Power decision-making system and method based on cloud side-end coordination

Abstract

The invention discloses a power decision-making system and method based on cloud side-end coordination, and the system comprises a terminal equipment node which is used for obtaining equipment monitoring data corresponding to power equipment, carrying out the preprocessing of the equipment monitoring data, obtaining target monitoring data, and transmitting the target monitoring data to an edge calculation node; and the edge computing node is used for determining a data analysis result corresponding to the power equipment according to the target monitoring data, and generating a short-time scheduling decision corresponding to the power equipment according to the short-time decision model, the data analysis result and the target monitoring data. Based on the above technical scheme, the equipment monitoring data corresponding to the power equipment is collected, and the edge computing node determines the short-time scheduling decision according to the processed target monitoring data, so that the scheduling decision corresponding to the power equipment is determined at the cloud side end, the decision making efficiency of the power system is improved, and the power system experience is improved. And the data processing pressure of the cloud is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management, and in particular to a power decision-making system and method based on cloud-edge-end coordination. Background Art

[0002] In the field of modern power grids, with the continuous advancement of smart grid technology, the operation and management of power grids have entered a new stage. Smart grids monitor and manage the operation status of power grids in real time by introducing a large number of sensors, smart devices and user terminals in various links of power transmission, distribution and user end.

[0003] However, traditional power grid management systems usually rely on a centralized data processing model, that is, all data needs to be uploaded to a central server or cloud for centralized processing; however, the centralized processing model relies on cloud computing resources, and as the amount of data increases, the processing capacity of the cloud also faces bottlenecks. The increased complexity of data processing makes it difficult for a single centralized computing resource to effectively cope with high-concurrency data processing needs. Summary of the invention

[0004] The present invention provides an electric power decision-making system and method based on cloud-edge-end coordination, so as to realize the dispatching decision of electric power equipment through cloud-edge-end computing, thereby improving the decision-making efficiency of the electric power system.

[0005] According to one aspect of the present invention, a power decision-making system based on cloud-edge-end coordination is provided, the system comprising: a terminal device node and an edge computing node; wherein,

[0006] The terminal device node is used to obtain device monitoring data corresponding to the power equipment, pre-process the device monitoring data to obtain target monitoring data, and send the target monitoring data to the edge computing node;

[0007] The edge computing node is used to determine a data analysis result corresponding to the power equipment according to the target monitoring data, and generate a short-term scheduling decision corresponding to the power equipment according to a short-term decision model, the data analysis result and the target monitoring data;

[0008] Among them, the short-term decision-making model is a lightweight neural network model trained based on historical monitoring data and data analysis results.

[0009] According to another aspect of the present invention, a power decision method based on cloud-edge-end coordination is provided, and the method is applied to a power decision system based on cloud-edge-end coordination, and the system includes: a terminal device node and an edge computing node; wherein the method includes:

[0010] Acquire device monitoring data corresponding to the power equipment through the terminal device node, pre-process the device monitoring data to obtain target monitoring data, and send the target monitoring data to the edge computing node;

[0011] The edge computing node determines a data analysis result corresponding to the power equipment according to the target monitoring data, and generates a short-term scheduling decision corresponding to the power equipment according to a short-term decision model, the data analysis result and the target monitoring data;

[0012] Among them, the short-term decision-making model is a lightweight neural network model trained based on historical monitoring data and data analysis results.

[0013] The technical solution of the embodiment of the present invention obtains the equipment monitoring data corresponding to the power equipment through the terminal device node, and pre-processes the equipment monitoring data to obtain the target monitoring data, and sends the target monitoring data to the edge computing node, and then the edge computing node determines the data analysis results corresponding to the power equipment according to the target monitoring data, and generates a short-term scheduling decision corresponding to the power equipment according to the short-term decision model, the data analysis results and the target monitoring data. Based on the above technical solution, by collecting the equipment monitoring data corresponding to the power equipment, the edge computing node determines the short-term scheduling decision according to the processed target monitoring data, and then realizes the determination of the scheduling decision corresponding to the power equipment at the cloud edge, improves the efficiency of the power system in making decisions, and reduces the pressure on the cloud to process data.

[0014] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 It is a structural diagram of a power decision-making system based on cloud-edge-end coordination provided according to an embodiment of the present invention;

[0017] Figure 2 It is a structural diagram of a power decision-making system based on cloud-edge-end coordination provided according to an embodiment of the present invention;

[0018] Figure 3It is a flow chart of a power decision-making method based on cloud-edge-end coordination provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] Embodiment 1

[0022] Figure 1 The structure diagram of a power decision-making system based on cloud-edge-end coordination provided by an embodiment of the present invention. Figure 1 As shown, the system includes: a terminal device node and an edge computing node; wherein,

[0023] The terminal device node is used to obtain the equipment monitoring data corresponding to the power equipment, pre-process the equipment monitoring data to obtain the target monitoring data, and send the target monitoring data to the edge computing node; the edge computing node is used to determine the data analysis results corresponding to the power equipment according to the target monitoring data, and generate the short-term scheduling decision corresponding to the power equipment according to the short-term decision model, the data analysis results and the target monitoring data;

[0024] Among them, power equipment can be equipment that performs power transmission and distribution, voltage conversion, protection and control in the power system. For example, power equipment can be divided into primary equipment and secondary equipment. Primary equipment directly participates in the production, conversion, transmission, distribution and consumption of electric energy, such as generators, transformers, circuit breakers, etc.; secondary equipment is used to protect, measure, monitor and control the operating status of primary equipment, such as measuring meters, relay protection devices, etc. Equipment monitoring data can be understood as data generated by power equipment during operation. Target monitoring data can be data obtained after processing equipment monitoring data. The short-term decision model is a lightweight neural network model trained based on historical monitoring data and data analysis results.

[0025] Specifically, the terminal device includes a data acquisition module, a preprocessing module and a dynamic monitoring module; the data acquisition module is used to collect real-time environmental data and user behavior data; the data preprocessing module is used to denoise and format convert the collected data; the dynamic monitoring module is used to monitor data changes and transmit the monitored data to the edge computing node; the edge computing node includes a data processing engine, a dynamic decision module, a feedback adjustment module and a cache and data synchronization module; the data processing engine receives real-time data from the terminal device, and performs preliminary processing and dynamic analysis on the received data based on preset rules and models; the dynamic decision module generates preliminary decision recommendations based on a lightweight machine learning model combined with real-time data; the feedback adjustment module dynamically adjusts processing rules and parameters based on changes in real-time data; the cache and data synchronization module locally caches the received real-time data and synchronizes the cached real-time data with the cloud platform.

[0026] On the basis of the above technical solution, the terminal device node includes: a detection data acquisition module and a data preprocessing module; wherein the data acquisition module is used to collect equipment monitoring data corresponding to the power equipment based on a preset data acquisition cycle, wherein the equipment monitoring data includes operating status data, operating environment data and power behavior data; the data preprocessing module is used to perform data preprocessing on the equipment monitoring data to determine the target monitoring data, wherein the data preprocessing includes data screening and format standardization.

[0027] Specifically, the data acquisition module integrates various types of sensors and user terminal devices, and can obtain data from all aspects of the power grid in real time. These data include but are not limited to the power grid operation status (such as voltage, current, temperature), equipment operation information (such as equipment health status, load status), environmental parameters (such as humidity, air pressure) and user behavior data (such as power consumption mode, power consumption time period, etc.). The data acquisition module can collect data at a high frequency to ensure that the acquired data can fully reflect the real-time operation status of the power grid. High-frequency collection also provides a basis for subsequent real-time analysis and decision-making. It should be noted that the data acquisition module supports multiple data acquisition protocols, such as Modbus, IEC 61850, etc., to ensure compatibility with different devices and systems, and to ensure the breadth and accuracy of data collection. The data acquisition module covers key devices and nodes in the power grid through distributed deployment, ensuring the comprehensiveness and real-time nature of the data; multi-protocol support enhances the flexibility of the system, can adapt to various types of power equipment and sensors, and ensures the diversity of data sources.

[0028] The data preprocessing module performs denoising on the collected data, filters out noise data caused by equipment failure, environmental interference, etc., and preprocesses the data to ensure the accuracy and reliability of the data for subsequent processing; and can identify and exclude abnormal data, for example, extreme values ​​caused by instantaneous equipment failure or other unexpected situations. By filtering abnormal data, the consistency and availability of the data are guaranteed. In order to adapt to different processing links, the preprocessing module is also responsible for converting the collected data into a unified format. The format conversion includes data standardization, unit conversion, and data type conversion (such as converting analog signals into digital signals), thereby ensuring the compatibility and efficiency of subsequent processing. The data preprocessing module greatly improves the data quality, reduces the interference of noise and abnormal data on subsequent analysis, and enhances the stability of the system; the format conversion function ensures the standardization of data, so that the data can be seamlessly transmitted and processed between different processing nodes, such as edge computing nodes and cloud platforms. It should be noted that the environmental data and behavioral data collected in real time by the data acquisition module are transmitted to the edge computing node through the MQTT protocol; and the data page processed by the preprocessing module is transmitted to the edge computing node through the MQTT protocol.

[0029] Based on the above technical solution, the terminal device node includes: a data monitoring module and a data upload module; wherein the data monitoring module is used to determine the data change trend according to the target monitoring data, and determine the power equipment status based on the data change trend; the data upload module is used to send the target monitoring data to the edge computing node when the power equipment status is abnormal.

[0030] Specifically, the data monitoring module monitors the collected data in real time, identifies the changing trend of the data, such as voltage fluctuations, load changes, etc., and predicts possible anomalies or faults in advance by monitoring the data changes; and can automatically trigger data transmission according to preset conditions (such as data exceeding a certain threshold or a specific change pattern), and immediately transmit key data or abnormal data to the edge computing node for further processing. The above data processing mechanism ensures the efficiency and response speed of the system; when key changes or anomalies are detected, the data monitoring module immediately transmits the data to the edge computing node. The transmitted data is preprocessed to ensure the accuracy and timeliness of the data. For example, the data monitoring module monitors data changes in real time. If the detected data exceeds the preset threshold or anomalies occur, the terminal device immediately sends the alarm or event data to the edge computing node through the MQTT protocol. After the edge computing node receives the alarm information, it initiates dynamic analysis and decision-making, generates decision suggestions through the dynamic decision module, and then enables the dynamic decision module to generate decision suggestions and send the decision suggestions to the terminal device through the MQTT protocol. The terminal device executes relevant decisions based on the received decision suggestions.

[0031] On the basis of the above technical solution, the edge computing node includes: a data analysis module and a short-term decision module; wherein the data analysis module is used to call the data detection algorithm corresponding to the target monitoring data from the rule database according to the target monitoring data, and determine the data analysis result corresponding to the power equipment according to the data detection algorithm; the short-term decision model is used to input the data analysis result and the target monitoring data into the short-term decision model to determine the short-term scheduling decision corresponding to the power equipment.

[0032] Specifically, the data processing engine receives pre-processed data from the terminal device and further processes it according to the system preset rules and models, including further filtering of the data, such as removing redundant information, data aggregation, such as comprehensive processing of data from multiple sensors, and simple real-time analysis, such as trend analysis, pattern recognition, etc. It should be noted that by pre-setting a rule database, the rule database is used to store data processing rules and algorithm models corresponding to different data types, and then call the corresponding data monitoring algorithm from the rule database according to the data type. For example, for data such as voltage and current, specific threshold filtering and anomaly detection algorithms are applied; for environmental data, trend analysis models are applied. Furthermore, the short-term decision module is based on lightweight machine learning models, which are optimized to run efficiently on edge nodes with relatively limited computing resources. The model can process the analysis results from the data processing engine and generate preliminary decision recommendations in combination with real-time data. It should be noted that the short-term decision model is suitable for generating short-term decisions, such as local adjustment of power grid load, issuance of equipment control commands, etc. The above decisions can be completed within milliseconds to seconds and executed immediately, which significantly improves the response speed and reliability of the system. The short-term decision module also has self-learning capabilities and can adaptively adjust as environmental data changes, and maintain the effectiveness of the decision model through regular or real-time model updates.

[0033] The technical solution provided by the embodiment of the present invention deploys lightweight models on edge computing nodes, so that the short-term decision-making module can run efficiently on edge nodes with limited resources and generate accurate decision suggestions in a timely manner; through self-learning and model updating, the system can dynamically adapt to changing environmental conditions to ensure that reliable decisions can be provided in different scenarios.

[0034] On the basis of the above technical solution, the edge computing node also includes: a feedback adjustment module and a data synchronization module, wherein the feedback adjustment module is used to execute short-term scheduling decisions, obtain scheduling feedback data corresponding to the short-term scheduling decisions, and correct the short-term scheduling decisions according to the scheduling feedback data; the data synchronization module is used to store the target monitoring data and data analysis results in the local database, and synchronously send the target monitoring data and data analysis results to the cloud platform.

[0035] Specifically, the feedback adjustment module monitors the effects of data processing and decision-making in real time, and collects feedback data after the decision is executed. The effectiveness of the current processing rules and decision-making models is evaluated by analyzing the feedback data; and the data processing rules and model parameters can be dynamically adjusted based on the feedback data. For example, if a certain processing rule does not work well in actual applications, the decision effect can be optimized by calculating the loss value or adjusting the parameters of the rule according to the feedback parameters or selecting a more appropriate processing model. The data synchronization module is responsible for caching real-time data on the local edge node to ensure that the data can be quickly accessed and processed; this local caching mechanism reduces the dependence on the remote cloud and improves the processing speed of the system; the data synchronization module is also responsible for synchronizing the cached real-time data with the cloud platform. The data synchronization mechanism ensures that the cloud has global control over all edge node data, enabling the cloud to perform comprehensive analysis and intelligent decision-making. It should be noted that the data synchronization module synchronously sends the target monitoring data and data analysis results to the cloud platform through the RESTful API or WebSocket channel.

[0036] Based on the above technical solution, Figure 2 As shown, the system also includes: a cloud platform; wherein the cloud platform is used to obtain data analysis results and target monitoring data uploaded by each edge computing node, and determine the global equipment scheduling decision corresponding to the target area based on the global decision model.

[0037] Specifically, the cloud platform includes a dynamic data analysis module, a global decision-making module, a model update module and a global scheduling module. The dynamic data analysis module uses big data analysis technology to comprehensively analyze the data uploaded by the edge computing nodes to identify long-term trend changes or abnormal situations; the global decision-making module uses a global decision-making model to generate global intelligent decisions based on the dynamic data analysis results; the model update module regularly updates the machine learning model and the global decision model, and sends the updated machine learning model to the edge computing node for updating; the global scheduling module is used for global resource scheduling and optimization, and adjusts the allocation of resources and task priorities of the entire system according to dynamic data and decision results.

[0038] On the basis of the above technical solution, the cloud platform includes: a dynamic data analysis module and a global decision-making module; wherein the dynamic data analysis module is used to obtain the target monitoring data and data analysis results corresponding to each edge computing node, and integrate the target monitoring data and data analysis results corresponding to each edge computing node to determine the global monitoring data, and determine the global data analysis results based on the global monitoring data; the global decision-making module is used to process the global data analysis results and the global monitoring data according to the global decision-making model, and determine the global equipment scheduling decision corresponding to the target area.

[0039] Specifically, the dynamic data analysis module first receives and integrates data from multiple edge computing nodes. The above data may include real-time operation data, historical data, and analysis results after preliminary processing by edge nodes. Then, in the process of data integration, the module will remove duplicates, unify data formats, align time series, and other processes to ensure data consistency and availability. It should be noted that the dynamic data analysis module uses big data analysis technology, such as distributed computing frameworks (such as Hadoop, Spark) and data mining algorithms (such as clustering analysis, association rule mining) to conduct in-depth analysis of massive data of the power grid. The goals of the analysis include identifying long-term trend changes, abnormal conditions (such as potential faults, load abnormalities), and changes in other key indicators of the power grid. Through in-depth analysis of the data, the module can identify long-term trends and potential risks that may exist in the power grid. For example, identifying the periodic trend of load changes with seasons, or detecting abnormal modes of equipment operation, provides important basis for subsequent intelligent decision-making. The global decision module makes global intelligent decisions based on the global decision model. The global decision model can optimize the decision strategy by continuously learning the operating status of the power grid and the historical decision effects. The model input includes comprehensive data analysis results from the dynamic data analysis module, and the output is a specific decision plan; the intelligent decision engine takes into account the overall operating status of the power grid and generates decision plans including resource scheduling, load balancing, equipment maintenance, etc. For example, when the power grid load is too high, the global decision module may suggest adjusting part of the load to the backup line, or dispatching backup generators to balance the power supply. The global decision module has adaptive capabilities and can continuously adjust and optimize decision strategies under new data and environments. This feature enables the system to make the best global decision when facing different power grid operating conditions.

[0040] Based on the above technical solution, the cloud platform also includes: a global scheduling module and a model update module; wherein the global scheduling module is used to schedule the power equipment in the target area according to the global equipment scheduling decision; the model update module is used to adjust the short-term decision model and the global decision model according to the preset model update cycle, and send the adjusted short-term decision model to each edge computing node.

[0041] Specifically, the model update module is responsible for regularly updating the machine learning model and reinforcement learning model in the cloud. It should be noted that the model may become invalid or degrade over time due to changes in the power grid environment or the addition of new data, so it needs to be updated regularly to maintain its effectiveness. The updated model is not only applied in the cloud, but also sent to each edge computing node. This ensures the consistency and real-time performance of the entire system from cloud to edge, so that edge nodes can also process data and make preliminary decisions based on the latest model. Before updating the model, the module will conduct comprehensive testing and optimization to ensure that the new model can perform well in the actual environment. The test content includes the accuracy, robustness and performance of the model under different loads.

[0042] The global scheduling module is responsible for optimizing and scheduling system resources according to the global decisions generated by the cloud, including adjusting the priority of each task, allocating computing resources, optimizing the allocation of power resources, etc. The module ensures the best use of resources and the efficiency of power grid operation through global control of system resources; the module dynamically adjusts the priority of tasks according to the real-time needs and decision suggestions of the power grid; for example, in the event of an emergency, the module may prioritize the allocation of resources for fault handling and postpone the execution of non-critical tasks; the global optimization scheduling module is also responsible for optimizing the overall performance of the system to ensure that the response speed and stability of the power grid system can be guaranteed under different load conditions; the module achieves load balancing and performance improvement of the power grid system through dynamic adjustment of resources.

[0043] The present invention realizes efficient data processing and intelligent decision-making between terminal devices, edge computing nodes and cloud platforms. By working together, each part not only reduces the processing burden of a single node, but also improves the response speed and decision-making quality of the system, adapting to the high concurrency and high complexity requirements of modern smart grids.

[0044] It should be noted that TLS (Transport Layer Security) or SSL (Secure Sockets Layer) encryption protocols are used in all communication links between the cloud platform, edge computing nodes, and terminal devices. The TLS / SSL protocol can provide encrypted transmission of data during the communication process to prevent data from being eavesdropped or tampered with during transmission; in order to ensure the security of the communication link, the system will use the latest version of the TLS protocol and adopt strong encryption algorithms (such as AES-256) and certificate verification mechanisms to ensure the confidentiality, integrity, and identity authentication of the communication. In addition, OAuth2.0 or JWT (JSON Web Token) is used for identity authentication to ensure that only authenticated terminal devices and edge computing nodes can communicate with the cloud platform; OAuth2.0 is a widely used authorization framework that allows third-party applications to access the resources of the resource server with the permission of the resource owner. The OAuth2.0 framework is used in the system to manage access tokens to ensure that only legitimate entities can access the resources of the cloud platform. Among them, WT is a compact and self-contained token format that is often used to transmit authenticated identity information between different services. JWT can be used in the system to sign and verify transmitted tokens to ensure that each request in the communication carries legitimate identity information.

[0045] For example, the functional modules of edge computing nodes and cloud platforms are implemented as independent microservices, and each microservice can be independently developed, deployed and expanded; different microservices are exposed to external systems through API gateways, where API gateways are responsible for request routing, authentication, load balancing and current limiting. Each microservice implements specific functions, such as data processing, decision generation, model update, global scheduling, etc. Microservices run independently of each other and communicate through well-defined API interfaces. Communication between microservices uses lightweight protocols (such as HTTP / HTTPS) and is remotely called through RESTful API or gRPC. API gateways are introduced into the system as the interaction entry point between microservices and external systems. API gateways are responsible for handling routing, authentication, load balancing and current limiting of external requests. API gateways simplify the interaction between clients and microservices by distributing requests to corresponding microservices and provide unified security controls (such as authentication and authorization).

[0046] In microservices, services are dynamically registered and discovered through the service discovery mechanism, so that each service can dynamically call other services as needed. When each microservice starts, it will automatically register its own information (such as service name, IP address, port number, etc.) with the service registry. The service registry usually uses a distributed system (such as Consul, Eureka or Zookeeper) to store and manage this registration information; microservices report their own status to the service registry by sending heartbeat signals regularly. If a microservice is down or unavailable, the service registry will automatically remove it from the registration list. When a microservice needs to call another microservice, it can obtain the latest location information (such as IP and port) of the target service through the service registry to achieve dynamic discovery of services. The service discovery mechanism makes the calls between microservices independent of fixed IP and port configurations, and can dynamically adjust the call targets according to actual conditions, thereby improving the flexibility and robustness of the system.

[0047] Alternatively, container orchestration tools can be used to define the resource requirements of services to achieve automatic adjustment and dynamic expansion of services. Based on the resource requirements and the service and storage conditions of the cluster, the container orchestration tools automatically allocate resources and deploy them to suitable nodes. The system uses container orchestration tools (such as Kubernetes and Docker Swarm) to manage the deployment and expansion of microservices. The container orchestration tool can automatically adjust the number and deployment location of service instances according to the defined resource requirements. The container orchestration tool monitors the resource usage (such as CPU and memory) in the cluster, and automatically allocates resources according to the load conditions to deploy service instances to suitable nodes. By setting automatic expansion policies, the system can dynamically increase or decrease the number of service instances according to the real-time load of the service. For example, when the number of requests for a service increases, the container orchestration tool will automatically start a new service instance to cope with the increased load. By introducing secure communication and authentication mechanisms, microservice architectures, service discovery mechanisms, and container orchestration and dynamic expansion technologies, the system can achieve efficient, flexible, and scalable cloud-edge collaborative data processing and decision-making while ensuring security. The integration of these technical solutions ensures the stable operation of the system in a complex environment and lays a solid foundation for future expansion and optimization.

[0048] The technical solution of the embodiment of the present invention obtains the equipment monitoring data corresponding to the power equipment through the terminal device node, and pre-processes the equipment monitoring data to obtain the target monitoring data, and sends the target monitoring data to the edge computing node, and then the edge computing node determines the data analysis results corresponding to the power equipment according to the target monitoring data, and generates a short-term scheduling decision corresponding to the power equipment according to the short-term decision model, the data analysis results and the target monitoring data. Based on the above technical solution, by collecting the equipment monitoring data corresponding to the power equipment, the edge computing node determines the short-term scheduling decision according to the processed target monitoring data, and then realizes the determination of the scheduling decision corresponding to the power equipment at the cloud edge, improves the efficiency of the power system in making decisions, and reduces the pressure on the cloud to process data.

[0049] Embodiment 2

[0050] Figure 3 A flow chart of a power decision-making method based on cloud-edge-end coordination provided for an embodiment of the present invention. This embodiment can be applicable to the case where the corresponding short-term scheduling decision is determined based on the collected equipment monitoring data corresponding to the power equipment, and the method can be executed by a power decision-making system based on cloud-edge-end coordination, and the power decision-making system based on cloud-edge-end coordination can be implemented in the form of hardware or software. Specifically, the system includes: a terminal device node and an edge computing node; wherein the method includes:

[0051] S110, acquiring equipment monitoring data corresponding to the power equipment through the terminal device node, preprocessing the equipment monitoring data to obtain target monitoring data, and sending the target monitoring data to the edge computing node;

[0052] S120, the edge computing node determines a data analysis result corresponding to the power equipment according to the target monitoring data, and generates a short-term scheduling decision corresponding to the power equipment according to a short-term decision model, the data analysis result and the target monitoring data;

[0053] Among them, the short-term decision-making model is a lightweight neural network model trained based on historical monitoring data and data analysis results.

[0054] The present invention effectively solves the bottleneck problem of single centralized computing resources in high-concurrency data processing through a dynamic data processing and decision-making system that collaborates with the cloud, edge and end. The system uses terminal devices to collect and pre-process data in real time, and transmits the preliminarily processed data to the edge computing node, reducing the computing and transmission burden on the cloud. The edge computing node uses a data processing engine and a lightweight machine learning model to achieve rapid dynamic analysis and decision-making, reducing the pressure on response time and network bandwidth. In addition, the feedback adjustment module adjusts the processing rules according to real-time data to ensure that the system can flexibly respond to dynamic changes in the power grid. The cloud platform is responsible for global data analysis, intelligent decision generation and model updates to ensure that the system has global optimization capabilities when processing complex tasks. Through the collaborative work of the cloud, edge and end, the system can efficiently allocate computing resources and quickly respond to the dynamic needs of the power grid, solving the delay and efficiency problems of traditional centralized systems in high-concurrency environments.

[0055] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0056] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A power decision-making system based on cloud-edge-end coordination, characterized in that The system includes: a terminal device node and an edge computing node; wherein, The terminal device node is used to obtain device monitoring data corresponding to the power equipment, pre-process the device monitoring data to obtain target monitoring data, and send the target monitoring data to the edge computing node; The edge computing node is used to determine a data analysis result corresponding to the power equipment according to the target monitoring data, and generate a short-term scheduling decision corresponding to the power equipment according to a short-term decision model, the data analysis result and the target monitoring data; Among them, the short-term decision-making model is a lightweight neural network model trained based on historical monitoring data and data analysis results.

2. The system according to claim 1, characterized in that The terminal device node includes: a detection data acquisition module and a data preprocessing module; wherein, The data acquisition module is used to collect equipment monitoring data corresponding to the power equipment based on a preset data acquisition cycle, wherein the equipment monitoring data includes operating status data, operating environment data and power behavior data; The data preprocessing module is used to perform data preprocessing on the equipment monitoring data to determine the target monitoring data, wherein the data preprocessing includes data screening and format standardization.

3. The system according to claim 2, characterized in that The terminal device node includes: a data monitoring module and a data uploading module; wherein, The data monitoring module is used to determine the data change trend according to the target monitoring data, and determine the state of the power equipment based on the data change trend; The data uploading module is used to send the target monitoring data to the edge computing node when the power equipment is in an abnormal state.

4. The system according to claim 1, characterized in that The edge computing node includes: a data analysis module and a short-term decision module; wherein, The data analysis module is used to call a data detection algorithm corresponding to the target monitoring data from a rule database according to the target monitoring data, and determine a data analysis result corresponding to the power equipment according to the data detection algorithm; The short-term decision model is used to input the data analysis results and the target monitoring data into the short-term decision model to determine the short-term scheduling decision corresponding to the power equipment.

5. The system according to claim 4, characterized in that The edge computing node further includes: a feedback adjustment module and a data synchronization module, wherein: The feedback adjustment module is used to execute the short-term scheduling decision, obtain scheduling feedback data corresponding to the short-term scheduling decision, and modify the short-term scheduling decision according to the scheduling feedback data; The data synchronization module is used to store the target monitoring data and the data analysis results in a local database, and synchronously send the target monitoring data and the data analysis results to a cloud platform.

6. The system according to claim 5, characterized in that The data synchronization module synchronously sends the target monitoring data and the data analysis results to the cloud platform through a RESTful API or a WebSocket channel.

7. The system according to claim 1, characterized in that The system also includes: a cloud platform; wherein, The cloud platform is used to obtain the data analysis results and target monitoring data uploaded by each edge computing node, and determine the global equipment scheduling decision corresponding to the target area according to the global decision model.

8. The system according to claim 7, characterized in that The cloud platform includes: a dynamic data analysis module and a global decision-making module; wherein, The dynamic data analysis module is used to obtain the target monitoring data and data analysis results corresponding to each edge computing node, integrate the target monitoring data and data analysis results corresponding to each edge computing node to determine the global monitoring data, and determine the global data analysis results based on the global monitoring data; The global decision module is used to process the global data analysis results and the global monitoring data according to the global decision model to determine the global equipment scheduling decision corresponding to the target area.

9. The system according to claim 8, characterized in that The cloud platform also includes: a global scheduling module and a model updating module; wherein, The global scheduling module is used to schedule the power equipment in the target area according to the global equipment scheduling decision; The model updating module is used to adjust the short-term decision model and the global decision model according to a preset model updating cycle, and send the adjusted short-term decision model to each edge computing node.

10. A power decision-making method based on cloud-edge-end coordination, characterized in that: The method is applied to a power decision-making system based on cloud-edge-end coordination, the system comprising: a terminal device node and an edge computing node; wherein the method comprises: Acquire device monitoring data corresponding to the power equipment through the terminal device node, pre-process the device monitoring data to obtain target monitoring data, and send the target monitoring data to the edge computing node; The edge computing node determines a data analysis result corresponding to the power equipment according to the target monitoring data, and generates a short-term scheduling decision corresponding to the power equipment according to a short-term decision model, the data analysis result and the target monitoring data; Among them, the short-term decision-making model is a lightweight neural network model trained based on historical monitoring data and data analysis results.

Images