential fault), the main control unit will automatically execute corresponding control measures in the real-time control kernel according to a pre-set strategy: reducing the load or speed to decrease vibration, adjusting the opening of relevant valves, triggering the intervention of backup equipment, or issuing alarm prompts to maintenance personnel, etc. In this way, a local autonomous closed loop from "monitoring to diagnosis to control" is achieved, enabling timely response to abnormal situations and eliminating faults in their infancy without waiting for remote server analysis results or manual decision-making intervention.
[0083] Optionally, if the edge controller detects abnormal patterns in both vibration and acoustic data of a water pump, it can immediately reduce the operating power of that pump and activate a backup pump to prevent the fault from escalating. This millisecond-level automatic response speed is far faster than traditional methods relying on host computers or manual intervention, significantly improving the system's responsiveness and on-site operational safety. Even when communication is disrupted or connection to the dispatch master station (central control room) is lost, the edge controller can still autonomously execute predetermined control strategies, ensuring uninterrupted critical control processes.
[0084] For example, this application can also realize data storage and collaboration: the main control unit uses a built-in solid-state drive to locally store a large amount of raw data and analysis and diagnostic results, which can be used for in-depth offline analysis and model optimization in the future. At the same time, the system supports uploading selected key information to the host computer server or cloud platform through communication interfaces such as Ethernet, realizing collaborative work between the edge and the cloud. Optionally, referring to Figure 5, the communication connection between the edge controller and field devices (sensors, actuators) and the host system is shown. Among them, the Zone II access system communicates with the multi-protocol heterogeneous conversion device through IEC104, and the edge control device communicates with the multi-protocol heterogeneous conversion device through Modbus, MQTT, and IEC61850, and finally communicates and interacts with the big data integrated platform.
[0085] For example, the edge controller can periodically send equipment health assessment reports to the remote operation and maintenance center for management personnel to refer to in decision-making; the cloud can also distribute new model parameters or software updates to the edge controller to continuously improve its analytical capabilities. This architecture ensures both the independence and integrity of local control and provides an interface for centralized management and large-scale data mining. In summary, the software architecture and algorithm flow design of the main control unit integrates data acquisition, intelligent analysis, and control decision-making, which is the core guarantee for solving the deficiency of existing technologies where "analysis results cannot be directly used for field control".
[0086] In one exemplary embodiment, the multi-source data acquisition module includes a vibration acquisition module, an acoustic signature acquisition module, a visual acquisition module, and an electrical quantity acquisition module.
[0087] In one exemplary embodiment, the system further includes:
[0088] The host computer is connected to the edge controller. The host computer includes a display module for displaying key information of the edge controller. The key information includes at least: device status indicators, readings of each sensor, current alarms and action status.
[0089] Optionally, this application also includes a human-machine interface. To facilitate on-site personnel monitoring and debugging of the system, this invention designs rich local human-machine interface (HMI) methods on the edge controller. The main control unit comes with standard video and input interfaces (such as HDMI and USB interfaces), allowing maintenance personnel to connect temporary monitors and mice and keyboards when needed, using the controller as an industrial computer to directly view system operating status, sensor data, alarm information, and adjust parameters locally. This approach makes on-site debugging and fault diagnosis more intuitive and convenient.
[0090] In addition, referring to Figure 3, the system provides a dedicated display module (optional) for users to choose from. This display module is installed on the main control unit via a side slot in the chassis, sharing the backplane power supply and communication interface. Once installed, this module is equivalent to a local operation panel embedded in the control cabinet, which can display key information collected and analyzed by the main control unit in real time (such as equipment status indicators, sensor readings, current alarm and action status, etc.). The display interface also supports simple touch or button operations, allowing field personnel to perform basic operations such as mode switching, manually triggering certain control commands, and resetting alarms. Since the display module is integrated with the main control unit and installed in the control cabinet, operators do not need to carry additional terminal equipment and can directly observe the equipment operation status during inspections.
[0091] Through the above HMI design, this system can be directly viewed and controlled by on-site personnel in local mode, and also supports remote monitoring and intervention by the central control room or remote operation and maintenance center via network. The two operating modes can be flexibly switched as needed, serving as backups for each other. This improves the system's usability and maintenance efficiency. It should be noted that the HMI interface does not participate in the implementation of automatic control functions; it is an additional improvement feature for solving the main technical problem of this invention (its purpose is to optimize user experience and maintenance convenience; the system functions normally even without this module).
[0092] In one exemplary embodiment, the system further includes:
[0093] The power supply unit is connected to the edge controller and the multi-source data acquisition module. The power supply unit has at least two independent power supplies.
[0094] Uninterruptible power supply, connected to edge controller and multi-source data acquisition module.
[0095] Optionally, referring to Figures 2 and 3, the system of this application also employs redundancy and expansion design. To address the stringent reliability and availability requirements of unattended hydropower station operations, this application adopts a multi-layered redundancy design and expansion scheme in its hardware architecture to ensure continuous and stable operation even in the event of partial failures or performance bottlenecks, and to provide flexible expansion capabilities. These measures include, but are not limited to:
[0096] Power Redundancy: The system is equipped with redundant power supply modules, with two independent power supply units installed on the edge control device rack. Under normal circumstances, the two power supplies share the load; if one fails, the other can seamlessly switch over to provide all power, ensuring the main controller continues to operate without power interruption. Furthermore, a UPS (Uninterruptible Power Supply) module is installed on the backplane of the edge control device to prevent the controller from shutting down directly due to momentary power failure. Through dual power supply hot standby and UPS backup power, the system's resilience to power fluctuations and outages is enhanced. This redundant power supply design is primarily used to optimize system reliability and is considered an additional invention point (the system can still function without it, but its resilience is reduced).
[0097] Redundancy / Coordination of Main Control Units: When the computing power or resources of a single main control CPU are insufficient to handle all tasks, or to further improve system reliability, a second main control unit module can be inserted into the backplane, forming a dual-main control architecture. The two main control units are interconnected via a high-speed link on the backplane and can operate in two modes as needed: The first is a hot standby redundancy mode, i.e., one main and one standby. The primary main control unit undertakes real-time control and analysis tasks, while the standby main control unit mirrors and synchronizes the primary unit's key data and status in real time. When the primary unit fails, the standby unit can automatically take over control in a very short time, ensuring a smooth transition of control strategies and processes without affecting the normal operation of field equipment. The second is a resource coordination mode (main-main coordination), where both main control units are online simultaneously and share tasks. For example, traditional PLC logic operations can be primarily performed by one main control unit, while big data analysis and model inference tasks can be handled by the other. Both units share sensor data and control commands through a high-speed backplane channel, jointly completing the control and monitoring of field equipment. This collaborative multi-CPU architecture significantly improves the overall processing power and real-time performance of the system. When the task load increases or the algorithm complexity rises, the system performance can be linearly scaled by adding master control units and adopting task scheduling strategies. It should be noted that the dual-master control redundancy / cooperative design is one of the key innovations in this application: in the field of hydropower station field control, traditional solutions typically only use a single PLC or dual-machine hot standby, while this application introduces a cooperative parallel processing concept, which ensures reliability and enhances intelligent analysis capabilities. As an optimization method, it greatly improves the system's robustness and performance under high load and abnormal conditions.
[0098] The above redundancy design ensures the high reliability of the system in unattended, remote monitoring scenarios of hydropower stations. However, even without enabling all redundancy configurations, the multi-source fusion and intelligent control functions of this invention can be implemented on a single control unit, although the performance in terms of fault risk and performance limits is not as good as with redundancy. Therefore, these redundancy features can be regarded as optimization points of the invention, used to further improve system performance and reliability, but are not the only necessary conditions for realizing the basic functions of this application.
[0099] Furthermore, this application adopts a pluggable modular structure design, allowing users to flexibly configure input and output modules as needed. The controller can accept conventional electrical signals such as voltage, current, temperature, pressure, and switching signals, and can also acquire physical quantities required for device status diagnosis through expansion modules. This design not only enhances the system's adaptability to multiple scenarios but also greatly simplifies the expansion process and reduces system integration and maintenance costs.
[0100] Optionally, the edge controller of this application also has a rich array of industrial interface configurations, supporting multiple communication methods such as RS485, Ethernet, CAN bus, Wi-Fi, and 4G, ensuring its compatibility with the access of various new and old equipment in hydropower stations, meeting the needs of interconnection and interoperability of multi-protocol devices, and improving the overall compatibility and flexibility of the system.
[0101] Optionally, the edge controller of this application supports local human-machine interaction and remote maintenance management. The controller itself provides interfaces such as HDMI and USB, which can be connected to peripherals such as displays, mice, and keyboards, allowing on-site personnel to directly view the operating status, set parameters, or perform debugging operations. At the same time, the controller can be connected to the dispatch center or remote operation and maintenance platform via network, supporting remote configuration updates and operating status monitoring, significantly improving operation and maintenance efficiency and reducing the frequency of manual inspections.
[0102] In summary, the edge controller device provided in this application organically integrates traditional PLC control, edge intelligent analysis, modular data acquisition, communication adaptation, and remote operation and maintenance, providing strong technical support for building a future-oriented intelligent hydropower station control system.
[0103] In one embodiment, as shown in Figure 6, a hydropower station edge control method is provided, characterized by comprising:
[0104] Step 202: Obtain multi-source data from the hydropower station;
[0105] Step 204: Analyze multi-source data in real time using an artificial intelligence model to obtain evaluation results; the artificial intelligence model is used for fault diagnosis and prediction of equipment in hydropower stations.
[0106] Step 206: Perform real-time PLC control on the equipment of the hydropower station based on the evaluation results and / or control logic.
[0107] In one exemplary embodiment, the artificial intelligence model includes:
[0108] Regression models are used to predict the remaining working time of devices using multi-source data.
[0109] Classification models are used to classify the operating status of equipment using multi-source data to determine whether the equipment is malfunctioning.
[0110] Deep learning models are used to determine whether a device has visual anomalies by analyzing visual data from multiple sources.
[0111] Time series prediction models are used to identify potential risks of equipment failure by processing time series data from multiple sources.
[0112] Anomaly detection models are used to detect abnormal patterns in the operation of equipment using multi-source data.
[0113] For example, in this application, the artificial intelligence (AI) model performs local data analysis through the embedded computing unit of an edge controller. The system collects various types of data, including acoustic signature data, vibration data, visual data, and conventional electrical signal data (such as pressure, flow rate, and temperature). After preprocessing, this data is input into the trained AI model for real-time analysis.
[0114] Artificial intelligence models mainly include the following types to adapt to different data analysis needs:
[0115] Regression models are used to predict continuous values such as the remaining useful life (RUL) of equipment. For example, based on equipment vibration and temperature data, regression models can predict the remaining operating time of the equipment and identify potential failures in advance.
[0116] Classification Model: By classifying the operating status of equipment, AI models can determine whether the equipment is in a normal, warning, or faulty state. For example, based on vibration signals or soundprint data, classification models can identify whether the equipment is malfunctioning in real time and issue an alarm.
[0117] Deep learning models (such as Convolutional Neural Networks, CNNs) are used to analyze visual data (such as device images or video streams). CNN models can detect cracks, damage, or other visual anomalies on the surface of a device, providing more intuitive fault diagnosis.
[0118] Temporal prediction models (such as Long Short-Term Memory networks, LSTM): are used to process time-series data of equipment, such as changes in vibration, temperature, and pressure. LSTM models can capture time-correlated patterns during equipment operation, predict future equipment conditions, and determine potential risks of failure.
[0119] Anomaly detection models, such as Isolation Forest and One-Class SVM, are used to detect abnormal patterns in equipment operation. For example, if the vibration of the equipment exceeds the normal range, the anomaly detection model can identify this anomaly and issue a real-time warning.
[0120] The edge controller of this application boasts powerful computing capabilities, supporting hardware acceleration units such as CPUs, GPUs, and NPUs, and features a modular design that allows for on-demand expansion. This design not only enhances computing performance but also improves the controller's flexibility and scalability. Through modular expansion, the controller can flexibly adjust its computing power, functional modules, and communication interfaces according to different application scenarios to meet the needs of hydropower stations of varying sizes.
[0121] The edge controller's computing scalability enables the system to handle the demands of large-scale equipment and complex data processing, while also enabling real-time data analysis and intelligent decision-making through locally deployed AI models. This mechanism allows the system to perform intelligent fault prediction and equipment status assessment without cloud dependency, improving the stability and maintenance efficiency of hydropower station equipment.
[0122] In one specific embodiment, the hydropower station edge control system of this application adopts a modular installation method of rack + backplane. The core of the entire device is the main control unit, which is fixed in the center of the rack, and different functional expansion modules are plugged into the two sides as needed. These modules are electrically connected to the main control unit through an internal high-speed backplane. The backplane not only transmits data, but also distributes power and provides clock synchronization signals to ensure that the data collected by each module has a unified time stamp.
[0123] The main control unit is equipped with a high-performance embedded CPU board, memory, and solid-state drive, and comes pre-installed with an embedded real-time operating system and edge computing software platform. Its front panel integrates multiple communication ports, including a Gigabit Ethernet interface (for connecting to the plant network, host computer, or industrial cameras), a fiber optic interface (for long-distance connections or substation systems), a switchable RS485 / RS232 serial port, a CAN bus interface, and an industrial wireless interface supporting secure wireless protocols (for direct communication with wireless inspection robots or wireless sensors). These interface combinations can be replaced or expanded according to field needs, and the interface layout facilitates on-site wiring and maintenance. The side of the main control unit also has reserved HDMI and USB interfaces for direct connection to a monitor, mouse, and keyboard during maintenance.
[0124] On one side of the main control unit, a vibration acquisition module and an acoustic signature acquisition module can be plugged in. The vibration module is used to connect eddy current sensors and acceleration sensors installed on the bearings or base of the hydro-generator set to acquire vibration waveforms in real time. After signal conditioning and high-speed analog-to-digital conversion within the module, the signals are sent to the main control unit. The acoustic signature module connects to an industrial microphone or microphone array in the plant via a shielded cable to amplify, filter, and digitize the acquired operating sound signals before transmitting them to the main control unit for analysis. Industrial cameras can be directly connected to the main control unit via Ethernet ports, eliminating the need for an additional video acquisition module.
[0125] The main control unit can also be plugged into conventional electrical quantity I / O modules, and can be connected to conventional electrical signals via remote I / O substations. This allows for the input of various analog quantities such as voltage, current, temperature, and pressure, as well as switch signals such as circuit breaker status and gate opening. The module can also provide relay or transistor output channels for directly driving actuators such as solenoid valves in the turbine speed control system, pump start / stop control circuits, or transformer cooler fan control mechanisms.
[0126] The internal high-speed backplane data bus is divided into two paths, capable of simultaneously transmitting multiple large data streams such as vibration, sound, and video, while ensuring millisecond-level response latency. Thus, when the device is deployed within the control cabinet of a hydropower station, devices such as vibration sensors, sound collectors, cameras, and temperature sensor networks can directly connect to this edge controller. Data is fused and analyzed locally in real time, and simultaneously uploaded to the monitoring center via wired or wireless means. The industrial wireless interface can also establish an encrypted wireless connection with inspection robots, enabling the robot to instantly transmit infrared thermal images and environmental data collected back to the controller, and receive action commands issued by the controller.
[0127] When field equipment needs to be controlled, the main control unit can directly output control signals through the I / O module to perform actions such as load adjustment and start / stop switching. These actions can be executed according to traditional PLC logic or triggered directly by intelligent analysis results. If the system adopts a dual-main-control configuration, the two main control units are interconnected through a backplane to form a hot standby redundancy or collaborative working mode. Even if one fails, the other can take over control in a very short time to ensure uninterrupted operation.
[0128] Optionally, the edge control system utilizes multi-source data for intelligent analysis and directly influences control decisions. Taking vibration fault early warning for a hydroelectric generator unit as an example: the vibration module continuously collects acceleration signals from the bearing housing, while the acoustic signature module simultaneously collects the unit's operating sound. The main control unit performs fast Fourier transform and feature extraction on these signals, then inputs them into an embedded deep learning model for pattern recognition. Once abnormal frequency components or abnormal acoustic signature features are identified, the system comprehensively judges the operating condition and issues an early warning when the abnormality persists to a certain threshold. Simultaneously, it can automatically execute strategies such as load reduction and shutdown, and save key data segments for future analysis.
[0129] For main transformer monitoring, the system connects to the temperature measurement network via a wireless interface to acquire the temperature in real time, while also collecting the transformer's operating acoustic signature. If an abnormal temperature rise is detected accompanied by typical discharge noise, the system will determine that there is a risk of partial discharge, immediately notify maintenance personnel, reduce the load, and trigger backup equipment switching if necessary.
[0130] Optionally, the system of this application can be directly installed in the control cabinet of the unit layer. During installation, simply insert the main control unit and the required acquisition modules into the backplane, connect the sensors, actuators, and network optical cables or network cables, and connect the dual power supply and UPS module. After power-on, the system will automatically execute a self-test program and display the operating status of each module on the local display terminal. Maintenance personnel can connect an industrial monitor or laptop through the HDMI interface of the main control unit to enter the graphical configuration interface to perform operations such as parameter configuration of the acquisition channels, alarm threshold setting, intelligent model selection and loading.
[0131] When new expansion modules are added to the system, such as vibration acquisition modules, acoustic fingerprint acquisition modules, or additional I / O modules, the expanded modules will automatically connect to the data bus and display key data and status information from the new modules in real time on the interface, achieving functional expansion without modifying other parts. This allows maintenance personnel to immediately monitor the real-time data and health status of the newly added measurement points during expansion, enabling plug-and-play functionality and visualized management.
[0132] The module supports hot-swapping, so the system will not be interrupted when it is replaced or maintained. This ensures that core data can be continuously collected and displayed during maintenance or expansion, meeting the high reliability requirements of hydropower stations.
[0133] In summary, the technical solution proposed in this application innovatively integrates control, sensing and monitoring, and intelligent analysis. Through the collaborative design of modular hardware and edge computing software, it achieves a comprehensive upgrade of the hydropower station's on-site control device. Compared with existing technologies, this system can directly connect to various existing sensors and actuators by deploying the edge control device of this invention without changing the original power station main control system architecture. This forms a comprehensive monitoring network covering multiple parameters such as electrical parameters, mechanical status, and environmental information. It also performs real-time big data analysis and autonomous control decisions on-site, effectively overcoming the limitations of traditional solutions in terms of perception, response lag, and reliability. This solution addresses the main technical problems existing in hydropower station operation and achieves the technical goals of multi-source data fusion, intelligent diagnosis, and safe and efficient control.
[0134] The edge controller device provided in this application has significant advantages in terms of functional integration, operating efficiency, system compatibility, deployment flexibility, and intelligence level, and can effectively improve the overall performance of hydropower station automation systems.
[0135] Specifically, this application integrates the high real-time control capabilities of traditional PLCs with the data processing and intelligent analysis capabilities of modern edge computing platforms, achieving integrated on-site control and predictive maintenance. A single device can complete the entire process from signal acquisition and logic control to data analysis and intelligent decision-making, greatly simplifying the system architecture. By deploying AI models locally, the operating status and trend prediction of key equipment in hydropower stations can be assessed, providing early warnings when fault symptoms first appear, preventing the fault from escalating or causing sudden shutdowns, and significantly enhancing the system's proactive protection capabilities.
[0136] The edge controller in this application performs data processing and model inference through local computing, reducing reliance on the cloud or remote servers and avoiding the risks of latency and communication interruptions caused by large-scale data transmission, thus ensuring real-time control and response. Its millisecond-level processing speed enables the control system to have rapid response capabilities. At the same time, the predictive maintenance mechanism can schedule maintenance plans in advance, reducing unplanned downtime events and improving equipment availability and the overall operating efficiency of the hydropower station.
[0137] This application's design fully considers the characteristics of hydropower stations, such as a wide variety of equipment and a large generational span, providing multiple industrial communication interfaces that are compatible with existing mainstream protocols, ensuring interoperability with older equipment. Simultaneously, through modular design, it supports dynamic expansion of computing power and functionality, allowing users to flexibly connect various sensors or functional modules, achieving smooth upgrades of system functions, avoiding redundant investment in existing systems, and reducing transformation costs.
[0138] This application offers ease of deployment. The controller device adopts a plug-and-play modular structure, supporting flexible configuration and on-site assembly. Maintenance personnel can select the necessary modules based on site requirements, quickly expanding equipment capacity and functionality without complex wiring or large-scale adjustments. The system supports remote access and online configuration, allowing maintenance personnel to remotely perform debugging, parameter settings, and program updates without frequent on-site visits, significantly reducing maintenance costs. Simultaneously, the local HMI interface enhances the intuitiveness and efficiency of on-site operations, making it suitable for diverse application environments.
[0139] This application demonstrates advanced intelligence, leveraging an embedded AI inference engine and model management capabilities to enable the control system to learn autonomously and make intelligent judgments. It can not only detect abnormal changes in operational data in real time but also continuously optimize control strategies based on long-term operational data, gradually evolving towards adaptive and intelligent control. This capability provides a foundation for "preventative" intelligent operation and maintenance in hydropower stations, significantly improving operational safety and economy.
[0140] In summary, the edge controller device of this application, while maintaining the stability of traditional control, introduces advanced intelligent features and modular architecture, providing reliable technical support for the efficient, safe and intelligent operation of hydropower stations, and has significant practical application value and promotion prospects.
[0141] It should be understood that although the steps in the flowchart of Figure 6 are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in Figure 6 may include multiple steps or multiple stages, which are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.
[0142] In one embodiment, referring to FIG7, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.
[0143] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.
[0144] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0145] In the description of this specification, references to terms such as "some embodiments," "other embodiments," and "ideal embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative descriptions of the above terms do not necessarily refer to the same embodiments or examples.
[0146] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0147] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.
Claims
1. A hydropower station edge control system, characterized in that, include: The multi-source data acquisition module is used to acquire multi-source data from the hydropower station. Multiple controller nodes are communicatively connected to the multi-source data acquisition module, used to receive the multi-source data and analyze the multi-source data in real time through an artificial intelligence model to obtain evaluation results, and to perform real-time PLC control on the equipment of the hydropower station based on the evaluation results and / or control logic; the artificial intelligence model is used for fault diagnosis and prediction of the equipment of the hydropower station.
2. The system according to claim 1, characterized in that, The controller nodes are distributed in different physical locations and are interconnected to exchange data and schedule tasks, forming a distributed architecture.
3. The system according to claim 2, characterized in that, The controller node includes: at least one edge controller, the edge controller including a real-time control kernel and an edge intelligent analysis engine; wherein, the edge intelligent analysis engine is used to output the evaluation result of the health status of the hydropower station equipment based on the multi-source data; the real-time control kernel is used to perform real-time PLC control of the hydropower station equipment based on the control logic, and is also used to perform real-time PLC control of the hydropower station equipment based on the evaluation result.
4. The system according to any one of claims 1 to 3, characterized in that, The multi-source data acquisition module includes a vibration acquisition module, an acoustic fingerprint acquisition module, a visual acquisition module, and an electrical quantity acquisition module.
5. The system according to claim 3, characterized in that, Also includes: A host computer is connected to the edge controller, and the host computer includes a display module for displaying key information of the edge controller; The key information includes at least: equipment status indicators, sensor readings, current alarms, and operational status.
6. The system according to claim 3, characterized in that, Also includes: A power supply unit is connected to the edge controller and the multi-source data acquisition module, and the power supply unit has at least two independent power supplies. An uninterruptible power supply is connected to the edge controller and the multi-source data acquisition module.
7. A method for edge control of a hydropower station, characterized in that, include: Acquire multi-source data from hydropower stations; The multi-source data is analyzed in real time using an artificial intelligence model to obtain evaluation results; the artificial intelligence model is used to diagnose and predict faults in the equipment of the hydropower station; and the equipment of the hydropower station is controlled in real time by PLC based on the evaluation results and / or control logic.
8. The method according to claim 7, characterized in that, The artificial intelligence model includes: a regression model for predicting the remaining operating time of the device using the multi-source data; a classification model for classifying the operating status of the device using the multi-source data to determine whether the device is malfunctioning; a deep learning model for determining whether there are visual anomalies in the device by analyzing visual data in the multi-source data; a time-series prediction model for determining the potential risk of device failure by processing time-series data in the multi-source data; and an anomaly detection model for detecting abnormal patterns in the operation of the device using the multi-source data.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 7 to 8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 7 to 8.
