Power grid heterogeneous flexible resource management scheduling method based on cloud side-end cooperation

Abstract

The invention discloses a power grid heterogeneous flexible resource management scheduling method based on cloud side-end collaboration, and relates to the technical field of resource management, and the method comprises the steps: building a power grid heterogeneous resource node network; obtaining a plurality of power grid resource node monitoring blocks; determining and identifying a power grid resource node through abnormal prediction perception; performing association auxiliary analysis on the identified power grid resource nodes according to the power grid heterogeneous resource node network to generate an auxiliary power grid resource node network; and performing auxiliary resource scheduling decision on the identified power grid resource nodes, generating a resource auxiliary scheduling scheme, and performing resource auxiliary management on the identified power grid resource nodes. The technical problems that in the prior art, when power grid heterogeneous resource management scheduling is carried out, resource management scheduling for abnormal power grid resource nodes is low in timeliness and poor in accuracy are solved, and the technical effects that the resource management scheduling timeliness and accuracy of the abnormal power grid resource nodes are improved, and the power grid heterogeneous resource management scheduling quality is improved are achieved.

Description

Technical Field

[0001] The present invention relates to the field of resource management technology, and in particular to a method for managing and scheduling heterogeneous flexible resources in a power grid based on cloud-edge-end collaboration. Background Art

[0002] With the development of modern power systems and the popularization of smart grid technologies, the types and quantities of heterogeneous flexible resources in power grids are increasing. These resources include renewable energy generation devices, energy storage equipment, distributed power sources, load response resources, etc. The access of these heterogeneous resources has brought new challenges to the dispatch and management of power grids. Traditional grid resource dispatch methods are usually based on centralized decision-making systems and rely on preset dispatch rules. However, this method is unable to cope with the complexity and dynamics of the current power grid environment. Summary of the invention

[0003] The present application provides a method for heterogeneous and flexible resource management and scheduling of power grids based on cloud-edge-end collaboration, which is used to solve the technical problems of low timeliness and poor accuracy in resource management and scheduling for abnormal power grid resource nodes in the prior art when performing heterogeneous resource management and scheduling of power grids.

[0004] In view of the above problems, the present application provides a method for managing and scheduling heterogeneous flexible resources in power grids based on cloud-edge-end collaboration.

[0005] The present application provides a method for managing and scheduling heterogeneous flexible resources in a power grid based on cloud-edge-end collaboration, the method comprising:

[0006] According to the cloud-edge-end collaborative platform, multiple power grid heterogeneous resource nodes are feature-driven integrated to establish a power grid heterogeneous resource node network; according to the power grid heterogeneous resource node network, the cloud-edge-end collaborative platform is interacted to obtain multiple power grid resource node monitoring blocks; based on the node abnormality constraint conditions, the power grid heterogeneous resource node network is abnormally predicted and perceived according to the multiple power grid resource node monitoring blocks to determine the identified power grid resource node; according to the power grid heterogeneous resource node network, the identified power grid resource node is associated with auxiliary analysis to generate an auxiliary power grid resource node network, wherein the auxiliary power grid resource node network includes a task auxiliary power grid resource node subnetwork and an operation and maintenance auxiliary power grid resource node subnetwork; based on the resource-assisted scheduling adaptation constraint rules, the auxiliary power grid resource node network is combined with the auxiliary power grid resource node network to make auxiliary resource scheduling decisions for the identified power grid resource nodes to generate a resource-assisted scheduling plan; based on the cloud-edge-end collaborative platform, the identified power grid resource nodes are resource-assisted managed according to the auxiliary power grid resource node network and the resource-assisted scheduling plan.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] This application performs feature-driven integration of multiple power grid heterogeneous resource nodes according to the cloud-edge-end collaborative platform to establish a power grid heterogeneous resource node network; based on the power grid heterogeneous resource node network, the cloud-edge-end collaborative platform is interacted to obtain multiple power grid resource node monitoring blocks; based on the node abnormality constraint conditions, the power grid heterogeneous resource node network is abnormally predicted and perceived according to multiple power grid resource node monitoring blocks to determine the identified power grid resource node; according to the power grid heterogeneous resource node network, the identified power grid resource node is associated with auxiliary analysis to generate an auxiliary power grid resource node network, wherein the auxiliary power grid resource node network includes a task auxiliary power grid resource node subnetwork and an operation and maintenance auxiliary power grid resource node subnetwork; based on the resource auxiliary scheduling adaptation constraint rules, the auxiliary power grid resource node network is combined with the auxiliary power grid resource node network to make auxiliary resource scheduling decisions for the identified power grid resource nodes, and a resource auxiliary scheduling plan is generated; based on the cloud-edge-end collaborative platform, the identified power grid resource nodes are resource-assisted managed according to the auxiliary power grid resource node network and the resource auxiliary scheduling plan. The present invention solves the technical problems of low timeliness and poor accuracy of resource management scheduling for abnormal power grid resource nodes in the prior art when performing power grid heterogeneous resource management and scheduling. Through the cloud-edge-end collaborative platform, heterogeneous resource nodes of the power grid are integrated to conduct real-time monitoring and anomaly prediction, generate task and operation and maintenance auxiliary sub-networks, improve the timeliness and accuracy of resource management and scheduling of abnormal power grid resource nodes, and improve the technical effect of improving the quality of heterogeneous resource management and scheduling of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] 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.

[0010] Figure 1 A schematic diagram of a process flow of a power grid heterogeneous flexible resource management and scheduling method based on cloud-edge-end collaboration provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of the process of generating a resource-assisted scheduling scheme in the cloud-edge-end collaboration-based power grid heterogeneous flexible resource management and scheduling method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] This application provides a method for heterogeneous and flexible resource management and scheduling of power grids based on cloud-edge-end collaboration, which solves the technical problem of low timeliness and poor accuracy of resource management and scheduling for abnormal power grid resource nodes in the prior art when performing heterogeneous power grid resource management and scheduling. The cloud-edge-end collaborative platform integrates heterogeneous power grid resource nodes, performs real-time monitoring and abnormality prediction, generates tasks and operation and maintenance auxiliary sub-networks, improves the timeliness and accuracy of resource management and scheduling of abnormal power grid resource nodes, and improves the technical effect of heterogeneous power grid resource management and scheduling quality.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0015] Embodiment 1

[0016] like Figure 1 As shown, the present application provides a method for managing and scheduling heterogeneous flexible resources in a power grid based on cloud-edge-end collaboration, the method comprising:

[0017] Step S100: Perform feature-driven integration of multiple power grid heterogeneous resource nodes based on the cloud-edge-end collaborative platform to establish a power grid heterogeneous resource node network.

[0018] In the embodiments of the present application, the cloud-edge-end collaborative platform combines the advantages of cloud computing and edge computing. Cloud computing provides powerful computing and storage capabilities, while edge computing performs processing close to the data source, with the advantages of low latency and local real-time processing. Grid heterogeneous resource nodes refer to various types of resource nodes in the grid, such as power stations, substations, battery storage facilities, wind power generation, etc.

[0019] Feature-driven integration refers to data processing and integration based on the feature data of these power grid resource nodes to build a unified power grid resource node network. Specifically, first, the feature data of each power grid resource node is obtained. These feature data include geographic location, capacity, current status, historical operation data, fault records, etc. Next, the acquired feature data is cleaned and preprocessed. Then, key features are extracted from the cleaned and preprocessed data to construct feature vectors. These feature vectors include the power output, load demand, and operation stability of the node. Then, through cluster analysis, principal component analysis, feature selection, etc., the features of different types of nodes are fused to form a unified feature representation.

[0020] After completing the feature-driven integration, the next step is to build a network of power grid heterogeneous resource nodes based on the integrated feature data. Specifically, the connection relationship between nodes is determined based on information such as node characteristics and geographical distribution. For example, geographically adjacent nodes have a strong connection relationship, and functionally similar nodes are also interconnected. Finally, the network topology of power grid heterogeneous resource nodes is constructed to form a graph structure, in which nodes represent power grid resources and edges represent the connection relationship between nodes.

[0021] Through the above steps, feature-driven integration of multiple power grid heterogeneous resource nodes was carried out based on the cloud-edge-end collaborative platform, and a power grid heterogeneous resource node network was established.

[0022] Furthermore, in the method provided in the embodiment of the application, a plurality of power grid heterogeneous resource nodes are feature-driven integrated according to the cloud-edge-end collaborative platform to establish a power grid heterogeneous resource node network, and the method also includes:

[0023] Based on the multiple power grid heterogeneous resource nodes, the cloud-edge collaborative platform is interacted to obtain the power grid resource node attribute feature flow and the power grid resource node association feature flow; data cleaning is performed according to the power grid resource node attribute feature flow to obtain the first feature-driven integration condition; data cleaning is performed according to the power grid resource node association feature flow to obtain the second feature-driven integration condition; according to the first feature-driven integration condition, the multiple power grid heterogeneous resource nodes are integrated with node attribute feature to generate an initial power grid resource node network; according to the second feature-driven integration condition, the initial power grid resource node network is integrated with node association feature to obtain the power grid heterogeneous resource node network.

[0024] In the embodiment of the present application, first, data is exchanged with the cloud-edge collaborative platform through the power grid heterogeneous resource nodes. During the data interaction process, the power grid resource node attribute feature stream and the power grid resource node association feature stream are obtained. The power grid resource node attribute feature stream contains the attribute information of each resource node, such as geographical location, capacity, type, current status, etc.; the power grid resource node association feature stream contains the association information between nodes, such as the power transmission relationship between nodes, the geographical proximity relationship, etc.

[0025] Next, the two feature streams are cleaned. For the grid resource node attribute feature stream, data cleaning algorithms such as outlier detection and missing value filling are used to ensure that the attribute data of each node is accurate and obtain the first feature-driven integration condition. For the grid resource node association feature stream, data fusion and consistency verification are used to ensure that the association data between nodes is accurate and obtain the second feature-driven integration condition.

[0026] Then, based on the first feature-driven integration condition, multiple power grid heterogeneous resource nodes are integrated with node attribute features to generate an initial power grid resource node network. In this step, by integrating the cleaned attribute feature data, clustering algorithms such as K-Means or DBSCAN and network construction algorithms such as Delaunay triangulation or Voronoi diagram are used to construct an initial network containing all nodes. The position and attributes of each node in the network are determined by its attribute feature data.

[0027] Next, based on the second feature-driven integration condition, the node association feature-driven integration is performed on the initial power grid resource node network. By integrating the cleaned association feature data, the initial network is further optimized and adjusted to more accurately reflect the actual association relationship between nodes. This step uses graph theory algorithms, such as minimum spanning tree, network flow algorithm, and relational network analysis technology, to ensure the accuracy of the association between nodes and the optimization of the network.

[0028] Finally, after two feature-driven integrations, a grid heterogeneous resource node network is generated.

[0029] Step S200: According to the power grid heterogeneous resource node network, interact with the cloud-edge collaborative platform to obtain multiple power grid resource node monitoring blocks.

[0030] In an embodiment of the present application, data is first interacted with the cloud-edge collaborative platform through a grid heterogeneous resource node network. Next, detailed monitoring data is collected from each grid resource node. These nodes are various types of resource nodes in the power grid, such as power stations, substations, battery storage facilities, etc. The collected monitoring data include voltage, current, power, temperature, operating status, etc. The collected monitoring data is then processed and analyzed in the cloud, including data cleaning and data fusion. Specifically, data cleaning algorithms such as outlier detection and missing value filling are first used to remove noise data and invalid data to ensure the accuracy and consistency of the data. Then data fusion techniques such as multi-source data fusion and data integration are used to integrate data from different nodes.

[0031] After processing and analyzing the data, the data is divided into multiple monitoring blocks based on a specific time window or geographical range to generate multiple power grid resource node monitoring blocks. A power grid resource node monitoring block is a data unit that contains the operating status and key parameters of each node within a specific time period, including voltage, current, power, temperature, operating status and other parameters.

[0032] Step S300: Based on the node abnormality constraint conditions, the abnormality prediction and perception of the power grid heterogeneous resource node network is performed according to the multiple power grid resource node monitoring blocks to determine and identify the power grid resource node.

[0033] In the embodiment of the present application, the node abnormality constraint condition is a preset indicator for identifying abnormal behavior of power grid resource nodes. These conditions include voltage upper and lower limits, current upper and lower limits, power fluctuation range and temperature range, etc. For example, the node voltage must be within the range of 220V±10%, and it is considered abnormal if it exceeds this range; similarly, current and temperature also have corresponding normal operating ranges.

[0034] Then, based on these node abnormality constraints, the monitoring block data is analyzed, abnormal prediction perception is performed, and the monitoring data of each node is compared with the preset abnormality constraints to find out the data points that do not meet the conditions. In the process of abnormal prediction perception, an abnormality coefficient is generated for each power grid resource node. The abnormality coefficient is an indicator that quantifies the degree of node abnormality. The specific steps include calculating the deviation of each monitoring parameter from the normal range, and quantifying the degree of abnormality according to the deviation value. For example, if the voltage of a node continues to exceed the normal range, its abnormality coefficient will be higher, reflecting that its degree of abnormality is larger. Finally, based on the generated abnormality coefficient, it is determined whether the abnormality coefficient of each node exceeds the preset threshold. If it exceeds, it is marked as an abnormal node. Through the above process, the identified power grid resource node is determined, and the identified power grid resource node is the abnormal node.

[0035] Furthermore, in the method provided in the embodiment of the application, based on the node abnormality constraint condition, the abnormality prediction perception of the heterogeneous resource node network of the power grid is performed according to the multiple power grid resource node monitoring blocks, and the identification of the power grid resource node is determined, and the method also includes:

[0036] Anomaly prediction and perception are performed according to the multiple power grid resource node monitoring blocks to generate multiple power grid resource node anomaly coefficients; it is determined whether the multiple power grid resource node anomaly coefficients meet the node anomaly constraint conditions; if any one of the multiple power grid resource node anomaly coefficients does not meet the node anomaly constraint conditions, an identified power grid resource node anomaly coefficient is generated; the power grid heterogeneous resource node network is located according to the identified power grid resource node anomaly coefficient to obtain the identified power grid resource node.

[0037] In the embodiment of the present application, the monitoring block data is analyzed using a machine learning algorithm and a statistical analysis method to generate an abnormal coefficient for each power grid resource node. The abnormal coefficient is an indicator that quantifies the degree of deviation of the node operation state from the normal range. For example, using time series analysis and anomaly detection algorithms such as isolation forests and support vector machines, data points that do not conform to the normal operating mode are detected, and the degree of abnormality of these data points is calculated, and finally an abnormal coefficient is generated.

[0038] After the abnormal coefficient is generated, it is determined whether the abnormal coefficient of each grid resource node meets the node abnormal constraint condition. The node abnormal constraint condition is an indicator range set according to the safety and stability requirements of the grid operation, such as the normal operating range of voltage, current or temperature. If the abnormal coefficient of a node exceeds the preset range, it is deemed to not meet the node abnormal constraint condition.

[0039] If it is detected that the abnormal coefficient of any power grid resource node does not meet the node abnormality constraint condition, the identification power grid resource node abnormal coefficient of the node is generated. The identification power grid resource node abnormal coefficient is a special mark that records the specific abnormal situation of the node and the degree of exceeding the normal range. For example, if the voltage abnormal coefficient of a node is significantly higher than the normal range, its identification abnormal coefficient will record this abnormal situation in detail.

[0040] According to the abnormal coefficient of identifying power grid resource nodes, these abnormal nodes are located in the power grid heterogeneous resource node network using graph theory algorithms and relational network analysis technology. The location of abnormal nodes is accurately located through the association information between power grid resource nodes. Specifically, graph search algorithms such as depth-first search and breadth-first search are used to find and locate the location of abnormal nodes in the network.

[0041] Through these steps, based on the data of multiple power grid resource nodes monitoring blocks, abnormal prediction perception is performed, an abnormal coefficient is generated, and it is judged whether it meets the node abnormal constraint conditions. If not, an identification abnormal coefficient is generated, and the abnormal node is located in the power grid heterogeneous resource node network, so as to obtain the identification power grid resource node.

[0042] Furthermore, in the method provided in the embodiment of the application, abnormal prediction perception is performed according to the plurality of power grid resource node monitoring blocks to generate a plurality of power grid resource node abnormality coefficients, and further includes:

[0043] According to the power grid heterogeneous resource node network, multiple node anomaly prediction and perception dual channels are established, wherein each node anomaly prediction and perception dual channel includes a node anomaly detection channel and a node anomaly risk prediction channel; based on the multiple power grid resource node monitoring blocks, the multiple power grid resource node anomaly coefficients are output according to the multiple node anomaly prediction and perception dual channels.

[0044] In the embodiment of the present application, firstly, the grid heterogeneous resource node network is used to identify and determine the key nodes that need to be monitored and analyzed. Through this network, nodes that are critical to the operation of the grid are identified, and an abnormal prediction perception dual channel is built for these nodes, including a node abnormality detection channel and a node abnormality risk prediction channel.

[0045] The main function of the node anomaly detection channel is to monitor the operating status of the power grid resource nodes in real time and detect abnormal situations. When building the node anomaly detection channel, the Internet of Things technology is used to collect data from each power grid resource node in real time. These data include key parameters such as voltage, current, power, and temperature. After data collection, the data is preprocessed, including data cleaning, denoising, and normalization. Then, the sliding window technology is used to split the time series data into windows of fixed length, and each window is used as a sample. Next, a machine learning algorithm such as isolation forest is used to analyze the preprocessed data to detect abnormalities in the operating status in real time. Isolation forest is an unsupervised learning algorithm based on random forests, which is suitable for anomaly detection of high-dimensional data. Each data point is isolated by building a set of random trees and its anomaly score is calculated. Data points with higher anomaly scores are considered abnormal. The detection results are fed back to the system in real time to generate the current anomaly detection coefficient, which reflects the current operating status of the node.

[0046] The main function of the node abnormal risk prediction channel is to predict the possible abnormal risk of the node in the future based on historical data and real-time data. When constructing the node abnormal risk prediction channel, key features are extracted from the monitoring block data. These features include statistical features such as mean, variance, maximum, minimum and time series features such as autocorrelation coefficient and Fourier transform features. These features help capture trends and abnormal patterns in the data. The sliding window technology is used to divide the time series data into windows of fixed length, and each window is used as a sample. The time series model is trained using historical monitoring data. The ARIMA model can capture trends and seasonal changes in time series, and is modeled by combining three parts: difference, autoregression and moving average. After training the ARIMA model with historical data, the model can predict future data points. Based on the trained ARIMA model, real-time data is predicted to evaluate the abnormal risk of the node. Based on the trained ARIMA model, real-time data is predicted to evaluate the abnormal risk of the node. The prediction results are quantified as risk coefficients, reflecting the possibility of abnormality of the node in the future. The risk assessment results are combined with the current abnormal detection coefficient to generate a comprehensive abnormality coefficient.

[0047] Finally, based on multiple power grid resource node monitoring blocks, the node anomaly prediction perception dual channel is used to output the anomaly coefficient of each power grid resource node. Specifically, the abnormal situation of the power grid resource node is identified through the node anomaly detection channel to obtain the current anomaly detection coefficient; the future abnormal risk of the node is evaluated through the node abnormal risk prediction channel to obtain the future risk coefficient. The current anomaly detection coefficient and the future abnormal risk coefficient are normalized, and the anomaly coefficient is generated by weighted summation.

[0048] Step S400: performing association auxiliary analysis on the identified power grid resource node according to the power grid heterogeneous resource node network to generate an auxiliary power grid resource node network, wherein the auxiliary power grid resource node network includes a task auxiliary power grid resource node subnetwork and an operation and maintenance auxiliary power grid resource node subnetwork.

[0049] In an embodiment of the present application, when performing association-assisted analysis on the identified power grid resource nodes according to the power grid heterogeneous resource node network, graph theory technology is first used to establish a connection relationship between nodes by analyzing information such as the geographical location, functional similarity, and historical data between the nodes, and to construct a relationship graph, in which the nodes represent power grid resource nodes and the edges represent the relationship between the nodes. Then, an association analysis algorithm, such as the Apriori algorithm or the FP-Growth algorithm, is used to analyze the relationship graph to find other nodes related to the identified node. Through association-assisted analysis, an auxiliary power grid resource node network is generated, which includes a task-assisted power grid resource node subnetwork and an operation and maintenance-assisted power grid resource node subnetwork. The task-assisted subnetwork identifies nodes that collaborate with the identified node during task execution, while the operation and maintenance-assisted subnetwork identifies nodes associated with the identified node during maintenance and management.

[0050] Furthermore, in the method provided in the embodiment of the application, the associated auxiliary analysis is performed on the identified power grid resource node according to the power grid heterogeneous resource node network to generate an auxiliary power grid resource node network, and further includes:

[0051] According to the identified power grid resource nodes, the power grid heterogeneous resource node network is cleaned of abnormal nodes to obtain an optimized heterogeneous resource node network; according to the identified power grid resource nodes, the optimized heterogeneous resource node network is analyzed for task execution association to obtain a plurality of task-assisted power grid resource nodes that meet task execution association constraints; according to the identified power grid resource nodes, the optimized heterogeneous resource node network is analyzed for operation and maintenance association to obtain a plurality of operation and maintenance-assisted power grid resource nodes that meet operation and maintenance execution association constraints; according to the plurality of task-assisted power grid resource nodes, the task-assisted power grid resource node subnetwork is constructed; according to the plurality of operation and maintenance-assisted power grid resource nodes, the operation and maintenance-assisted power grid resource node subnetwork is constructed; according to the task-assisted power grid resource node subnetwork and the operation and maintenance-assisted power grid resource node subnetwork, the auxiliary power grid resource node network is obtained.

[0052] In an embodiment of the present application, when cleaning abnormal nodes of a power grid heterogeneous resource node network according to the identified power grid resource nodes, the nodes identified as abnormal are filtered out and removed or isolated from the network to obtain an optimized heterogeneous resource node network.

[0053] Next, according to the identified power grid resource nodes, the optimized heterogeneous resource node network is subjected to task execution association analysis. The purpose of task execution association analysis is to identify those auxiliary nodes that are closely associated with the identified nodes during task execution. Specifically, first collect and analyze task execution data, including task dependencies and task execution history records. By analyzing these data, understand the task relationship between nodes. Then use association rule mining algorithms (such as the Apriori algorithm) or graph theory algorithms (such as the minimum spanning tree algorithm) to identify nodes with high association with the identified nodes. Through these algorithms, the association patterns between nodes are discovered, thereby obtaining multiple task-assisted power grid resource nodes that meet the task execution association constraints.

[0054] After the task execution association analysis, the operation and maintenance execution association analysis is performed on the optimized heterogeneous resource node network according to the identified power grid resource nodes. The purpose of the operation and maintenance execution association analysis is to identify those auxiliary nodes that are closely associated with the identified nodes in the operation and maintenance management process. First, the operation and maintenance management data, including maintenance records and equipment dependencies, are collected and analyzed. By analyzing these data, the operation and maintenance relationship between nodes is understood. Then, a method similar to the task execution association analysis is used to identify nodes that are highly associated with the identified nodes in operation and maintenance management, and the operation and maintenance association pattern between nodes is discovered, thereby obtaining multiple operation and maintenance auxiliary power grid resource nodes that meet the operation and maintenance execution association constraints.

[0055] Based on the identified multiple task-assisted power grid resource nodes, a task-assisted power grid resource node sub-network is constructed. Specifically, based on the results of the task execution association analysis, the task-assisted nodes are connected in the network to form a sub-network. By connecting the task-assisted nodes, a sub-network dedicated to task execution is formed. Then, a graph theory optimization algorithm, such as the minimum spanning tree algorithm, is used to optimize the task-assisted sub-network to ensure its efficiency and stability.

[0056] Similarly, based on the identified multiple operation and maintenance auxiliary power grid resource nodes, an operation and maintenance auxiliary power grid resource node subnetwork is constructed. Specifically, based on the results of the operation and maintenance execution association analysis, the operation and maintenance auxiliary nodes are connected in the network to form a subnetwork. By connecting the operation and maintenance auxiliary nodes, a subnetwork dedicated to operation and maintenance management is formed. Then, a similar method is used to optimize the operation and maintenance auxiliary subnetwork to ensure its efficiency and reliability in operation and maintenance management. By optimizing the network structure, the efficiency and reliability of operation and maintenance management can be improved.

[0057] Finally, the task-assisted subnetwork and the operation-assisted subnetwork are integrated to construct an auxiliary power grid resource node network.

[0058] Step S500: Based on the resource-assisted scheduling adaptation constraint rule, in combination with the auxiliary power grid resource node network, an auxiliary resource scheduling decision is made on the identified power grid resource node to generate a resource-assisted scheduling plan.

[0059] In the embodiment of the present application, resource-assisted scheduling adaptive constraint rules are predefined rules and standards used to guide the resource scheduling process and ensure that the scheduling scheme is optimized under various constraints. These rules include task execution priority, resource utilization efficiency, cost control, etc.

[0060] Next, the task-assisted power grid resource node subnetwork and the operation-assisted power grid resource node subnetwork are combined to conduct a detailed analysis and evaluation of the identified power grid resource nodes. During the analysis and evaluation process, a variety of technical means, such as optimization algorithms and simulation technologies, are used to assist resource scheduling decisions. Optimization algorithms, such as linear programming, integer programming, and heuristic algorithms are used to solve the optimal scheduling scheme under various constraints. The execution effects of different scheduling schemes are simulated through simulation technology to evaluate their performance in practical applications.

[0061] Through the above analysis and evaluation, a preliminary resource-assisted scheduling plan is generated. Then, the plan is optimized and adjusted in multiple rounds to further improve its effectiveness and feasibility. This includes repeated testing and modification of the scheduling plan to ensure that it can meet the expected performance indicators under various operating conditions.

[0062] Finally, the final resource-assisted scheduling plan is generated, which specifies in detail the specific scheduling measures for the identified power grid resource nodes, including resource allocation, task execution plan and operation and maintenance management arrangements.

[0063] Further, such as Figure 2 As shown, in the method provided by the embodiment of the application, based on the resource-assisted scheduling adaptation constraint rule, in combination with the auxiliary power grid resource node network, an auxiliary resource scheduling decision is made on the identified power grid resource node to generate a resource-assisted scheduling plan, and also includes:

[0064] The resource-assisted scheduling adaptation constraint rules include task-assisted resource scheduling adaptation constraints and operation-and-maintenance-assisted resource scheduling adaptation constraints; based on the task-assisted resource scheduling adaptation constraints, a task-assisted resource scheduling decision is made on the identified power grid resource node according to the task-assisted power grid resource node subnetwork to obtain a task-assisted resource scheduling plan; based on the operation-and-maintenance-assisted resource scheduling adaptation constraints, an operation-and-maintenance-assisted resource scheduling decision is made on the identified power grid resource node according to the operation-and-maintenance-assisted power grid resource node subnetwork to obtain an operation-and-maintenance-assisted resource scheduling plan; the task-assisted resource scheduling plan and the operation-and-maintenance-assisted resource scheduling plan are added to the resource-assisted scheduling plan.

[0065] In the embodiment of the present application, task-assisted resource scheduling adaptation constraints include resource capacity limitations, task priorities, time constraints, task dependencies, resource utilization, etc. Operation and maintenance-assisted resource scheduling adaptation constraints include maintenance time windows, maintenance priorities, resource availability, node dependencies, and cost constraints, etc.

[0066] Based on the task-assisted resource scheduling adaptation constraints, when making task-assisted resource scheduling decisions for the identified power grid resource nodes according to the task-assisted power grid resource node sub-network, a resource scheduling decision model based on task-assisted resource scheduling adaptation constraints is established, and the optimal resource allocation and scheduling scheme is determined using optimization algorithms such as linear programming or integer programming. The optimization goal is to minimize the task execution time, maximize resource utilization or minimize the scheduling cost, and solve according to the constraints such as resource capacity limit, task priority, time constraint and task dependency, and finally obtain a specific task-assisted resource scheduling scheme.

[0067] Based on the adaptive constraints of operation and maintenance auxiliary resource scheduling, when making operation and maintenance auxiliary resource scheduling decisions for the identified power grid resource nodes according to the operation and maintenance auxiliary power grid resource node sub-network, a resource scheduling decision model based on the adaptive constraints of operation and maintenance auxiliary resource scheduling is established, and the optimal resource allocation and scheduling scheme is determined using optimization algorithms such as linear programming or integer programming. The optimization goal is to minimize the operation and maintenance time, maximize resource utilization or minimize maintenance costs, and solve according to constraints such as maintenance time window, maintenance priority, resource availability and node dependency, and finally obtain a specific operation and maintenance auxiliary resource scheduling scheme.

[0068] Finally, the task-assisted resource scheduling scheme and the operation and maintenance-assisted resource scheduling scheme are added to the resource-assisted scheduling scheme.

[0069] Furthermore, in the method provided in the embodiment of the application, based on the task-assisted resource scheduling adaptation constraint, a task-assisted resource scheduling decision is made on the identified power grid resource node according to the task-assisted power grid resource node subnetwork to obtain a task-assisted resource scheduling solution, and further includes:

[0070] According to the task-assisted power grid resource node subnetwork, random task-assisted resource scheduling decisions are made on the identified power grid resource node to obtain a task-assisted resource scheduling decision particle group; according to the task-assisted resource scheduling decision particle group, a first task-assisted resource scheduling decision particle is obtained; according to the first task-assisted resource scheduling decision particle, fitting task-assisted resource scheduling is performed on the identified power grid resource node to obtain a first task-assisted resource scheduling fitness; it is determined whether the first task-assisted resource scheduling fitness satisfies the task-assisted resource scheduling adaptation constraint; if the first task-assisted resource scheduling fitness satisfies the task-assisted resource scheduling adaptation constraint, the first task-assisted resource scheduling decision particle is added to the task-assisted resource scheduling decision space; according to the task-assisted resource scheduling adaptation constraint, the task-assisted resource scheduling decision particle group is continued to be traversed and screened to construct the task-assisted resource scheduling decision space; according to the task-assisted resource scheduling decision space, the task-assisted resource scheduling fitness is maximized and optimized to generate the task-assisted resource scheduling plan.

[0071] In the embodiment of the present application, when making random task-assisted resource scheduling decisions, a group of scheduling decisions is randomly generated to form a task-assisted resource scheduling decision particle group. Each particle represents a possible scheduling solution, including specific resource allocation and task execution order.

[0072] Next, a particle is randomly selected from the task-assisted resource scheduling decision particle group as the first task-assisted resource scheduling decision particle. This particle represents a preliminary scheduling scheme with specific resource allocation and task execution order. Then, according to the first task-assisted resource scheduling decision particle, the identified power grid resource node is fitted with task-assisted resource scheduling, and its fitness is calculated by simulating the execution of the scheduling scheme. It is judged whether the fitness of the first task-assisted resource scheduling satisfies the task-assisted resource scheduling adaptation constraint. If it does, the first task-assisted resource scheduling decision particle is added to the task-assisted resource scheduling decision space. The task-assisted resource scheduling adaptation constraints include resource capacity constraints, task priorities, time constraints, task dependencies, and resource utilization. These constraints ensure the feasibility and effectiveness of the scheduling scheme in practical applications.

[0073] Then continue to traverse and screen the task-assisted resource scheduling decision particle swarm, evaluate the fitness of each particle one by one, and determine whether it meets the task-assisted resource scheduling fitness constraints. Add the particles that meet the constraints to the task-assisted resource scheduling decision space. This process evaluates the fitness of each particle to ensure that only scheduling solutions that meet the constraints are retained.

[0074] According to the constructed task-assisted resource scheduling decision space, the task-assisted resource scheduling fitness is maximized and optimized. Use optimization algorithms, such as genetic algorithms or particle swarm optimization algorithms, to optimize the particles in the decision space and find the scheduling solution with the highest fitness.

[0075] Finally, a task-assisted resource scheduling solution is generated, which is the best solution after optimizing the fitness of particles in the task-assisted resource scheduling decision space.

[0076] Furthermore, in the method provided in the embodiment of the application, the task-assisted resource scheduling is performed on the identified power grid resource node according to the first task-assisted resource scheduling decision particle to obtain the first task-assisted resource scheduling fitness, and further includes:

[0077] According to the first task-assisted resource scheduling decision particle, a fitting task-assisted resource scheduling is performed on the identified power grid resource node to obtain a first fitting resource scheduling cost coefficient, a first fitting resource scheduling task-assisted efficiency coefficient and a first fitting resource scheduling task-assisted quality coefficient; according to the first fitting resource scheduling cost coefficient, the first fitting resource scheduling task-assisted efficiency coefficient and the first fitting resource scheduling task-assisted quality coefficient, normalization processing is performed to generate a first particle fitting resource scheduling analysis result; according to the task-assisted resource scheduling adaptation weight condition, a weighted calculation is performed on the first particle fitting resource scheduling analysis result to generate the first task-assisted resource scheduling fitness.

[0078] In an embodiment of the present application, when performing fitting task auxiliary resource scheduling, the first fitting resource scheduling cost coefficient, the first fitting resource scheduling task auxiliary efficiency coefficient and the first fitting resource scheduling task auxiliary quality coefficient are calculated by simulating the execution of the scheduling plan.

[0079] Calculate the cost coefficient of the first-fit resource scheduling by activity-based costing. Specifically, by analyzing the resource usage records, evaluate the economic cost of the scheduling scheme, including the cost of resource usage, equipment operating costs, and labor time costs. For example, by recording the frequency and time of use of each resource, calculate the corresponding cost. Use data envelopment analysis tools such as DEAP or MaxDEA to evaluate the first-fit resource scheduling task auxiliary efficiency coefficient. Specifically, by analyzing the total time required for task execution and the utilization rate of each resource, evaluate the execution efficiency of the scheduling scheme. For example, by counting the execution time and resource utilization of each task, calculate the overall efficiency coefficient. Use total quality management or six sigma methods to evaluate the first-fit resource scheduling task auxiliary quality coefficient through quality management software such as Minitab or JMP. Specifically, by analyzing the accuracy, reliability, and failure rate of task completion, evaluate the quality of the scheduling scheme. For example, by recording the success rate and failure rate of each task, calculate the corresponding quality coefficient.

[0080] Next, the first fitting resource scheduling cost coefficient, the first fitting resource scheduling task auxiliary efficiency coefficient and the first fitting resource scheduling task auxiliary quality coefficient are subjected to Min-Max normalization processing to generate the first particle fitting resource scheduling analysis result.

[0081] Then, according to the task-assisted resource scheduling adaptation weight condition, the normalized analysis results are weighted, where the weight of each coefficient is pre-set. By combining the weighted values ​​of cost, efficiency and quality, the first task-assisted resource scheduling fitness is generated.

[0082] Step S600: Based on the cloud-edge-terminal collaborative platform, perform resource-assisted management on the identified power grid resource node according to the auxiliary power grid resource node network and the resource-assisted scheduling scheme.

[0083] In the embodiment of the present application, based on the cloud-edge-end collaborative platform, when performing resource-assisted management according to the auxiliary power grid resource node network and the resource-assisted scheduling scheme, first allocate necessary resources, such as electricity and equipment to the identification node according to the resource-assisted scheduling scheme to ensure its normal operation. Then, according to the task relationship and priority provided by the task-assisted subnetwork, arrange the specific task execution plan to ensure the efficient completion of the task. According to the operation and maintenance information of the operation and maintenance auxiliary subnetwork, formulate a detailed operation and maintenance management plan, including regular maintenance, troubleshooting, and performance monitoring.

[0084] Through the above steps, based on the coordination and management of the cloud-edge collaborative platform, the identified power grid resource nodes have received effective resource support and management, ensuring the efficient operation and stability of the power grid.

[0085] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0086] This application performs feature-driven integration of multiple power grid heterogeneous resource nodes according to the cloud-edge-end collaborative platform to establish a power grid heterogeneous resource node network; based on the power grid heterogeneous resource node network, the cloud-edge-end collaborative platform is interacted to obtain multiple power grid resource node monitoring blocks; based on the node abnormality constraint conditions, the power grid heterogeneous resource node network is abnormally predicted and perceived according to multiple power grid resource node monitoring blocks to determine the identified power grid resource node; according to the power grid heterogeneous resource node network, the identified power grid resource node is associated with auxiliary analysis to generate an auxiliary power grid resource node network, wherein the auxiliary power grid resource node network includes a task auxiliary power grid resource node subnetwork and an operation and maintenance auxiliary power grid resource node subnetwork; based on the resource auxiliary scheduling adaptation constraint rules, the auxiliary power grid resource node network is combined with the auxiliary power grid resource node network to make auxiliary resource scheduling decisions for the identified power grid resource nodes, and a resource auxiliary scheduling plan is generated; based on the cloud-edge-end collaborative platform, the identified power grid resource nodes are resource-assisted managed according to the auxiliary power grid resource node network and the resource auxiliary scheduling plan. The present invention solves the technical problems of low timeliness and poor accuracy of resource management scheduling for abnormal power grid resource nodes in the prior art when performing power grid heterogeneous resource management and scheduling. Through the cloud-edge-end collaborative platform, heterogeneous resource nodes of the power grid are integrated to conduct real-time monitoring and anomaly prediction, generate task and operation and maintenance auxiliary sub-networks, improve the timeliness and accuracy of resource management and scheduling of abnormal power grid resource nodes, and improve the technical effect of improving the quality of heterogeneous resource management and scheduling of the power grid.

[0087] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0089] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for managing and scheduling heterogeneous resources in power grids based on cloud-edge-end collaboration, characterized in that: The method comprises: Based on the cloud-edge-end collaborative platform, multiple power grid heterogeneous resource nodes are feature-driven integrated to establish a power grid heterogeneous resource node network; According to the power grid heterogeneous resource node network, the cloud-edge-end collaborative platform is interacted to obtain multiple power grid resource node monitoring blocks; Based on the node abnormality constraint condition, the abnormality prediction perception of the power grid heterogeneous resource node network is performed according to the multiple power grid resource node monitoring blocks to determine and identify the power grid resource node; Performing an auxiliary correlation analysis on the identified power grid resource node according to the power grid heterogeneous resource node network to generate an auxiliary power grid resource node network, wherein the auxiliary power grid resource node network includes a task auxiliary power grid resource node subnetwork and an operation and maintenance auxiliary power grid resource node subnetwork; Based on the resource-assisted scheduling adaptation constraint rule, in combination with the auxiliary power grid resource node network, an auxiliary resource scheduling decision is made on the identified power grid resource node to generate a resource-assisted scheduling plan; Based on the cloud-edge collaborative platform, resource-assisted management is performed on the identified power grid resource nodes according to the auxiliary power grid resource node network and the resource-assisted scheduling scheme.

2. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration according to claim 1 is characterized in that: Based on the cloud-edge-end collaborative platform, multiple power grid heterogeneous resource nodes are feature-driven integrated to establish a power grid heterogeneous resource node network, including: Based on the multiple power grid heterogeneous resource nodes, interact with the cloud-edge-end collaborative platform to obtain a power grid resource node attribute feature flow and a power grid resource node associated feature flow; Perform data cleaning according to the grid resource node attribute feature flow to obtain a first feature-driven integration condition; Perform data cleaning according to the grid resource node associated characteristic flow to obtain a second characteristic driven integration condition; Perform node attribute feature-driven integration on the plurality of power grid heterogeneous resource nodes according to the first feature-driven integration condition to generate an initial power grid resource node network; The initial power grid resource node network is integrated by node association feature driving according to the second feature driving integration condition to obtain the power grid heterogeneous resource node network.

3. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration according to claim 1 is characterized in that: Based on the node abnormality constraint condition, the abnormality prediction perception of the power grid heterogeneous resource node network is performed according to the multiple power grid resource node monitoring blocks, and the power grid resource node is determined and identified, including: Performing abnormal prediction and perception according to the plurality of power grid resource node monitoring blocks, generating a plurality of power grid resource node abnormality coefficients; Determining whether the abnormality coefficients of the plurality of power grid resource nodes satisfy the node abnormality constraint condition; If any one of the plurality of power grid resource node abnormality coefficients does not satisfy the node abnormality constraint condition, generating an identification power grid resource node abnormality coefficient; The power grid heterogeneous resource node network is located according to the abnormal coefficient of the identified power grid resource node to obtain the identified power grid resource node.

4. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration as claimed in claim 3 is characterized in that: Performing abnormal prediction and perception according to the plurality of power grid resource node monitoring blocks, generating a plurality of power grid resource node abnormality coefficients, including: According to the power grid heterogeneous resource node network, multiple node anomaly prediction perception dual channels are established, wherein each node anomaly prediction perception dual channel includes a node anomaly detection channel and a node anomaly risk prediction channel; Based on the multiple power grid resource node monitoring blocks and according to the multiple node abnormality prediction perception dual channels, the multiple power grid resource node abnormality coefficients are output.

5. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration according to claim 1 is characterized in that: Performing an auxiliary correlation analysis on the identified power grid resource node according to the power grid heterogeneous resource node network to generate an auxiliary power grid resource node network, including: Cleaning abnormal nodes of the power grid heterogeneous resource node network according to the identified power grid resource node to obtain an optimized heterogeneous resource node network; Performing task execution association analysis on the optimized heterogeneous resource node network according to the identified power grid resource node to obtain a plurality of task auxiliary power grid resource nodes that meet task execution association constraints; Performing operation and maintenance execution association analysis on the optimized heterogeneous resource node network according to the identified power grid resource node to obtain a plurality of operation and maintenance auxiliary power grid resource nodes that meet the operation and maintenance execution association constraints; constructing the mission-assisted power grid resource node subnetwork according to the plurality of mission-assisted power grid resource nodes; Constructing the operation and maintenance auxiliary power grid resource node subnetwork according to the plurality of operation and maintenance auxiliary power grid resource nodes; The auxiliary power grid resource node network is obtained according to the task auxiliary power grid resource node subnetwork and the operation and maintenance auxiliary power grid resource node subnetwork.

6. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration according to claim 1, characterized in that: Based on the resource-assisted scheduling adaptation constraint rule, in combination with the auxiliary power grid resource node network, an auxiliary resource scheduling decision is made on the identified power grid resource node to generate a resource-assisted scheduling plan, including: The resource-assisted scheduling adaptation constraint rules include task-assisted resource scheduling adaptation constraints and operation-assisted resource scheduling adaptation constraints; Based on the task-assisted resource scheduling adaptation constraint, a task-assisted resource scheduling decision is made on the identified power grid resource node according to the task-assisted power grid resource node subnetwork to obtain a task-assisted resource scheduling solution; Based on the operation and maintenance auxiliary resource scheduling adaptation constraint, an operation and maintenance auxiliary resource scheduling decision is made on the identified power grid resource node according to the operation and maintenance auxiliary power grid resource node subnetwork to obtain an operation and maintenance auxiliary resource scheduling plan; The task-assisted resource scheduling scheme and the operation and maintenance-assisted resource scheduling scheme are added to the resource-assisted scheduling scheme.

7. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration according to claim 6 is characterized in that: Based on the task-assisted resource scheduling adaptation constraint, a task-assisted resource scheduling decision is made on the identified power grid resource node according to the task-assisted power grid resource node subnetwork to obtain a task-assisted resource scheduling solution, including: Performing random task-assisted resource scheduling decisions on the identified power grid resource nodes according to the task-assisted power grid resource node subnetwork to obtain a task-assisted resource scheduling decision particle swarm; Obtaining a first task-assisted resource scheduling decision particle according to the task-assisted resource scheduling decision particle swarm; Performing a fitting task-assisted resource scheduling on the identified power grid resource node according to the first task-assisted resource scheduling decision particle to obtain a first task-assisted resource scheduling fitness; Determining whether the first task-assisted resource scheduling fitness satisfies the task-assisted resource scheduling fitness constraint; If the first task-assisted resource scheduling fitness satisfies the task-assisted resource scheduling fitness constraint, adding the first task-assisted resource scheduling decision particle to the task-assisted resource scheduling decision space; According to the task-assisted resource scheduling adaptation constraint, continue to traverse and screen the task-assisted resource scheduling decision particle swarm to construct the task-assisted resource scheduling decision space; The task-assisted resource scheduling fitness is maximized and optimized according to the task-assisted resource scheduling decision space to generate the task-assisted resource scheduling solution.

8. The method for managing and scheduling heterogeneous resources in a power grid based on cloud-edge-end collaboration according to claim 7 is characterized in that: Performing a fitting task-assisted resource scheduling on the identified power grid resource node according to the first task-assisted resource scheduling decision particle to obtain a first task-assisted resource scheduling fitness includes: Performing a fitting task auxiliary resource scheduling on the identified power grid resource node according to the first task auxiliary resource scheduling decision particle to obtain a first fitting resource scheduling cost coefficient, a first fitting resource scheduling task auxiliary efficiency coefficient, and a first fitting resource scheduling task auxiliary quality coefficient; Performing normalization processing according to the first fitting resource scheduling cost coefficient, the first fitting resource scheduling task auxiliary efficiency coefficient, and the first fitting resource scheduling task auxiliary quality coefficient to generate a first particle fitting resource scheduling analysis result; The first particle fitting resource scheduling analysis result is weightedly calculated according to the task-assisted resource scheduling adaptation weight condition to generate the first task-assisted resource scheduling fitness.

Images