Power grid dispatching method and device based on cloud edge collaboration and computer equipment

Abstract

The invention relates to a power grid dispatching method and device based on cloud edge collaboration, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: collecting power grid operation data; dividing the power grid operation data to obtain real-time operation data and historical operation data; respectively constructing at least one edge processing model according to the real-time operation data and the historical operation data; at least one edge processing model is utilized to pre-process the real-time operation data and the historical operation data, and the pre-processing mode comprises duplicate removal processing and data cleaning; based on the preprocessed real-time operation data and historical operation data, controlling the cloud to predict the operation state of the power grid, generating a power grid dispatching instruction, and sending the power grid dispatching instruction to an edge processing model; and utilizing at least one edge processing model to respond and process the power grid dispatching instruction. By adopting the method, the real-time performance and efficiency of power grid dispatching can be improved.

Description

Technical Field

[0001] The present application relates to the field of cloud-edge collaboration technology, and particularly to a power grid scheduling method, device, computer device, computer-readable storage medium, and computer program product based on cloud-edge collaboration. Background Art

[0002] As an important development direction of the power industry, the smart grid is facing the challenge of massive data processing. Cloud-edge collaboration technology provides a new solution for smart grid scheduling by combining the advantages of cloud computing and edge computing. This technology can reduce latency, optimize bandwidth utilization, and improve data privacy and security. In power grid intelligent scheduling, cloud-edge collaboration technology constructs a two-level fusion platform of "cloud + edge", realizing comprehensive monitoring, analysis and prediction, and collaborative control of the power system.

[0003] However, traditional cloud-edge collaboration technology has problems in processing power grid operation data. The main problem is that the unified and centralized processing of data causes real-time type data not to be fed back in a timely manner, affecting the rapid response to power grid faults. In addition, not all clouds have powerful computing capabilities, which limits the solution of problems.

[0004] Therefore, there is an urgent need for a power grid scheduling method, device, computer device, computer-readable storage medium, and computer program product based on cloud-edge collaboration, which can improve the real-time performance and efficiency of power grid scheduling. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a power grid scheduling method, device, computer device, computer-readable storage medium, and computer program product based on cloud-edge collaboration, which can improve the real-time performance and efficiency of power grid scheduling.

[0006] In a first aspect, the present application provides a power grid scheduling method based on cloud-edge collaboration, including:

[0007] Collect power grid operation data;

[0008] Perform partitioning processing on the power grid operation data to obtain real-time operation data and historical operation data;

[0009] Construct at least one edge processing model according to the real-time operation data and the historical operation data;

[0010] Use at least one edge processing model to preprocess the real-time operation data and the historical operation data, where the preprocessing methods include duplicate removal processing and data cleaning;

[0011] Based on the preprocessed real-time operation data and historical operation data, control the cloud to predict the operation state of the power grid, generate power grid dispatching instructions, and send the power grid dispatching instructions to the edge processing model;

[0012] Use at least one edge processing model to respond to and process the power grid dispatching instructions.

[0013] In one embodiment, the partitioning process of the power grid operation data to obtain real-time operation data and historical operation data includes:

[0014] According to the data complexity level, initially partition the power grid operation data to obtain operation data of the first complexity level and operation data of the second complexity level;

[0015] Partition the operation data of the first complexity level to obtain first real-time operation data and first historical operation data, and partition the operation data of the second complexity level to obtain second real-time operation data and second historical operation data.

[0016] In one embodiment, the constructing of at least one edge processing model according to the real-time operation data and historical operation data includes:

[0017] Construct a real-time edge processing model according to the first real-time operation data and the second real-time operation data;

[0018] According to the data importance level, partition the first historical operation data to obtain first operation data of the first difficulty level and second operation data of the first difficulty level, and partition the second historical operation data to obtain first operation data of the second difficulty level and second operation data of the second difficulty level;

[0019] Construct corresponding first, second, third, and fourth edge processing models according to the first operation data of the first difficulty level, the second operation data of the first difficulty level, the first operation data of the second difficulty level, and the second operation data of the second difficulty level.

[0020] In one embodiment, the partitioning of the first historical operation data according to the data importance level to obtain first operation data of the first difficulty level and second operation data of the first difficulty level, and the partitioning of the second historical operation data to obtain first operation data of the second difficulty level and second operation data of the second difficulty level includes:

[0021] Obtain the importance level weight value corresponding to each item of data in the first historical operation data and the second historical operation data;

[0022] Based on the importance level weight value corresponding to each item of data in the first historical operation data, the data is divided according to the first preset threshold to obtain the first operation data of the first difficulty and the second operation data of the first difficulty;

[0023] Based on the importance level weight value corresponding to each item of data in the second historical operation data, the data is divided according to the second preset threshold to obtain the first operation data of the second difficulty and the second operation data of the second difficulty.

[0024] In one embodiment, after initially dividing the power grid operation data to obtain the operation data of the first complexity level and the operation data of the second complexity level, it further includes:

[0025] According to the data importance level, each item of data in the operation data of the first complexity level is sorted to obtain a first sorting result, and the first sorting result is in turn: equipment status information, environmental parameters, infrastructure information, protection setting values, planned data, communication status information, and statistical data; among them, the serial number levels of the equipment status information and the environmental parameters are the same, the serial number levels of the infrastructure information and the protection setting values are the same, and the serial number levels of the communication status information and the statistical data are the same.

[0026] In one embodiment, after initially dividing the power grid operation data to obtain the operation data of the first complexity level and the operation data of the second complexity level, it further includes:

[0027] According to the data importance level, the weight values of each item of data in the operation data of the second complexity level are sorted to obtain a second sorting result, and the second sorting result is in turn: distributed energy resource data, power grid operation logs, power grid simulation data, user electricity consumption behavior data, power market data, network security data, advanced metering infrastructure data, and geographic information data; among them, the serial number levels of the power grid simulation data and the user electricity consumption behavior data are the same, the serial number levels of the power market data and the network security data are the same, and the serial number levels of the advanced metering infrastructure data and the geographic information data are the same.

[0028] In a second aspect, the present application further provides a power grid dispatching device based on cloud-edge collaboration, including:

[0029] A data acquisition module for acquiring power grid operation data;

[0030] A data division module for dividing and processing the power grid operation data to obtain real-time operation data and historical operation data;

[0031] A model construction module for respectively constructing at least one edge processing model according to the real-time operation data and the historical operation data;

[0032] A preprocessing module, configured to preprocess real-time operation data and historical operation data by using at least one edge processing model, wherein the preprocessing methods include duplicate removal processing and data cleaning;

[0033] A cloud computing and big data analysis module, further configured to control the cloud to predict the grid operation status based on the preprocessed real-time operation data and historical operation data, generate a grid scheduling instruction, and send the grid scheduling instruction to the edge processing model;

[0034] An edge computing and response module, further configured to use at least one edge processing model to respond to and process the grid scheduling instruction.

[0035] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0036] Collect grid operation data;

[0037] Perform partitioning processing on the grid operation data to obtain real-time operation data and historical operation data;

[0038] Respectively construct at least one edge processing model according to the real-time operation data and the historical operation data;

[0039] Use at least one edge processing model to preprocess the real-time operation data and the historical operation data, wherein the preprocessing methods include duplicate removal processing and data cleaning;

[0040] Based on the preprocessed real-time operation data and historical operation data, control the cloud to predict the grid operation status, generate a grid scheduling instruction, and send the grid scheduling instruction to the edge processing model;

[0041] Use at least one edge processing model to respond to and process the grid scheduling instruction.

[0042] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0043] Collect grid operation data;

[0044] Perform partitioning processing on the grid operation data to obtain real-time operation data and historical operation data;

[0045] Respectively construct at least one edge processing model according to the real-time operation data and the historical operation data;

[0046] Use at least one edge processing model to preprocess the real-time operation data and the historical operation data, wherein the preprocessing methods include duplicate removal processing and data cleaning;

[0047] Based on the pre - processed real - time operation data and historical operation data, control the cloud to predict the power grid operation state, generate power grid dispatching instructions, and send the power grid dispatching instructions to the edge processing model;

[0048] Use at least one edge processing model to respond to and process the power grid dispatching instructions.

[0049] In a fifth aspect, the present application also provides a computer program product, including a computer program, which when executed by a processor, implements the following steps:

[0050] Collect power grid operation data;

[0051] Perform partitioning processing on the power grid operation data to obtain real - time operation data and historical operation data;

[0052] Construct at least one edge processing model respectively according to the real - time operation data and historical operation data;

[0053] Use at least one edge processing model to pre - process the real - time operation data and historical operation data, where the pre - processing methods include duplicate removal processing and data cleaning;

[0054] Based on the pre - processed real - time operation data and historical operation data, control the cloud to predict the power grid operation state, generate power grid dispatching instructions, and send the power grid dispatching instructions to the edge processing model;

[0055] Use at least one edge processing model to respond to and process the power grid dispatching instructions.

[0056] The above grid dispatching method, device, computer equipment, computer-readable storage medium, and computer program product based on cloud-edge collaboration divide data processing into two stages: edge preprocessing and cloud deep processing, making data processing more efficient. According to the real-time nature and complexity of the data, computing resources are reasonably allocated between the edge and the cloud. The edge processes tasks with strong real-time requirements and relatively small computational loads, while the cloud processes complex tasks that require a large amount of computing resources and historical operation data support, such as long-term load forecasting and optimized dispatching strategies, achieving reasonable allocation and utilization of computing resources and improving the overall processing efficiency. By building a processing model at the edge to preprocess and respond to real-time operation data, rapid feedback on real-time operation data can be achieved. The edge computing device is deployed close to the data source, reducing the latency of data transmission to the cloud, enabling events with high real-time requirements such as grid faults to be responded to and processed immediately, and improving the real-time nature of grid dispatching. The edge processing model can monitor real-time operation data in real time, promptly detect abnormal situations, and perform real-time control according to preset rules or models. The edge processing model works in collaboration with the cloud and can flexibly adjust the dispatching strategy based on the real-time operation status and historical operation data of the power grid. For example, during peak load periods, the edge can quickly respond to the grid's dispatching instructions, start energy storage devices or distributed power sources, increase power supply capacity, and relieve the power supply pressure on the grid; during off-peak load periods, the cloud can optimize the dispatching strategy and perform charging of energy storage devices or equipment maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] To more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0058] Figure 1 FIG. is an application environment diagram of the grid dispatching method based on cloud-edge collaboration in an embodiment;

[0059] Figure 2 FIG. is a flowchart of the grid dispatching method based on cloud-edge collaboration in an embodiment;

[0060] Figure 3 FIG. is a flowchart of the grid dispatching method based on cloud-edge collaboration in another embodiment;

[0061] Figure 4 FIG. is a flowchart block diagram of the grid dispatching method based on cloud-edge collaboration in an embodiment;

[0062] Figure 5The structural block diagram of a power grid dispatching device based on cloud-edge collaboration in an embodiment;

[0063] Figure 6 The internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0064] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] With the transformation of the energy structure and the deepening of the power market, smart grids have become an important development direction in the power industry. Smart grids integrate advanced communication, computing, and control technologies to achieve intelligent, automated, and efficient operation of the power grid. However, the complexity of smart grids also brings challenges, especially when dealing with massive, multi-source heterogeneous data, traditional dispatching methods are difficult to meet the requirements.

[0066] As an emerging computing model, cloud-edge collaboration technology combines the powerful data processing capabilities of cloud computing and the low-latency, high-reliability advantages of edge computing, providing a new solution for smart grid dispatching. By pushing data processing to edge devices, cloud-edge collaboration technology reduces the time for data to be transmitted to remote data centers, reduces latency, and is suitable for applications with high real-time requirements. At the same time, cloud-edge collaboration can perform preliminary data processing on edge devices and only transmit the required summary data to the cloud, reducing a large amount of data transmission and optimizing bandwidth utilization. In addition, cloud-edge collaboration can also improve data privacy and security, because some applications involving sensitive data can be processed locally without transmitting sensitive information to the cloud.

[0067] In power grid intelligent dispatching, the system and method based on cloud-edge collaboration realizes the dispatching operation support for various services such as comprehensive monitoring, analysis and prediction, collaborative control, and market trading of the source-network-load-storage of the new power system by building an intelligent power grid dispatching operation platform with a two-level integration of "cloud + edge". The system integrates and divides functions, and adopts a microservices architecture to realize service-oriented business, which is convenient for result reuse and on-demand expansion. The edge-side system adopts an intensive concept and uses a lightweight hyper-converged intelligent hardware platform, which integrates many functions such as network switching, computing, storage, IO, and industrial communication buses, forming an integrated device with low cost, high reliability, diverse functions, and flexible configuration.

[0068] Based on the above, although existing cloud-edge collaboration technologies can reduce the time for data transmission to a remote data center by pushing data processing to edge devices, thereby reducing latency, and at the same time, preliminary data processing can be performed on edge devices, only transmitting the required summary data to the cloud, reducing a large amount of data transmission and optimizing bandwidth utilization, there are still the following problems in the current cloud-edge collaboration technology during the actual application process: First, when processing power grid operation data, it usually collects data uniformly and centrally, and then hands it over to edge devices for preprocessing into a summary form. In this process, the importance of real-time type data is not given priority. As a result, when the preprocessed data is transmitted to the cloud and the cloud gives corresponding feedback, it is easier for some real-time type data not to be fed back in a timely manner. This is also one of the main reasons why when a node in the power grid fails, the staff in the operation center do not receive alarm information in a timely manner. Therefore, based on the above, this phenomenon can be attributed to the fact that unified and centralized data processing is prone to data confusion, and the order of data processing has not been properly resolved. Although only when the cloud has very powerful computing and other processing capabilities can the above problems be effectively improved or solved, considering regional and cost issues, not all clouds have powerful computing and other capabilities, which makes the above problems still urgently need to be solved.

[0069] The power grid dispatching method based on cloud-edge collaboration provided by the embodiments of the present application can be applied to an application environment as Figure 1 shown. Among them, the terminal 102 communicates with the server 104 through a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers.

[0070] The server 104 controls the terminal 102 to collect power grid operation data; the server 104 performs partitioning processing on the power grid operation data to obtain real-time operation data and historical operation data; at least one edge processing model is constructed respectively according to the real-time operation data and the historical operation data; the real-time operation data and the historical operation data are preprocessed by using at least one edge processing model; based on the preprocessed real-time operation data and historical operation data, the cloud is controlled to predict the power grid operation state, generate a power grid dispatching instruction, and send the power grid dispatching instruction to the edge processing model; at least one edge processing model is used to respond to and process the power grid dispatching instruction.

[0071] Among them, the terminal 102 can be but is not limited to various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0072] In an exemplary embodiment, as Figure 2 shown, a power grid dispatching method based on cloud-edge collaboration is provided. Taking the 104 server in Figure 1 as an example, the method includes the following steps S202 to step S212. Among them:

[0073] Step S202, collect power grid operation data.

[0074] Specifically, collecting power grid operation data means obtaining various data generated during the operation of the power grid in real time through sensors and monitoring devices deployed at each node of the power grid. These data include but are not limited to:

[0075] Infrastructure information: such as the location of substations, the model and capacity of transformers, the starting and ending points of transmission lines, etc. This information is relatively stable, changes less, and has a relatively simple structure.

[0076] Equipment status information: such as the opening and closing status of switches, the on and off status of circuit breakers, etc. These status information are usually represented by boolean values or simple enumeration values and are easy to process.

[0077] Power grid operation logs: record various events and status changes during the operation of the power grid, contain a large amount of information, and require complex algorithms for parsing and analysis.

[0078] Power grid simulation data: used to simulate the operation of the power grid under various conditions. These data usually contain multiple variables and complex logical relationships and require complex models for calculation.

[0079] User power consumption behavior data: including the power consumption, power consumption time, power consumption mode of users, etc. These data have a high degree of diversity and uncertainty and require complex machine learning algorithms for mining and analysis.

[0080] Real-time operation data: such as real-time readings of current, voltage, power, etc. of some simple measurement devices; real-time status of power grid equipment such as transformers, generators, circuit breakers; real-time reports of faults and abnormal events; real-time weather data such as wind speed, temperature, humidity, rainfall, etc.; real-time power consumption of users, etc.

[0081] Step S204, perform partitioning processing on the power grid operation data to obtain real-time operation data and historical operation data.

[0082] Specifically, real-time operation data refers to data generated at the current moment or very close to the current moment, which needs to be processed and responded to immediately or quickly. This type of data reflects the current operation status and situation of the power grid.

[0083] Real-time operation data has strong time sensitivity and is usually generated and updated at intervals of seconds, milliseconds or even shorter, such as real-time measurement data like current, voltage, power, etc., as well as real-time status information of equipment. Since real-time operation data reflects the immediate state of the power grid, it needs to be processed and analyzed quickly in order to make scheduling decisions and control responses in a timely manner to ensure the stable operation and safety of the power grid.

[0084] Real-time operation data is often used in scenarios such as real-time monitoring of the power grid, fault detection, emergency response, load forecasting, etc. For example, by real-time monitoring data such as current and voltage, abnormal situations in the power grid such as overload and short circuit can be detected in a timely manner, and measures can be quickly taken to deal with them.

[0085] Historical operation data refers to the data generated within a certain period in the past. After a period of accumulation and storage, it is used for analyzing and mining the long-term trends, laws and characteristics of the power grid operation. Historical operation data usually covers a relatively long time span, such as data at the hour level, day level, month level or even year level, such as past power consumption data, equipment operation logs, fault records, etc.

[0086] Historical operation data is often used in scenarios such as power grid operation analysis, load forecasting, equipment maintenance plan formulation, fault diagnosis and prevention, energy management, etc. For example, by analyzing historical power consumption data, future power demands can be predicted, providing a reference for the scheduling and resource allocation of the power grid.

[0087] Step S206, construct at least one edge processing model according to the real-time operation data and the historical operation data.

[0088] Specifically, construct at least one edge processing model according to the real-time operation data, and process the real-time operation data quickly and efficiently to meet the application scenarios with high real-time requirements such as real-time monitoring of the power grid, fault detection, and emergency response. The characteristics include:

[0089] Low latency: The model needs to be executed quickly on the edge device, reducing the time for data to be transmitted to the cloud, thereby reducing the overall processing latency.

[0090] Lightweight: Since the computing and storage resources of the edge device are relatively limited, the model needs to be simplified as much as possible to adapt to the resource constraints of the edge environment.

[0091] Real-time performance: It can capture and respond to anomalies and changes in the power grid operation in a timely manner. For example, by real-time analyzing data such as current and voltage, faults such as overload and short circuit can be quickly detected.

[0092] Application: For example, deploying an edge computing platform in a substation, and using intelligent algorithms such as deep learning and neural networks to perform real-time analysis and processing on data such as device status information and power grid operation logs collected in real time, so as to achieve substation area protection control.

[0093] Construct at least one edge processing model based on historical operation data, and conduct in-depth analysis and mining on the accumulated historical power grid operation data to discover long-term trends, patterns and potential problems in power grid operation, and provide support for power grid planning, optimization and decision-making. Features:

[0094] High precision: The model needs to be able to accurately analyze and mine complex patterns and correlation relationships in historical operation data. For example, load forecasting is performed on historical electricity consumption data through machine learning algorithms.

[0095] Scalability: As historical operation data continues to accumulate, the model needs to have good scalability and be able to adapt to the growth of data scale.

[0096] Data fusion: It may be necessary to fuse and integrate historical operation data from different data sources to obtain more comprehensive and accurate analysis results.

[0097] Application: For example, deploying an edge computing server inside a power plant, training a deep learning model on historical generator set operation data, load side data, etc., and establishing a performance model and a load model of the generator set for optimizing power generation scheduling and load management.

[0098] Step S208, using at least one edge processing model to preprocess real-time operation data and historical operation data, where the preprocessing methods include duplicate removal processing and data cleaning.

[0099] Specifically, during the process of real-time operation data collection, due to reasons such as sensor failures and abnormal network transmissions, duplicate data records may be generated. For example, current, voltage and other data at the same moment are sent repeatedly. The edge processing model can identify duplicate data records by comparing features such as data timestamps and data contents, and delete or merge them, only retaining one valid piece of data. Duplicate removal processing can reduce data redundancy, reduce the overhead of data storage and transmission, improve the efficiency and accuracy of subsequent data processing, and avoid incorrect analysis and decision-making caused by duplicate data.

[0100] During the storage and management of historical operation data, data duplication may occur due to reasons such as data import, migration, backup, etc. For example, power consumption data for the same time period may be stored repeatedly in different database tables. The edge processing model can use technologies such as hash algorithms and data fingerprints to deduplicate the historical operation data and ensure data uniqueness. Deduplication helps improve the quality and usability of historical operation data, providing accurate basic data support for subsequent data analysis, mining, and decision-making.

[0101] During the acquisition process of real-time operation data and historical operation data, various noises, interferences, and abnormal factors may be encountered, such as sensor failures, signal interferences, etc., resulting in incorrect, abnormal, or incomplete records in the data, such as abnormal fluctuations or missing values in data such as current and voltage.

[0102] The edge processing model can use technologies such as statistical analysis and anomaly detection algorithms to clean the real-time operation data and historical operation data, such as identifying and correcting outliers, filling in missing values, and smoothing noisy data. Data cleaning can improve the quality of real-time operation data, ensuring the accuracy and reliability of subsequent real-time monitoring, fault detection, and emergency response operations, and avoiding misjudgments and incorrect decisions caused by data problems.

[0103] Step S210: Based on the preprocessed real-time operation data and historical operation data, control the cloud to predict the grid operation status, generate grid dispatching instructions, and send the grid dispatching instructions to the edge processing model.

[0104] Specifically, the cloud integrates the preprocessed real-time operation data and historical operation data, and uses big data analysis technology and machine learning algorithms to comprehensively analyze and predict the operation status of the grid.

[0105] The prediction content includes, but is not limited to, grid load prediction, equipment operation status prediction, fault prediction, etc. For example, by analyzing historical load data and current real-time load data, the load change trend in the future period can be predicted; by analyzing the historical operation data and real-time status data of equipment, possible equipment failures can be predicted. Prediction methods can use machine learning algorithms such as time series analysis, regression analysis, neural networks, and support vector machines, combined with the operation model and empirical knowledge of the grid, to make accurate predictions.

[0106] According to the prediction results and the operation goals of the grid (such as stability, economy, reliability, etc.), the cloud generates corresponding grid dispatching instructions. For example, if it is predicted that the load in a certain area will increase significantly, the cloud may generate dispatching instructions to increase the output of generating units or adjust the grid operation mode.

[0107] The content of the dispatching instruction includes the start and stop of generating units, output adjustment, switch operation of grid equipment, load distribution and transfer, charge and discharge of energy storage equipment, etc. Optimization algorithms (such as linear programming, integer programming, multi-objective optimization, etc.) can be used to optimize the dispatching scheme to achieve the best dispatching effect and meet the operation requirements and constraints of the power grid. The cloud transmits the generated power grid dispatching instructions to the edge device where the edge processing model is located through the communication network with the edge device.

[0108] Step S212, use at least one edge processing model to respond to and process the power grid dispatching instruction.

[0109] Specifically, after receiving the dispatching instruction, the edge processing model precisely controls and responds to the grid equipment according to the local actual situation and control logic. For example, it controls the output adjustment of generating units, operates the switches of grid equipment, and regulates the charge and discharge of energy storage equipment. During the execution of the dispatching instruction, the edge processing model will monitor the operation status of the power grid and the feedback information of the equipment in real time and transmit this information back to the cloud. The cloud dynamically adjusts and optimizes the dispatching instruction according to the real-time feedback information to adapt to the actual changes and requirements of the power grid operation.

[0110] It should be noted that when the processing capacity of the cloud is strong enough, for some data, there is no need to go through the edge processing model until it is uploaded to the cloud, and the cloud completes the entire process including calculation.

[0111] In addition, for some simple power grid operation data processing tasks, they can be directly completed on the edge processing model without being transmitted to the cloud. This is related to the fact that conventional edge devices usually have a certain computing ability and can handle some basic data analysis and preprocessing tasks. The advantage of doing this is that it can reduce data transmission delay, improve real-time performance, and reduce the burden on cloud computing resources. For example, the second-level operation data with the second difficulty level divided by the application through a value lower than the first threshold is only processed through the second edge processing model for the entire process including calculation, which also includes generating instructions according to the calculated results and making corresponding responses according to the instructions.

[0112] Among them, it is worth noting that regarding the issue of the processing priority of real-time operation data in this application, the reason why real-time operation data can obtain the processing priority is mainly that the historical operation data after screening out the real-time operation data still needs to be further screened by means of weight setting and threshold comparison before it can enter the preprocessing step of the subsequent edge processing model. Since the number of steps is more than that of the real-time operation data at the step level, the real-time operation data has the processing priority.

[0113] In the above grid dispatching method based on cloud-edge collaboration, according to the real-time nature and complexity of data, computing resources are reasonably allocated between the edge side and the cloud side. The edge side processes tasks with strong real-time nature and relatively small computing volume, while the cloud side processes complex tasks that require a large amount of computing resources and historical operation data support, such as long-term load forecasting, optimizing dispatching strategies, etc., achieving reasonable allocation and utilization of computing resources and improving the overall processing efficiency. By building a processing model at the edge side to preprocess and respond to real-time operation data, rapid feedback on real-time operation data can be achieved. The edge computing device is deployed close to the data source, reducing the latency of data transmission to the cloud, enabling events with high real-time requirements such as grid faults to be responded to and processed in a timely manner, and improving the real-time nature of grid dispatching. The edge processing model can monitor real-time operation data in real time, promptly detect abnormal situations, and perform real-time control according to preset rules or models. The edge processing model works in collaboration with the cloud side and can flexibly adjust the dispatching strategy according to the real-time operation state and historical operation data of the power grid.

[0114] In an exemplary embodiment, the grid operation data is divided to obtain real-time operation data and historical operation data, including:

[0115] According to the data complexity level, the grid operation data is initially divided to obtain operation data of the first complexity level and operation data of the second complexity level;

[0116] The operation data of the first complexity level is divided to obtain first real-time operation data and first historical operation data, and the operation data of the second complexity level is divided to obtain second real-time operation data and second historical operation data.

[0117] Specifically, the data complexity level refers to the complexity degree in aspects such as the structure, content, and association relationship of the data. The high or low complexity level reflects the differences in the difficulty of data processing and analysis and the required resources. By initially dividing the grid operation data according to its complexity level, the data can be divided into relatively simple and relatively complex categories, facilitating subsequent adoption of different processing strategies and methods for data of different complexity levels, and improving the efficiency and accuracy of data processing.

[0118] Among them, in the operation data of the first complexity level, according to the time characteristics of the data, it is divided into first real-time operation data and first historical operation data. Among them, the first real-time operation data: This part of the data is the data generated in real time at the first complexity level and needs to be processed and responded to quickly to meet the needs of real-time monitoring, fault detection, etc. For example, the real-time status information of equipment, simple real-time measurement data, etc. The first historical operation data: This part of the data is the data generated in the past at the first complexity level. After a period of accumulation, it is used to analyze and mine the long-term trends and laws of power grid operation. For example, past equipment status records, simple measurement data records, etc.

[0119] In the operation data of the second complexity level, similarly according to the time characteristics of the data, it is divided into real-time operation data and historical operation data. Among them, the second real-time operation data: This part of the data is the data generated in real time at the second complexity level and has higher complexity and real-time requirements, and more advanced processing methods need to be used for analysis and processing. For example, real-time power grid operation logs, real-time user electricity consumption behavior data, etc. The second historical operation data: This part of the data is the data generated in the past at the second complexity level, which contains rich information and complex correlation relationships and is used to deeply analyze the long-term trends, potential problems and optimization opportunities of power grid operation. For example, historical power grid operation logs, user electricity consumption behavior data, etc.

[0120] Among them, the data structure of the operation data at the first complexity level is relatively simple, the content is relatively single, the correlation relationship is relatively clear, and the difficulty of processing and analysis is relatively low. For example, the real-time readings of current, voltage, power, etc. of some simple measurement devices; although these data change in real time, their structure and processing methods are relatively simple.

[0121] The data structure of the operation data at the second complexity level is complex, the content is rich and diverse, the correlation relationship is complex, and the difficulty of processing and analysis is high. For example, the real-time data in the power grid operation log: the real-time status of power grid equipment such as transformers, generators, and circuit breakers; real-time reports of faults and abnormal events; real-time weather data such as wind speed, temperature, humidity, rainfall, etc.; these data may have a direct impact on power grid operation; the real-time simulation data in the power grid simulation data: during the power grid simulation process, real-time simulation data under various conditions will be generated to evaluate the performance and stability of the power grid; these data can include the real-time simulation status of various equipment and lines, as well as the overall real-time simulation results of the power grid. The real-time data in the user electricity consumption behavior data: the real-time electricity consumption of users, usually in kilowatt-hours; the real-time electricity load of users, such as the power and energy consumption of each electrical equipment; real-time electricity price data.

[0122] In this embodiment, through such partitioning processing, the power grid operation data can be subdivided according to the complexity level and time characteristics, providing more targeted and efficient data support for subsequent data preprocessing, analysis, and scheduling, thereby improving the overall performance and effectiveness of the power grid intelligent scheduling system.

[0123] In an exemplary embodiment, as Figure 3 shown, at least one edge processing model is respectively constructed according to the real-time operation data and historical operation data, including:

[0124] Step S302, constructing a real-time edge processing model according to the first real-time operation data and the second real-time operation data;

[0125] Step S304, partitioning the first historical operation data according to the data importance level to obtain the first operation data of the first difficulty and the second operation data of the first difficulty, and partitioning the second historical operation data to obtain the first operation data of the second difficulty and the second operation data of the second difficulty;

[0126] Step S306, respectively constructing corresponding first, second, third, and fourth edge processing models according to the first operation data of the first difficulty, the second operation data of the first difficulty, the first operation data of the second difficulty, and the second operation data of the second difficulty.

[0127] Specifically, aiming at the high real-time and fast response requirements of real-time operation data, a dedicated edge processing model is constructed to achieve fast processing and analysis of real-time data. The characteristics of the real-time edge processing model include:

[0128] Low latency: The model needs to be executed quickly on edge devices, reducing the time for data to be transmitted to the cloud and ensuring fast processing and response of real-time data. Lightweight: Since the computing and storage resources of edge devices are relatively limited, the model needs to be simplified as much as possible to adapt to the resource constraints of the edge environment. Real-time: It can timely capture and respond to abnormalities and changes in the power grid operation, such as quickly detecting faults such as overload and short circuit by real-time analyzing data such as current and voltage.

[0129] Among them, the first historical operation data is divided into data of different difficulties according to the importance of the data. Including:

[0130] The first operation data of the first difficulty: This type of data is relatively important in the first historical operation data and may contain key historical information and long-term trends, requiring in-depth analysis and mining.

[0131] Second running data of the first difficulty: This type of data has a relatively low importance level in the first historical running data. It may contain some auxiliary information or secondary data, but still has certain analysis value.

[0132] Among them, according to the importance level of the data, the second historical running data is divided into data of different difficulties.

[0133] First running data of the second difficulty: This type of data is relatively important in the second historical running data. It may contain complex correlation relationships and potential problems, and more advanced analysis methods need to be used for processing.

[0134] Second running data of the second difficulty: This type of data has a relatively low importance level in the second historical running data. It may contain some redundant information or incomplete data, but still has certain reference value.

[0135] Such as Figure 4 shown, the model construction is based on data of different difficulties, including:

[0136] First edge processing model: Constructed for the first running data of the first difficulty, the model needs to have strong analysis and mining capabilities, and be able to accurately identify and extract key historical information and long-term trends.

[0137] Second edge processing model: Constructed for the second running data of the first difficulty, the model can be relatively simplified, but still needs to be able to effectively process and analyze auxiliary information or secondary data.

[0138] Third edge processing model: Constructed for the first running data of the second difficulty, the model needs to be able to handle complex correlation relationships and potential problems, and adopt more advanced analysis algorithms and models.

[0139] Fourth edge processing model: Constructed for the second running data of the second difficulty, the model can be further simplified, mainly processing redundant information or incomplete data, and extracting valuable reference information.

[0140] In this embodiment, through such a model construction process, different processing strategies can be adopted for different types and difficulties of data, giving full play to the advantages of edge computing, improving the efficiency and accuracy of data processing, and providing strong support for the intelligent management and optimized operation of the power grid.

[0141] In an exemplary embodiment, according to the data importance level, the first historical running data is divided to obtain the first running data of the first difficulty and the second running data of the first difficulty, and the second historical running data is divided to obtain the first running data of the second difficulty and the second running data of the second difficulty, including:

[0142] Obtain the importance level weight values corresponding to each piece of data in the first historical operation data and the second historical operation data;

[0143] Based on the importance level weight values corresponding to each piece of data in the first historical operation data, perform data division according to the first preset threshold to obtain the first operation data of the first difficulty and the second operation data of the first difficulty;

[0144] Based on the importance level weight values corresponding to each piece of data in the second historical operation data, perform data division according to the second preset threshold to obtain the first operation data of the second difficulty and the second operation data of the second difficulty.

[0145] Specifically, the importance level weight value is a quantitative indicator for measuring the importance of data, usually determined according to factors such as the impact degree on power grid operation, analysis value, and decision support of the data. The higher the weight value, the more important the data.

[0146] Perform feature analysis on each piece of data in the first historical operation data and the second historical operation data to identify the key attributes and influencing factors of the data, such as the type, source, time span, correlation relationship, etc. of the data.

[0147] According to the data characteristics and the actual requirements of power grid operation, assign an important level weight value to each piece of data. For example, equipment fault records may be given a relatively high weight value because they have an important impact on the stability and security of the power grid; while some conventional equipment status records may be given a relatively low weight value.

[0148] Exemplarily, according to the above-mentioned sorting results, assign weight values to the first historical operation data and the second historical operation data respectively; among them:

[0149] Assigning weight values to the first historical operation data includes:

[0150] Assign a weight value of 0.2 to the equipment status information; assign a weight value of 0.2 to the environmental parameters; assign a weight value of 0.16 to the infrastructure information; assign a weight value of 0.16 to the protection setting value; assign a weight value of 0.13 to the planned data; assign a weight value of 0.09 to the communication status information; assign a weight value of 0.06 to the simple statistical data;

[0151] Assigning weight values to the second historical operation data includes:

[0152] The weight value assigned to distributed energy resource data is 0.25; the weight value assigned to grid operation logs is 0.2; the weight value assigned to grid simulation data is 0.125; the weight value assigned to user electricity consumption behavior data is 0.125; the weight value assigned to electricity market data is 0.09; the weight value assigned to network security data is 0.09; the weight value assigned to advanced metering infrastructure data is 0.06; the weight value assigned to geographic information data is 0.06.

[0153] Exemplarily, a first preset threshold is set for the weight value assigned to the first historical operation data in the above example to be 0.15; similarly, a second preset threshold is set for the weight value assigned to the second historical operation data in the above example to be 0.1.

[0154] Based on the importance level weight value, data is partitioned, where:

[0155] Partitioning of the first historical operation data: Set a first preset threshold, compare the importance level weight value of each item of data in the first historical operation data with the first preset threshold, and partition the data according to the comparison result. Data with a weight value greater than or equal to the first preset threshold is partitioned into first operation data of the first difficulty level. This type of data has relatively high importance, may contain key historical information and long-term trends, and requires in-depth analysis and mining. Data with a weight value less than the first preset threshold is partitioned into second operation data of the first difficulty level. The importance of this type of data is relatively low, may contain some auxiliary information or secondary data, but still has certain analysis value.

[0156] Partitioning of the second historical operation data: Set a second preset threshold, which may be different from the first preset threshold. Compare the importance level weight value of each item of data in the second historical operation data with the second preset threshold, and partition the data according to the comparison result. Data with a weight value greater than or equal to the second preset threshold is partitioned into first operation data of the second difficulty level. This type of data is relatively important in the second historical operation data, may contain complex correlation relationships and potential problems, and requires more advanced analysis methods for processing. Data with a weight value less than the second preset threshold is partitioned into second operation data of the second difficulty level. The importance of this type of data is relatively low, may contain some redundant information or incomplete data, but still has certain reference value.

[0157] In this embodiment, in this way, the historical operation data can be reasonably partitioned according to the importance level weight value of the data and the preset threshold, classifying the data according to importance and analysis difficulty, providing a more targeted basis for the subsequent construction of the edge processing model and data processing.

[0158] In an exemplary embodiment, after initially dividing the power grid operation data to obtain the operation data at the first complexity level and the operation data at the second complexity level, the following steps are further included:

[0159] According to the data importance level, each piece of data in the operation data at the first complexity level is sorted to obtain a first sorting result. The first sorting result is, in sequence: equipment status information, environmental parameters, infrastructure information, protection settings, planned data, communication status information, and statistical data. Among them, the serial number levels of the equipment status information and the environmental parameters are the same, the serial number levels of the infrastructure information and the protection settings are the same, and the serial number levels of the communication status information and the statistical data are the same.

[0160] Specifically, the sorting is performed according to the importance and urgency of the data for the power grid operation. The more important the data, the earlier it is ranked and given priority for processing and analysis to ensure that the power grid can respond promptly to key events and changes.

[0161] Equipment status information and environmental parameters: Their serial number levels are the same and they are ranked at the top, indicating that these two types of data have the greatest impact on the power grid operation and need to be monitored in real time and given priority for processing. For example, the equipment status information can promptly detect abnormal situations of the equipment, and the environmental parameters can warn of environmental changes that may affect the power grid equipment.

[0162] Infrastructure information and protection settings: Their serial number levels are the same and they are ranked in the middle position, indicating that these two types of data have a certain impact on the long-term stable operation and safety of the power grid, but the processing priority is relatively low. The infrastructure information helps to understand the overall structure and layout of the power grid, and the protection settings are an important guarantee for the safe operation of the power grid.

[0163] Communication status information and statistical data: Their serial number levels are the same and they are ranked at the end, indicating that these two types of data have a relatively small impact on the power grid operation and the lowest processing priority. The communication status information is mainly used for communication coordination between devices, and the statistical data is mainly used for historical analysis and decision support.

[0164] In this embodiment, through the complexity level division and importance level sorting in the above manner, the data processing resources can be reasonably allocated to ensure that the key data is processed in a timely manner, improving the efficiency and safety of the power grid operation.

[0165] In an exemplary embodiment, after initially dividing the power grid operation data to obtain the operation data at the first complexity level and the operation data at the second complexity level, the following steps are further included:

[0166] According to the data importance level, each piece of data in the operation data of the second complexity level is sorted by weight value to obtain the second sorting result. The second sorting result is as follows: distributed energy resource data, power grid operation log, power grid simulation data, user electricity consumption behavior data, power market data, network security data, advanced metering infrastructure data, and geographic information data. Among them, the serial number levels of power grid simulation data and user electricity consumption behavior data are the same, the serial number levels of power market data and network security data are the same, and the serial number levels of advanced metering infrastructure data and geographic information data are the same.

[0167] Specifically, the weight value sorting mainly depends on the importance of the data to the power grid operation and the urgency of processing. The more important the data, the higher the weight value, and it is ranked in the front and given priority for processing and analysis to ensure that the power grid can respond to key events and changes in a timely manner. Different types of power grid operation data play different roles in power grid dispatching and management, and their importance and application scenarios also vary. For example, distributed energy resource data directly affects the energy supply and dispatching balance of the power grid, so the weight value is relatively high; while geographic information data, although it has a certain role in power grid planning and asset management, its processing priority is relatively low.

[0168] Among them, distributed energy resource data: has the highest weight value and is ranked first. This is because distributed energy resources (such as solar energy, wind energy, etc.) are intermittent and uncertain, and their data is crucial for the real-time dispatching and optimization of the power grid. Accurately grasping the operation status and power generation prediction data of distributed energy resources helps the power grid reasonably arrange the power generation plan, balance supply and demand, and improve energy utilization efficiency.

[0169] Power grid operation log: is ranked second. The power grid operation log records various events and operations during the power grid operation process and is an important basis for analyzing the power grid operation status and troubleshooting the cause of faults. By analyzing the log data, potential problems can be discovered in a timely manner, and the power grid operation strategy can be optimized.

[0170] Power grid simulation data and user electricity consumption behavior data: have the same weight value and are ranked third. Power grid simulation data is used to simulate the operation status and various scenarios of the power grid to help power grid dispatching personnel make decision support and risk assessment; user electricity consumption behavior data reflects the electricity consumption habits and demand changes of users and is of great significance for power grid load forecasting and demand response management.

[0171] Power market data and network security data: have the same weight value and are ranked fourth. Power market data involves information such as power transactions and price fluctuations, which has a certain impact on the economic dispatching and market strategy formulation of the power grid; network security data is related to the information security and protection measures of the power grid to ensure the stable operation of the power grid dispatching system.

[0172] Advanced metering infrastructure data and geographic information data: They have the same weight value and are ranked last. Advanced metering infrastructure data mainly comes from devices such as smart meters, providing detailed information on user electricity consumption, which helps the power grid conduct refined management; geographic information data provides basic support for power grid planning, equipment positioning, and asset management.

[0173] In this embodiment, through the weight value sorting in the above manner, data processing resources can be reasonably allocated, ensuring that key data is processed preferentially, improving the efficiency and accuracy of power grid dispatching, and providing strong data support for the stable operation and optimized management of the power grid.

[0174] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are displayed in sequence according to the arrows, these steps do not necessarily need to be executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily need to be executed at the same time, but can be executed at different times. The execution order of these steps or stages does not necessarily need to be sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0175] Based on the same inventive concept, the embodiments of the present application also provide a cloud-edge collaborative power grid dispatching device for implementing the above-mentioned cloud-edge collaborative power grid dispatching method. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more of the following embodiments of the cloud-edge collaborative power grid dispatching device can refer to the limitations on the cloud-edge collaborative power grid dispatching method in the above text, and will not be repeated here.

[0176] In an exemplary embodiment, as Figure 5 shown, a cloud-edge collaborative power grid dispatching device is provided, including:

[0177] A data acquisition module 502, configured to acquire power grid operation data;

[0178] A data partitioning module 504, configured to perform partitioning processing on the power grid operation data to obtain real-time operation data and historical operation data;

[0179] A model construction module 506, configured to respectively construct at least one edge processing model according to the real-time operation data and the historical operation data;

[0180] A preprocessing module 508, configured to preprocess real-time operation data and historical operation data by using at least one edge processing model, wherein the preprocessing methods include duplicate removal processing and data cleaning;

[0181] The cloud computing and big data analysis module 510 is further configured to control the cloud to predict the power grid operation status, generate a power grid scheduling instruction, and send the power grid scheduling instruction to the edge processing model based on the preprocessed real-time operation data and historical operation data;

[0182] The edge computing and response module 512 is further configured to use at least one edge processing model to respond to and process the power grid scheduling instruction.

[0183] In an exemplary embodiment, the data partitioning module 504 is further configured to initially partition the power grid operation data according to the data complexity level to obtain operation data of the first complexity level and operation data of the second complexity level; partition the operation data of the first complexity level to obtain first real-time operation data and first historical operation data, and partition the operation data of the second complexity level to obtain second real-time operation data and second historical operation data.

[0184] In an exemplary embodiment, the model construction module 506 is further configured to construct a real-time edge processing model according to the first real-time operation data and the second real-time operation data; partition the first historical operation data according to the data importance level to obtain first operation data of the first difficulty level and second operation data of the first difficulty level, and partition the second historical operation data to obtain first operation data of the second difficulty level and second operation data of the second difficulty level; construct corresponding first, second, third, and fourth edge processing models according to the first operation data of the first difficulty level, the second operation data of the first difficulty level, the first operation data of the second difficulty level, and the second operation data of the second difficulty level, respectively.

[0185] In an exemplary embodiment, the data partitioning module 504 is further configured to partition the first historical operation data according to the data importance level to obtain the first operation data of the first difficulty level and the second operation data of the first difficulty level, and partition the second historical operation data to obtain the first operation data of the second difficulty level and the second operation data of the second difficulty level, including: obtaining the importance level weight value corresponding to each piece of data in the first historical operation data and the second historical operation data; partitioning the data according to a first preset threshold based on the importance level weight value corresponding to each piece of data in the first historical operation data to obtain the first operation data of the first difficulty level and the second operation data of the first difficulty level; and partitioning the data according to a second preset threshold based on the importance level weight value corresponding to each piece of data in the second historical operation data to obtain the first operation data of the second difficulty level and the second operation data of the second difficulty level.

[0186] In an exemplary embodiment, the data partitioning module 504 is further configured to sort each piece of data in the operation data of the first complexity level according to the data importance level to obtain a first sorting result, and the first sorting result is successively: device status information, environmental parameters, infrastructure information, protection setting values, planned data, communication status information, and statistical data; wherein, the serial number levels of the device status information and the environmental parameters are the same, the serial number levels of the infrastructure information and the protection setting values are the same, and the serial number levels of the communication status information and the statistical data are the same.

[0187] In an exemplary embodiment, the data partitioning module 504 is further configured to sort the weight values of each piece of data in the operation data of the second complexity level according to the data importance level to obtain a second sorting result, and the second sorting result is successively: distributed energy resource data, power grid operation logs, power grid simulation data, user electricity consumption behavior data, power market data, network security data, advanced metering infrastructure data, and geographic information data; wherein, the serial number levels of the power grid simulation data and the user electricity consumption behavior data are the same, the serial number levels of the power market data and the network security data are the same, and the serial number levels of the advanced metering infrastructure data and the geographic information data are the same.

[0188] Each module in the above grid dispatching device based on cloud-edge collaboration can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in hardware form or independent of the processor, or stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to the above respective modules.

[0189] In an exemplary embodiment, a computer device is provided, and the computer device may be a server, and its internal structure diagram may be as Figure 6As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store power grid operation data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a power grid scheduling method based on cloud-edge collaboration.

[0190] Those skilled in the art can understand that Figure 6 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0191] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the above method are implemented.

[0192] In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0193] In an embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0194] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0195] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. 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. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0196] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, 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, it should be considered as the scope recorded in this application.

[0197] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A power grid dispatching method based on cloud-edge collaboration, characterized in that: Methods include: Collect power grid operation data; Divide and process the power grid operation data to obtain real-time operation data and historical operation data; Building at least one edge processing model according to the real-time operation data and the historical operation data; Preprocessing the real-time operation data and the historical operation data using at least one edge processing model, wherein the preprocessing method includes deduplication processing and data cleaning; Based on the pre-processed real-time operation data and historical operation data, the cloud is controlled to predict the operation status of the power grid, generate power grid dispatching instructions, and send the power grid dispatching instructions to the edge processing model; At least one edge processing model is used to respond to and process the power grid dispatch instruction.

2. The power grid dispatching method based on cloud-edge collaboration according to claim 1 is characterized in that: The power grid operation data is divided and processed to obtain real-time operation data and historical operation data, including: According to the data complexity level, the power grid operation data is initially divided to obtain operation data of the first complexity level and operation data of the second complexity level; The operation data of the first complexity level is divided into data to obtain first real-time operation data and first historical operation data, and the operation data of the second complexity level is divided into data to obtain second real-time operation data and second historical operation data.

3. The power grid dispatching method based on cloud-edge collaboration according to claim 2 is characterized in that: The step of constructing at least one edge processing model according to the real-time operation data and the historical operation data includes: Building a real-time edge processing model based on the first real-time operation data and the second real-time operation data; According to the data importance level, the first historical operation data is divided into data to obtain first operation data of a first difficulty and second operation data of a first difficulty, and the second historical operation data is divided into data to obtain first operation data of a second difficulty and second operation data of a second difficulty; According to the first operating data of the first difficulty, the second operating data of the first difficulty, the first operating data of the second difficulty, and the second operating data of the second difficulty, the corresponding first edge processing model, second edge processing model, third edge processing model and fourth edge processing model are constructed respectively.

4. The power grid dispatching method based on cloud-edge collaboration according to claim 3 is characterized in that: The step of dividing the first historical operation data according to the data importance level to obtain first operation data of a first difficulty and second operation data of a first difficulty, and dividing the second historical operation data to obtain first operation data of a second difficulty and second operation data of a second difficulty includes: Obtaining an importance level weight value corresponding to each item of data in the first historical operation data and the second historical operation data; Based on the importance level weight value corresponding to each data item in the first historical operation data, the data is divided according to the first preset threshold value to obtain first operation data of a first difficulty and second operation data of a first difficulty; Based on the importance level weight value corresponding to each data item in the second historical operation data, the data is divided according to the second preset threshold to obtain first operation data of a second difficulty and second operation data of a second difficulty.

5. The power grid dispatching method based on cloud-edge collaboration according to claim 2 is characterized in that: After the power grid operation data is initially divided to obtain operation data of the first complexity level and operation data of the second complexity level, the method further includes: According to the data importance level, each data in the operating data of the first complexity level is sorted to obtain a first sorting result, which is: equipment status information, environmental parameters, infrastructure information, protection settings, planning data, communication status information and statistical data; among them, the serial number level of the equipment status information and the environmental parameters is the same, the serial number level of the infrastructure information and the protection settings is the same, and the serial number level of the communication status information and the statistical data is the same.

6. The power grid dispatching method based on cloud-edge collaboration according to claim 2 is characterized in that: After the power grid operation data is initially divided to obtain operation data of a first complexity level and operation data of a second complexity level, the method further includes: According to the data importance level, each data in the second complexity level operation data is sorted by weight value to obtain a second sorting result, and the second sorting results are: distributed energy resource data, power grid operation log, power grid simulation data, user electricity consumption behavior data, electricity market data, network security data, advanced measurement system data and geographic information data; among them, the grid simulation data and the user electricity consumption behavior data have the same serial number level, the electricity market data and the network security data have the same serial number level, and the advanced measurement system data and the geographic information data have the same serial number level.

7. A power grid dispatching device based on cloud-edge collaboration, characterized in that: The device comprises: Data acquisition module, used to collect power grid operation data; The data partitioning module is used to partition and process the power grid operation data to obtain real-time operation data and historical operation data; A model building module, used to build at least one edge processing model according to the real-time operation data and the historical operation data; A preprocessing module, used to preprocess the real-time operation data and the historical operation data using at least one edge processing model, wherein the preprocessing method includes deduplication processing and data cleaning; The cloud computing and big data analysis module is also used to control the cloud to predict the power grid operation status based on the pre-processed real-time operation data and historical operation data, generate power grid dispatching instructions, and send the power grid dispatching instructions to the edge processing model; The edge computing and response module is also used to respond to and process the power grid dispatching instruction using at least one edge processing model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images