Resource management method and device for power private network, computer device and readable storage medium

Abstract

The application relates to a resource management method and device for a power special network, computer equipment and a readable storage medium, and relates to the field of 5G power special networks. The method comprises the following steps: collecting power service information, network information and computing power information of the power special network; constructing a power special network knowledge graph according to the power service information, the network information and the computing power information; periodically monitoring power service state changes and resource state changes in the power special network; wherein the resource state comprises the state of network resources and the state of computing power resources; updating the power special network knowledge graph according to the power service state changes and the resource state changes; screening power services whose resource demands are not satisfied from the latest power special network knowledge graph, and updating the resource scheduling strategy corresponding to the power services; and issuing the resource scheduling strategy to the network nodes corresponding to the power services for execution. The method can improve resource utilization.

Description

Technical Field

[0001] This application relates to the field of 5G power private networks, and in particular to a resource management method, apparatus, computer equipment, and computer-readable storage medium for power private networks. Background Technology

[0002] With the rapid development of communication technology, the power industry is gradually building dedicated virtual networks for power services and deploying multiple access edge computing nodes at the network edge to improve the processing efficiency and security of power services. The dedicated power network carries various services such as power generation, transmission, transformation, distribution, and consumption, and these different services have significantly different requirements for service resources such as network bandwidth and computing power.

[0003] Currently, resource management in the above scenarios mainly adopts manual static configuration. However, the actual load of power business has strong dynamism and uncertainty, and static configuration is difficult to match business demand and resource supply in real time, resulting in low resource utilization. Summary of the Invention

[0004] Therefore, it is necessary to provide a resource management method, apparatus, computer equipment, and computer-readable storage medium for private power grids that can improve resource utilization in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a resource management method for private power grids, including:

[0006] Collect power business information, network information, and computing power information from the dedicated power grid;

[0007] Construct a knowledge graph of the power private network based on power business information, network information, and computing power information;

[0008] Periodically monitor changes in the status of power services and resources within the dedicated power grid; the resource status includes the status of network resources and computing resources.

[0009] The knowledge graph of the private power network is updated based on changes in power business status and resource status.

[0010] The system filters out power services whose resource demands are not being met from the latest power private network knowledge graph and updates the corresponding resource scheduling strategies for these power services.

[0011] The resource scheduling strategy is distributed to the network nodes corresponding to the power business for execution.

[0012] In one embodiment, resource requirements include network resource requirements; filtering power services whose resource requirements are not met from the latest power private network knowledge graph includes:

[0013] For each power service in the latest power private network knowledge graph, obtain the transmission delay, transmission bandwidth, and packet loss rate of the power service;

[0014] If at least one of the following conditions is met: transmission delay is less than or equal to the delay threshold, transmission bandwidth is less than the bandwidth threshold, or packet loss rate is less than the packet loss threshold, then the network resource requirements of the power service are determined to be unmet.

[0015] In one embodiment, resource requirements also include computing power resource requirements; filtering power services whose resource requirements are not met from the latest power private network knowledge graph includes:

[0016] For each power service in the latest power private network knowledge graph, obtain the computing power and memory of the power service;

[0017] If at least one of the following conditions is met: computing power is below the capacity threshold or memory is below the memory threshold, it is determined that the computing resource requirements of the power business are not being met.

[0018] In one embodiment, the resource scheduling strategy includes a network resource scheduling strategy for the base station area where the power terminal carrying the power service is located; updating the resource scheduling strategy corresponding to the power service includes:

[0019] Extract the set of power terminals that are in the same base station area as power services from the latest power private network knowledge graph;

[0020] Configure the network resources of each power terminal according to the corresponding service weight of each power terminal in the power terminal set, so as to update the network resource scheduling strategy of the base station area.

[0021] In one embodiment, the resource scheduling strategy includes a computing resource scheduling strategy related to power services; updating the resource scheduling strategy corresponding to power services includes:

[0022] Extract the set of edge servers from the latest power private network knowledge graph, and obtain the available computing power resources corresponding to each edge server in the set, as well as the transmission path between the edge server and the power business.

[0023] Based on available computing resources and transmission paths, determine the target weights corresponding to edge servers;

[0024] The edge servers are sorted in descending order according to their target weights to obtain the computing power resource queue.

[0025] The computing resources in the computing resource queue are allocated to the power business in sequence to update the computing resource scheduling strategy related to the power business until the computing resource requirements of the power business are met.

[0026] In one embodiment, the change in power service status includes power service migration; updating the resource scheduling strategy corresponding to the power service includes:

[0027] Release the resources currently allocated to the migrating power service;

[0028] Identify the target power terminals corresponding to other power services that are located in the same base station area as the migrated power service and whose resource requirements have not been met;

[0029] The released resources are allocated to the target power terminal according to the weight of the bearer service corresponding to the target power terminal.

[0030] Determine the target base station area after the migration of power services, and reallocate resources according to the service weight of power terminals already carrying services in the target base station area and the service weight of power terminals carrying the migrated power services.

[0031] In one embodiment, resource status changes include adding new resources; updating the resource scheduling strategy corresponding to the power service includes:

[0032] Determine the base station area to which the power service belongs. If there is a closed power service in the base station area, release the resources allocated to the closed power service and obtain new resources.

[0033] The new resources will be allocated to the power services according to the weight of the power terminals that carry the power services.

[0034] Secondly, this application also provides a resource management device for private power grids, comprising:

[0035] The information acquisition module is used to collect power business information, network information, and computing power information from the dedicated power grid.

[0036] The knowledge graph construction and update module is used to construct a knowledge graph for the power private network based on power business information, network information, and computing power information.

[0037] The monitoring module is used to periodically monitor changes in the status of power services and resources in the private power grid; the resource status includes the status of network resources and computing resources.

[0038] The knowledge graph construction and updating module is also used to update the power private network knowledge graph based on changes in power business status and resource status.

[0039] The strategy update module is used to filter out power services whose resource demands are not being met from the latest power private network knowledge graph and update the resource scheduling strategy corresponding to the power services.

[0040] The policy update module is also used to distribute resource scheduling policies to the network nodes corresponding to the power business for execution.

[0041] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described resource management method for power private networks.

[0042] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described resource management method for power private networks.

[0043] The aforementioned resource management method, apparatus, computer equipment, and computer-readable storage medium for power private networks first collect power service information, network information, and computing power information from the power private network to construct a power private network knowledge graph. Next, it periodically monitors changes in the power service status and resource status within the power private network; the resource status includes the status of network resources and computing power resources. The power private network knowledge graph is updated based on these changes, and power services with unmet resource demands are selected from the latest knowledge graph, with corresponding resource scheduling strategies updated. Finally, the resource scheduling strategies are distributed to the network nodes corresponding to the power services for execution. The introduction of the power private network knowledge graph not only integrates dispersed power services and resource supply into a unified structured model but also significantly improves the responsiveness and accuracy of resource scheduling decisions to dynamic changes in the power private network. Therefore, resource allocation no longer relies on static manual configuration but adaptively adjusts based on actual changes in the power service status and resource status within the power private network, significantly improving overall resource utilization efficiency. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is an application environment diagram of a resource management method for a private power grid in one embodiment;

[0046] Figure 2 This is a flowchart illustrating a resource management method for a private power grid in one embodiment;

[0047] Figure 3This is a schematic diagram of the knowledge graph of a private power grid in one embodiment;

[0048] Figure 4 This is a schematic diagram of a power private network knowledge graph in one embodiment;

[0049] Figure 5 This is a schematic diagram of the architecture of a resource management system in one embodiment;

[0050] Figure 6 This is a structural block diagram of a resource management device for a private power grid in one embodiment;

[0051] Figure 7 This is an internal structural diagram of a computer device in one embodiment;

[0052] Figure 8 This is a diagram of the internal structure of a computer device in another embodiment. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0054] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0055] With the rapid development of the 5G mobile communication system, the power industry has gradually built a dedicated virtual network for power services, namely the 5G power virtual private network (VPN), and deployed multi-access edge computing (MEC) nodes at the network edge. By applying 5G to the power sector and utilizing network slicing technology, a 5G power VPN can be formed to provide isolated, high-bandwidth, low-latency, and massive connectivity services for different power services, thereby meeting differentiated quality of service requirements. On the other hand, traditional cloud computing paradigms suffer from long network transmission latency, data privacy and security risks, making it difficult to meet the power industry's demand for low-latency, high-security, and high-performance intelligent computing. Therefore, deploying computing servers within the 5G power VPN and migrating power computing tasks to the private network for MEC edge computing is an important way to reduce network transmission latency and improve the privacy and security of power data.

[0056] However, power private networks carry various services such as power generation, transmission, transformation, distribution, and consumption, including distribution automation, precise load control, and drone inspections. These different services have significantly different requirements for network bandwidth, computing power, and other service resources. To meet these needs, traditional technologies typically use manual configuration of 5G power virtual private network resources. This not only wastes network and computing resources but also, due to the dynamic nature of the 5G wireless network environment, the attributes of network nodes and computing nodes may change. Static manual configuration often cannot keep up with changes in network status, leading to insufficient resource allocation for some services and excessive resource allocation for others. This wastes resources and fails to meet the differentiated service needs of diverse services. Furthermore, the MEC edge cloud management of 5G power virtual private networks involves multiple fields such as 5G communication, edge computing, and power. Network administrators and power business personnel typically possess only expertise in one field, making it difficult to respond promptly to changes in power business and optimize resource scheduling strategies. Therefore, power private networks currently suffer from low resource utilization efficiency.

[0057] To address the aforementioned problems, embodiments of this application provide a resource management method for private power grids, which can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, smartphones, drones, and power devices. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0058] The resource management method for private power grids provided in this application embodiment can be... Figure 1 The terminal 102 or server 104 can execute independently, or they can interact. The following explanation uses server 104 executing independently as an example. Specifically, server 104 constructs a power private network knowledge graph based on power service information, network information, and computing power information; server 104 periodically monitors changes in the power service status and resource status within the power private network; the resource status includes the status of network resources and computing power resources; server 104 updates the power private network knowledge graph based on these changes; server 104 filters power services whose resource demands are not met from the latest power private network knowledge graph and updates the corresponding resource scheduling strategies; server 104 then distributes the resource scheduling strategies to the network nodes corresponding to the power services for execution.

[0059] In one exemplary embodiment, such as Figure 2 As shown, a resource management method for private power grids is provided, which is applied to... Figure 1 Taking server 104 as an example, the following steps are included:

[0060] Step S202: Collect power business information, network information, and computing power information from the dedicated power grid.

[0061] The power private network refers to a virtual private network for communication within the power industry. For example, a fifth-generation mobile communication virtual private network (VPN) uses slicing technology to logically isolate dedicated channels on the public 5G network, carrying differentiated services from power generation, transmission, transformation, distribution, and consumption. Power service information includes at least one of the following: type of power service, data volume, computing requirements, and quality of service (QoS) requirements. The type of power service can be, but is not limited to, distribution automation, precise load control, or drone inspection; data volume refers to the number of bytes generated per unit time; computing requirements can include at least one of the following: required processing unit (PU) or memory capacity; and QoS requirements can include at least one of the following: maximum end-to-end latency, minimum bandwidth, or minimum reliability. Network information includes at least one of the following: the network topology of the power private network and the communication capabilities of each network node. The network topology can include the connections between base stations, power terminals, and edge computing servers; communication capabilities can include at least one of the following: the maximum wireless bandwidth of the base station, link capacity, etc. Computing power information includes at least one of the following: the distribution location of edge servers, computing power, and storage capacity.

[0062] For example, the server collects power business information, network communication information, and computing power information from the power private network to construct a knowledge graph of the power private network based on the collected information.

[0063] Step S204: Construct a knowledge graph of the power private network based on power business information, network information, and computing power information.

[0064] Knowledge graphs are semantic networks that store entities and their relationships using a graph structure. In this embodiment, a knowledge graph ontology model for a power grid is constructed based on the collected information. A partial schematic diagram of the ontology model is shown below. Figure 3 As shown, in the ontology model, ontologies are represented by vertices, and the relationships between ontologies are represented by edges. The ontology model mainly includes five core ontologies: power services, network, nodes, resources, and network topology, as well as the relationships between them. The ontology model describes the concepts, characteristics, and relationships between the core ontologies and extended ontologies of the power private network.

[0065] The power business ontology describes power industry applications including power grid companies and power generation companies, and further includes five major business segments: power generation, transmission, transformation, distribution and consumption, and their corresponding differentiated power business scenarios. Different power business scenarios have different requirements in terms of end-to-end latency, bandwidth, reliability and computing.

[0066] The network entity mainly corresponds to the power private network. Its function is to connect isolated devices or nodes such as power terminals, base stations, core network slices, and computing nodes for power companies, and to provide information interaction and communication capabilities for devices or nodes within the power private network.

[0067] The node itself is the infrastructure of the dedicated power network, including power terminals, network nodes, and computing nodes. Power terminals are the infrastructure for executing power services, possessing functions such as power service processing, data acquisition, and forwarding. The main function of network nodes is to provide data or information forwarding services to power terminals, computing nodes, and edge servers within the dedicated power network. The main function of computing nodes is to provide MEC (Multi-access Edge Computing) edge computing services for power services within the dedicated power network.

[0068] The resource ontology describes the capacity to forward network data or information within a dedicated power grid, or the capacity to provide edge computing. Network topology describes the communication connections of nodes within the dedicated power grid.

[0069] The power service category further includes two attribute entities: network demand and computing power demand. Network demand describes the service's requirements for end-to-end latency, bandwidth, and reliability; computing power demand describes the service's requirements for data processing capabilities such as computing and storage. Resources further include subcategories of network resources and computing power resources. Network resources describe the network service capabilities that network nodes can provide, while computing power resources describe the edge computing capabilities that computing nodes can provide.

[0070] The five core ontologies and their attributes described above contain the following triplet relationships: <network, contains, node>, <network, contains, resource>, <network, contains, network topology>, <node, contains, power terminal>, <node, contains, network node>, <node, contains, compute node>, <power terminal, undertakes, power service>, <network node, contains, resource>, <compute node, contains, resource>, <network node, connection, network topology>, <compute node, connection, network topology>, <power terminal, connection, network topology>, <power service, attribute, network requirement>, <power service, attribute, computing power requirement>, <resource, subclass, network resource>, <resource, subclass, computing power resource>. Among these,<H,R,T> In this context, H, R, and T represent the head entity, relation, and tail entity, respectively. For example, in the triple <network, contain, node>, the network is the head entity, the node is the tail entity, and "contains" is the relation between the head entity and the tail entity.

[0071] For example, based on the basic relationships between entities in the ontology model, and using the collected information as graph data, the relationships between power services, network nodes, computing nodes, network topology, network resources, computing power resources entities and entity attributes are determined. The aforementioned entities and entity attributes are used as vertices, and the aforementioned relationships are used as edges to construct a knowledge graph for the power private network.

[0072] For example, the server collects the following 5G power virtual private network information: Base station 1 serves three power terminals. Power terminal 1 carries distribution automation services, power terminal 2 carries drone inspection services, and power terminal 3 carries precision load control services. Base station 1 connects wirelessly to power terminals 1, 2, and 3, providing a maximum downlink bandwidth of 1Gbps and an uplink bandwidth of 100Mbps to the terminals within its service area. Base station 2 serves three power terminals, two of which carry drone inspection services and one carries distribution automation services. Base station 3 serves three power terminals, two of which carry distribution automation services and one carries drone inspection services. The network requirements for distribution automation are: latency less than 2ms, bandwidth greater than or equal to 19.2kbps, and reliability of 99.99%, with a computing power requirement of 1GHz CPU and 1GB of memory. The network requirements for precision load control are: latency less than 50ms, bandwidth greater than or equal to 0.59kbps, and reliability of 99.99%, with a computing power requirement of 1GHz CPU and 1GB of memory. The network requirements for drone inspection are: latency less than 100ms, bandwidth greater than or equal to 25Mbps, and reliability of 99.9%, with a computing power requirement of 10GHz CPU and 10GB of memory. Three MEC servers are connected to three base stations via wired connections; base station 1 and base station 2, and base station 2 and base station 3 are also connected via wired connections. The computing power resources of all three MEC servers are 10GHz CPU and 16GB of memory. Based on the above information, the server can extract the network node entities: base stations 1 to 3, power terminals 1 to 9; computing node entities: MEC servers 1 to 3; the network resources and computing power resources of each entity; and power business entities: distribution network automation, precision load control, and drone inspection, as well as the network requirements and computing power requirements of these entities. Using these entities and attributes as vertices and the relationships between entities as edges, the following can be obtained: Figure 4 The diagram shown is a knowledge graph of the power grid.

[0073] Figure 4It comprises three base station entities (Base Station 1, Base Station 2, and Base Station 3), each connected to several power IoT terminal entities (e.g., Power Terminal 1 to Power Terminal 9). Each terminal entity carries a corresponding power service entity; for example, Terminal 1 carries distribution automation services, Terminal 2 carries drone inspection services, and Terminal 3 carries precision load control services. Simultaneously, each base station entity is wiredly connected to its corresponding MEC server entity (e.g., Base Station 1 connects to MEC Server 1, Base Station 2 connects to MEC Server 2, and Base Station 3 connects to MEC Server 3). Furthermore, base stations are interconnected via wired links (e.g., Base Station 1 connects to Base Station 2, and Base Station 2 connects to Base Station 3). Each power service entity contains attribute nodes, such as network requirements (e.g., latency, bandwidth, reliability) and computing power requirements (e.g., CPU, memory). Each power terminal contains network resource attributes (e.g., downlink rate, uplink rate). Each base station contains network resource attributes (e.g., total wireless downlink bandwidth, total wireless uplink bandwidth). Each MEC server entity contains computing power resource attributes (e.g., total CPU usage, total memory usage). This knowledge graph allows for intuitive queries of the terminals and services carried by each base station within its coverage area, the quality of service requirements of each service, the available resources of each MEC server, and the allocation relationship between services and resources. This provides global and structured knowledge support for subsequent resource reallocation decisions.

[0074] In some embodiments, the server can visualize the constructed knowledge graph. Specifically, the Neo4j graph database functionality is embedded in the visualization management module. Figure 4 The knowledge graph data shown can be input into the neo4j graph database and then displayed intuitively on a webpage.

[0075] Step S206: Periodically monitor changes in the power service status and resource status within the dedicated power grid. The resource status includes the status of network resources and the status of computing resources.

[0076] The monitored changes in the status of power services include at least one of the following: new power service requests, changes in the attributes of existing services, and changes in service quality. Changes in service attributes further include changes in at least one of the following: network requirements, computing power requirements, etc.; changes in service quality include at least one of the following: changes in the latency level experienced by the service, changes in allocated bandwidth, and changes in reliability. Changes in the status of network resources include, but are not limited to, at least one of the following: network node reachability or unreachability, link load, etc. Changes in the status of computing power resources include, but are not limited to, at least one of the following: computing power node availability or unavailability, computing power load, etc.

[0077] For example, the server periodically monitors changes in the status of power services and resources within the dedicated power grid. Changes in the status of power services can be obtained through automatic collection of power service operation data (such as the addition, termination, or changes in service quality requirements) or through receiving manually input instructions (such as modifying service attributes) via the resource management window. Changes in the status of resources can be obtained through automatic monitoring of the real-time load and available capacity of network nodes (such as base stations and core networks) and computing nodes (such as edge servers) or through receiving manually input instructions (such as adding resources or migrating services) via the resource management window. Of course, the above monitoring methods are merely examples and do not constitute a limitation on the monitoring methods of this embodiment.

[0078] Step S208: Update the power private network knowledge graph based on changes in power business status and resource status.

[0079] For example, the server maintains a global graph database that stores all entities and relationships in the current power private network. When changes in power service status and resource status are detected, the server first compares the detected status change data with the current power private network knowledge graph. If the status change data is inconsistent with the graph data, the graph data is updated with the status change data, and the visualization interface is refreshed to display the latest graph.

[0080] For example, if the power service changes, the power service graph data in the current power private network knowledge graph is updated according to the power service change. Specifically, if a new power service is added, the network and computing power requirements of the new service, as well as the terminal carrying the new service, are extracted from the monitored status change data, and the network and computing power requirements of that terminal in the power private network knowledge graph are updated. If the terminal already carries other services, the network or computing power requirements of the terminal are the sum of the network or computing power requirements of the existing services and the new service. If the terminal does not carry other services, the network or computing power requirements of the terminal are the network or computing power requirements of the new service.

[0081] If network resources change, the network and node graph data of the power private network knowledge graph are updated accordingly. Specifically, if a new network node is added (including at least one of power terminals, network nodes, and computing nodes), the triplet data related to the new node is extracted from the monitored network resource status change data and updated in the current power private network knowledge graph. If a network node is removed, the vertices and edges related to the removed node are deleted from the current power private network knowledge graph. If a node's attributes change, the data related to the node's attributes is updated in the current power private network knowledge graph based on the node's identifier. If the network topology changes, the data related to the network topology is updated in the current power private network knowledge graph.

[0082] If computing resources change, the network and node graph data of the current power grid knowledge graph are updated accordingly. Specifically, if a new computing resource entity is added, triplet data related to the computing resource are extracted from the monitored computing resource status change data and updated in the current power grid knowledge graph. If a computing resource entity is reduced, the vertices and edges related to the reduced computing resource are deleted from the current power grid knowledge graph. If computing resource attributes change, the data related to the computing resource is updated in the current power grid knowledge graph.

[0083] Step S210: Select power services whose resource demands are not met from the latest power private network knowledge graph, and update the resource scheduling strategy corresponding to the power services.

[0084] Resource demand information refers to the resource demand thresholds agreed upon in the quality of service agreement for power services, including but not limited to at least one of the following: maximum tolerable end-to-end latency, minimum required bandwidth, minimum required reliability, required PU threshold, and memory capacity threshold. This information is typically pre-configured in the power service entities within the power private network knowledge graph. In this embodiment, the power service can be a newly added power service or an existing power service within the power private network. Resource scheduling strategy refers to the allocation rules and adjustment schemes for network resources and computing power resources within the power private network to meet the quality of service requirements of the power services.

[0085] For example, the server searches for services with unmet network resource requirements from the latest power private network knowledge graph, and also searches for services with unmet computing resource requirements from the latest power private network knowledge graph. Here, network resource requirements can be understood as the power service's demand for network resources; computing resource requirements can be understood as the power service's demand for computing resources. For these power services, resource scheduling optimization is performed to obtain corresponding resource scheduling strategies.

[0086] In one embodiment, resource requirements include network resource requirements; filtering power services whose resource requirements are not met from the latest power private network knowledge graph includes: for each power service in the latest power private network knowledge graph, obtaining the transmission delay, transmission bandwidth, and packet loss rate of the power service; and determining that the network resource requirements of the power service are not met if at least one of the following conditions is met: transmission delay is less than or equal to a delay threshold, transmission bandwidth is less than a bandwidth threshold, and packet loss rate is less than a packet loss threshold.

[0087] For example, the server can record the end-to-end network latency experienced by the power service i within a statistical time period T that includes several statistical time slots t. And calculate the average network latency value. :

[0088]

[0089] like This indicates that the latency requirement of service i has not been met. This indicates the maximum tolerable end-to-end delay.

[0090] In addition, the server can also determine bandwidth. Let the power terminal of service i be k, and its wireless uplink rate be... The set of services carried on the power terminal k is The data volume of business j is The bandwidth allocated to service i is:

[0091]

[0092] in, This represents the amount of data for business i. If... This indicates that the bandwidth requirement of service i has not been met. This indicates the minimum bandwidth required.

[0093] In addition, the server can also perform reliability assessments. The total number of data packets sent by service i within the statistical period T. The number of lost data packets is Then the packet loss rate for:

[0094]

[0095] reliability for: .like This indicates that the reliability requirements of service i have not been met. This indicates the minimum required reliability.

[0096] If at least one of the above determination processes fails, the network resource requirements of the power service are determined to be unmet. If all requirements are met, the network resource requirements of the power service are determined to be met.

[0097] In one embodiment, the resource requirements also include computing power resource requirements; filtering power services whose resource requirements are not met from the latest power private network knowledge graph includes: for each power service in the latest power private network knowledge graph, obtaining the computing processing capacity and memory of the power service; and determining that the computing power resource requirements of the power service are not met if at least one of the computing processing capacity is lower than the capacity threshold or the memory is lower than the memory threshold.

[0098] For example, for each power service in the latest power private network knowledge graph, the server obtains the required computing power and memory capacity thresholds from the service entity, and simultaneously obtains the currently allocated computing power and memory capacity for that service from the computing node entities associated with that service. If the actually allocated computing power is lower than the service's required computing power threshold, or the actually allocated memory capacity is lower than the service's required memory threshold, or both are lower than the thresholds, then the service's computing resource requirements are determined to be unmet. If the allocated computing power and memory for the service are both not lower than the service's required computing power threshold and memory threshold, then the service's computing resource requirements are confirmed to be met.

[0099] In this embodiment, by determining whether the network resource requirements and computing power resource requirements of the power service are met, resource scheduling is performed on the power service if they are not met, effectively avoiding invalid resource scheduling operations, thereby improving the accuracy of resource scheduling decisions and the efficiency of resource utilization.

[0100] In one embodiment, the resource scheduling strategy includes a network resource scheduling strategy for the base station area where the power terminal carrying the power service is located; updating the resource scheduling strategy corresponding to the power service includes: extracting a set of power terminals in the same base station area as the power service from the latest power private network knowledge graph; configuring the network resources of each power terminal according to the weight of the service carried by each power terminal in the power terminal set, so as to update the network resource scheduling strategy of the base station area.

[0101] In this context, the network resource scheduling strategy specifically refers to the wireless resource block allocation scheme among power terminals within a base station area. The power terminal set refers to the collection of all power terminals located within the same base station coverage area as the power service-bearing terminal; these terminals share the base station's wireless resource blocks. The service weight is a parameter used to determine the proportion of wireless resource blocks each terminal should receive; specifically, it equals the proportion of the sum of all power service data on that terminal to the total data volume of all terminals within the same base station area. Terminals with larger data volumes have higher weights and are allocated more resource blocks. In this embodiment, network resources refer to the wireless resource blocks provided by the base station. Each resource block is the smallest unit of wireless resource allocation in a fifth-generation mobile communication system, and its quantity determines the uplink or downlink transmission rate that the terminal can obtain.

[0102] For example, when the server determines that the network resource requirements of a certain power service are not being met, the resource scheduling strategy specifically includes adjusting the allocation of wireless resources in the base station area where the service is located. First, the server extracts the set of all power terminals located in the same base station area as the terminal carrying the service from the latest power private network knowledge graph, and obtains the total data volume of all services carried on each terminal. Let there be a total of [number missing] power terminals under base station b. There are 1 terminal, and the total amount of service data carried by each terminal u is... The total number of radio resource blocks for the base station is The number of radio resource blocks allocated to terminal u is then calculated according to a weighted ratio:

[0103]

[0104] in, This indicates rounding down. The formula means that wireless resource blocks are allocated fairly according to the proportion of each terminal's service data volume to the total data volume; the larger the proportion, the higher the weight. If there are remainders that cannot be divided evenly, they can be temporarily stored or redistributed using a round-robin method.

[0105] After completing the allocation of wireless resource blocks, the server can recalculate the wireless uplink rate of each terminal based on the number of newly allocated resource blocks. The calculation formula is as follows:

[0106]

[0107] in, Indicates the number of multiple input multiple output layers in the terminal. Indicates the modulation order. Indicates the diffusion factor. Indicates encoding efficiency. Indicates the percentage of protocol overhead. This represents the average duration of an orthogonal frequency division multiplexing (OFDM) symbol at a base station, in seconds. This formula, based on the physical layer transmission principle of fifth-generation mobile communication systems, converts the number of resource blocks into the actual transmission rate.

[0108] After allocation is completed, the server updates the attributes of relevant resource entities in the current power private network knowledge graph, such as the allocated amount and remaining amount, and generates specific resource allocation instructions, such as adjusting the base station resource block allocation table or modifying the resource limit of the edge computing container.

[0109] In this embodiment, by fairly allocating network resources according to weights, the service needs of high-load or high-priority services can be prioritized, thereby improving resource utilization efficiency.

[0110] In one embodiment, the resource scheduling strategy includes a computing power resource scheduling strategy related to power services. Updating the resource scheduling strategy corresponding to power services includes: extracting an edge server set from the latest power private network knowledge graph, and obtaining the available computing power resources corresponding to each edge server in the edge server set, as well as the transmission path between the edge server and the power services; determining the target weight corresponding to the edge server based on the available computing power resources and the transmission path; arranging each edge server in descending order according to the target weight to obtain a computing power resource queue; and sequentially allocating each computing power resource in the computing power resource queue to the power services to update the computing power resource scheduling strategy related to power services until the computing power resource requirements of the power services are met.

[0111] The computing resource scheduling strategy refers to the allocation scheme for computing resources such as the central processing unit (CPU) and memory of edge computing servers (MEC servers) to meet the computing needs of power services. This includes selecting target edge servers for services and allocating specific resource quotas. The edge server set refers to the collection of all available multi-access edge computing servers in the power private network, each with computing and storage capabilities. Available computing resources refer to the currently unused computing processing units and memory capacity of the edge servers, i.e., the total server capacity minus the portion allocated to other services. The transmission path refers to the network link between the power service carrying terminal and the edge server, including the base stations and core network nodes along the path; the equivalent bandwidth of the path reflects the data transmission capacity. The target weight is a comprehensive score used to evaluate the suitability of an edge server for carrying the service, considering both the server's available computing resources and the equivalent bandwidth of the transmission path from the service to the server. A higher weight indicates that the server is more suitable for priority allocation. The computing resource queue is a list of edge servers sorted from high to low according to their target weights; servers attempt to allocate resources to services sequentially in this order.

[0112] For example, when the server determines that the computing power resource requirements of a certain power service are not being met, the resource scheduling strategy specifically includes reselecting or adjusting the computing power allocation of edge servers for that service. First, the server extracts all available edge server sets from the latest power private network knowledge graph. For each edge server e, it obtains its currently available computing processing units. and available memory At the same time, obtain the equivalent bandwidth of the transmission path from the service to the edge server e. Then, the server calculates the target weight for each edge server. :

[0113]

[0114] in, , , These are preset weighting coefficients for computing processing units, memory, and bandwidth, which can be dynamically adjusted according to management strategies.

[0115] The server follows The edge servers are sorted in descending order of resource availability to form a computing resource queue. The server then iterates through each edge server in the queue. If the available computing resources of the current server can meet the needs of the service, the service is assigned to that edge server, and the computing resource scheduling policy is updated. If the current edge server's resources are insufficient, the server continues to try the next edge server in the queue until the computing resource requirements of the service are met. If the available resources of all edge servers are insufficient to meet the service requirements, the server can trigger a resource expansion alarm or adjust the service requirements, such as downgrading the service.

[0116] After allocation, the server updates the available resource attributes of the relevant edge servers in the power private network knowledge graph and records the allocation relationship between power services and edge servers.

[0117] In this embodiment, edge servers with sufficient computing power and good network conditions are selected first by comprehensive weight ranking, so as to ensure the end-to-end service quality of the business while ensuring resource utilization.

[0118] In one embodiment, the change in power service status includes power service migration; updating the resource scheduling strategy corresponding to the power service includes: releasing the resources currently allocated to the migrated power service; identifying target power terminals corresponding to other power services located in the same base station area as the migrated power service and whose resource requirements have not been met; allocating the released resources to the target power terminals according to the service weights corresponding to the target power terminals; determining the target base station area after the migration of the power service, and reallocating resources according to the service weights of power terminals already carrying services in the target base station area and the service weights of power terminals carrying the migrated power service.

[0119] Power service migration refers to the process of moving a specific power service (referred to as the migrated service) from its current source base station area to another target base station area. Migration can be triggered manually, for example, by maintenance personnel entering a migration command through the resource management window. Releasing resources refers to releasing the radio resource blocks (such as the number of resource blocks) originally allocated to the terminal where the migrated service resides in the source base station area, making them available for reallocation. Other power services whose resource requirements are not met refer to power services within the source base station area, excluding the migrated service, whose current actual resource availability is lower than their service quality requirements. The target base station area refers to the coverage area of ​​the base station to which the migrated service is planned to be moved. After migration, the service will share the radio resources of that base station with existing power services within the target area.

[0120] For example, when the power service status changes to service migration, the server performs a two-stage resource scheduling policy update at both the source and target ends. At the source end, the server first retrieves the source base station and its bearer terminal currently located in the migrated service from the latest power private network knowledge graph, and releases all radio resource blocks occupied by that terminal. Let the number of resource blocks released be... Then, the server queries the knowledge graph for the set of terminals corresponding to other power services that are in the same source base station area as the migration service and whose resource requirements are not being met. For each terminal u, obtain the total amount of service data it carries. The released resource blocks are allocated to each terminal according to the weight of the services they carry, using the following allocation formula:

[0121]

[0122] in, This indicates the number of additional resource blocks acquired by terminal u. This represents the set of terminals corresponding to all services other than migration services on the source resource entity.

[0123] If there are still remaining resource blocks after allocation, they can be temporarily stored or allocated using a round-robin method. Through this step, the released resources are prioritized for alleviating insufficient demand from other services in the source base station area.

[0124] On the target side, the server determines the target base station area after the migration service, and retrieves the set of all power terminals under that target base station from the knowledge graph. This includes the migrated service bearer terminals (newly added) and the existing terminals of the target base station. Let the total number of radio resource blocks of the target base station be... Each terminal The total amount of business data carried is The server reallocates all resource blocks according to the service weight of each terminal:

[0125]

[0126] For example, if the resource management window detects the input "Move drone inspection 3 to base station 1", then... Figure 4As shown, the drone inspection service 3 is currently running on power terminal 5. According to the instruction, this service and the terminal will be migrated from base station 2 to base station 1. The radio resources allocated to power terminal 5 by base station 2 will be reclaimed and reallocated to power terminals 4 and 6 based on their resource requirements. Base station 1 will then add power terminal 5 and the drone inspection service 3 it carries. The radio resources of base station 1 will be reallocated according to the weights of the services carried by power terminals 1, 2, 3, and 5.

[0127] After allocation, the server updates the knowledge graph, deletes the connection relationship between the source base station and the migration service bearer terminal, establishes the connection relationship between the target base station and the terminal, and updates the resource usage attributes of all affected terminals.

[0128] In this embodiment, by reallocating resources according to weights between the migrated business and the existing business on the target resource entity during business migration, the service quality degradation caused by the new business crowding out the original resources is avoided. Thus, while maintaining fairness among businesses, a smooth transition and efficient utilization of resources are achieved.

[0129] In one embodiment, the resource status change includes adding resources; updating the resource scheduling strategy corresponding to the power service includes: determining the base station area to which the power service belongs, and when there is a closed power service in the base station area, releasing the resources allocated to the closed power service to obtain the added resources; and allocating the added resources to the power service according to the carrying service weight corresponding to the power terminal carrying the power service.

[0130] In this embodiment, newly added resources refer to the network or computing resources previously occupied by a power service being released and becoming available for reallocation after that service is shut down. Resource additions can also originate from node expansion. Shutdown power services refer to those services terminated by maintenance personnel through resource management windows or automatic policies; after shutdown, the service and its bearer terminals no longer require resources. Bearer service weight refers to the proportion of the sum of data volume of all power services on a terminal to the total data volume of all terminals within the same base station area, used to determine the allocation ratio of newly added resources among the terminals.

[0131] For example, when a resource status changes to "new resource," and this new resource originates from the shutdown of a certain power service, the server executes the following resource scheduling policy update. First, the server locates the shut-down power service and its bearer terminal from the knowledge graph, determining the base station area to which the terminal belongs. Then, the server releases all radio resource blocks occupied by the terminal; these released resources become the new resources, and their quantity is denoted as [number missing]. Next, the server determines whether there are other unmet power service requests within the base station area. If so, the server retrieves the set of terminals corresponding to all unmet power service requests within the base station area from the knowledge graph. And calculate the total amount of service data carried by each terminal u. And the total data volume of all terminals with insufficient demand. The server allocates the newly released resources to each terminal according to the weight of the services they carry, using the following formula:

[0132]

[0133]

[0134] in, This indicates the number of additional resource blocks acquired by terminal u. If there are still resources remaining after allocating them to all terminals with insufficient demand, the remaining resources can be kept idle or allocated to other terminals in a round-robin manner.

[0135] For example, if the system detects that the command "Close Distribution Automation Service 3" is entered in the resource management window, and the command contains the entity "Distribution Automation Service 3" from the graph, then... Figure 4 As shown, distribution automation service 3 is connected to power terminal 7 in the map. When the service is shut down, power terminal 7 no longer carries any service, and the radio resource blocks allocated by base station 3 to power terminal 7 will be released and reclaimed. Further, it is determined whether the radio resources currently allocated by base station 3 to power terminals 8 and 9 in the area can meet the network resource requirements for the terminals to carry services. If yes, the current resource allocation strategy remains unchanged. If not, the reclaimed radio resource blocks are fairly allocated to power terminals 8 and 9 according to the weight of the services they carry.

[0136] If the maintenance personnel input an attribute modification command, such as "Drone inspection 3 network bandwidth not less than 30Mbps", the server compares the current actual bandwidth with the requirement. If the requirement is met, the server only updates the bandwidth requirement attribute of the service in the knowledge graph and does not trigger resource scheduling.

[0137] After allocation is completed, the server updates the resource occupancy attributes of the corresponding terminals in the knowledge graph and generates a new base station resource scheduling strategy, such as the updated resource block allocation table.

[0138] Step S212: The resource scheduling strategy is distributed to the network nodes corresponding to the power business for execution.

[0139] For example, after generating the updated resource scheduling policy, the server enters the policy distribution and execution phase. If the policy is a base station radio resource block allocation scheme, the server converts the policy into instructions that the base station scheduler can recognize. After receiving the instructions, the base station allocates uplink or downlink radio resource blocks to each terminal according to the new allocation scheme in the next scheduling cycle. If the policy is a computing resource allocation scheme, the server calls the management interface of the edge computing platform to create or update the resource quotas of containers or virtual machines. After receiving the instructions, the edge server immediately adjusts the resource allocation. For service migration policies, the distribution process is divided into two steps: first, a resource release instruction is sent to the source node, and then a resource allocation instruction is sent to the target node. After all instructions are executed, the server collects the execution results, i.e., success or failure, and updates the actual resource occupancy status in the knowledge graph according to the execution results. If the execution fails, the server can trigger a retry or alarm mechanism.

[0140] In this embodiment, power service information, network information, and computing power information of the dedicated power grid are first collected to construct a knowledge graph of the dedicated power grid. Next, changes in the status of power services and resources within the dedicated power grid are periodically monitored; the resource status includes the status of network resources and computing power resources. The knowledge graph of the dedicated power grid is updated based on these changes. Power services whose resource demands are not being met are selected from the latest knowledge graph, and the corresponding resource scheduling strategies are updated. Finally, the resource scheduling strategies are distributed to the network nodes corresponding to the power services for execution. The introduction of the knowledge graph of the dedicated power grid not only integrates the dispersed power services and resource supply into a unified structured model but also significantly improves the responsiveness and accuracy of resource scheduling decisions to dynamic changes in the dedicated power grid. Therefore, resource allocation no longer relies on static manual configuration but adaptively adjusts based on actual changes in the status of power services and resources within the dedicated power grid, significantly improving the overall resource utilization efficiency.

[0141] In a specific embodiment, such as Figure 5 As shown, the resource management method for private power grids provided in this embodiment runs in a resource management system, which includes a resource management module and a data acquisition and policy distribution module. The resource management module serves as the core decision-making and execution entity, while the data acquisition and policy distribution module is responsible for underlying data interaction and instruction conversion.

[0142] The dedicated power network is specifically a 5G power virtual private network (VPN), connected to the resource management system. This 5G power VPN includes power IoT terminals, 5G base stations, and MEC servers. Base stations and terminals are connected wirelessly, while base stations and MEC servers, base stations and the 5G core network, and the core network and cloud computing center are connected via wired links. The 5G core network is the central network of the fifth-generation mobile communication system, responsible for terminal access control, mobility management, session management, and data routing and forwarding. The cloud computing center is a centralized data center deployed remotely, used to handle non-real-time or big data computing tasks, complementing edge computing.

[0143] The data acquisition and policy distribution module communicates with the 5G power virtual private network and the resource management module respectively. It is responsible for collecting power business information, network information and computing power information within the private network and forwarding them to the resource management module.

[0144] The resource management module comprises a visualization unit, a knowledge graph management unit, and another resource management unit. The visualization unit provides data visualization and a resource management window, allowing operations personnel to input manual management commands such as service migration, service shutdown, and resource addition. The knowledge graph management unit constructs and dynamically updates a knowledge graph based on collected information. This graph describes five core entities—power services, network, nodes, resources, and network topology—and the relationships between them using a graph structure. The resource management unit updates the power service resource scheduling strategy according to the latest power private network knowledge graph and converts the strategy into device-recognizable commands through the data acquisition and strategy distribution module, then distributes these commands to base stations or MEC servers for execution, thereby achieving adaptive resource management of the 5G power virtual private network.

[0145] When the system determines that the network resource demand of power services is not met, it reallocates wireless resource blocks according to the service weights of each terminal within the same base station area. When it determines that the computing resource demand is not met, it allocates computing resources in descending order according to the comprehensive weight of edge servers. When a service migration event is detected, resources are simultaneously reclaimed and reallocated at the source end, and all resources are reallocated according to weight at the target end. When new resources are detected (e.g., resources are released when services are closed), the released new resources are allocated according to weight to other services in the same area with insufficient demand. The above steps can be used individually or in combination as needed. For example, during service migration, the reallocation of resources after reclamation at the source end can adopt the network resource reallocation method, and the resource allocation at the target end can also adopt the weight allocation logic of network resources or computing resources. Through such integration, the system can flexibly invoke corresponding scheduling strategy update methods for different resource types, different resource change states, and different power service change states, realizing adaptive management of power private network resources.

[0146] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0147] Based on the same inventive concept, this application also provides a resource management device for power private networks for implementing the resource management method for power private networks described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the resource management device for power private networks provided below can be found in the limitations of the resource management method for power private networks described above, and will not be repeated here.

[0148] In one exemplary embodiment, such as Figure 6 As shown, a resource management device for a private power grid is provided, comprising:

[0149] The information acquisition module 602 is used to collect power business information, network information, and computing power information of the power private network;

[0150] The knowledge graph construction and update module 604 is used to construct a knowledge graph of the power private network based on power business information, network information and computing power information.

[0151] The monitoring module 606 is used to periodically monitor changes in the status of power services and resources in the private power grid; the resource status includes the status of network resources and computing resources.

[0152] The knowledge graph construction and updating module 604 is also used to update the power private network knowledge graph according to changes in power business status and resource status.

[0153] The strategy update module 608 is used to filter out power services whose resource demands are not being met from the latest power private network knowledge graph and update the resource scheduling strategy corresponding to the power services.

[0154] The policy update module 608 is also used to distribute resource scheduling policies to the network nodes corresponding to the power business for execution.

[0155] In one embodiment, the resource requirements include network resource requirements; the policy update module 608 is further configured to: for each power service in the latest power private network knowledge graph, obtain the transmission delay, transmission bandwidth and packet loss rate of the power service; and determine that the network resource requirements of the power service are not met if at least one of the following conditions is met: the transmission delay is less than or equal to the delay threshold, the transmission bandwidth is less than the bandwidth threshold, and the packet loss rate is less than the packet loss threshold.

[0156] In one embodiment, the resource requirements also include computing power resource requirements; the policy update module 608 is further configured to: obtain the computing power and memory of each power service in the latest power private network knowledge graph; and determine that the computing power resource requirements of the power service are not met if at least one of the computing power is lower than the capability threshold or the memory is lower than the memory threshold.

[0157] In one embodiment, the resource scheduling strategy includes a network resource scheduling strategy for the base station area where the power terminal carrying the power service is located; the strategy update module 608 is further configured to: extract a set of power terminals in the same base station area as the power service from the latest power private network knowledge graph; and configure the network resources of each power terminal according to the corresponding service weight of each power terminal in the power terminal set, so as to update the network resource scheduling strategy of the base station area.

[0158] In one embodiment, the resource scheduling strategy includes a computing power resource scheduling strategy related to power services; the strategy update module 608 is further configured to: extract an edge server set from the latest power private network knowledge graph, and obtain the available computing power resources corresponding to each edge server in the edge server set, as well as the transmission path between the edge server and the power services; determine the target weight corresponding to the edge server based on the available computing power resources and the transmission path; arrange the edge servers in descending order according to the target weight to obtain a computing power resource queue; and allocate each computing power resource in the computing power resource queue to the power services in sequence to update the computing power resource scheduling strategy related to the power services until the computing power resource requirements of the power services are met.

[0159] In one embodiment, the change in power service status includes power service migration; the policy update module 608 is further configured to: release the resources currently allocated to the migrated power service; identify the target power terminal corresponding to other power services that are in the same base station area as the migrated power service and whose resource requirements have not been met; allocate the released resources to the target power terminal according to the bearer service weight corresponding to the target power terminal; determine the target base station area after the migration of the power service, and reallocate resources according to the bearer service weight of the power terminal already carrying the service in the target base station area and the bearer service weight of the power terminal carrying the migrated power service.

[0160] In one embodiment, the resource status change includes the addition of new resources; the policy update module 608 is further configured to: determine the base station area to which the power service belongs, and when there is a closed power service in the base station area, release the resources allocated to the closed power service to obtain the new resources; and allocate the new resources to the power service according to the carrying service weight corresponding to the power terminal carrying the power service.

[0161] The modules in the aforementioned resource management device for dedicated power grids can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0162] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores network resource management data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a resource management method for a dedicated power grid.

[0163] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a resource management method for a dedicated power grid.

[0164] Those skilled in the art will understand that Figure 7 or Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0165] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0166] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program being executed by a processor to perform the steps in the above method embodiments.

[0167] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the steps described in the above method embodiments.

[0168] 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 used for analysis, data stored, data displayed, 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 the relevant data must comply with relevant regulations.

[0169] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media 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), magnetic 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 take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0170] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0171] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A resource management method for private power grids, characterized in that, The method includes: Collect power business information, network information, and computing power information from the dedicated power grid; A knowledge graph of the private power network is constructed based on the power business information, the network information, and the computing power information. Periodically monitor changes in the power service status and resource status within the dedicated power grid; wherein, the resource status includes the status of network resources and the status of computing resources; The power private network knowledge graph is updated based on the changes in the power service status and the changes in the resource status. The system filters out power services whose resource demands are not being met from the latest power private network knowledge graph and updates the resource scheduling strategy corresponding to the power services. The resource scheduling strategy is distributed to the network nodes corresponding to the power service for execution.

2. The method according to claim 1, characterized in that, The resource requirements include network resource requirements; the process of filtering power services whose resource requirements are not met from the latest power private network knowledge graph includes: For each power service in the latest power private network knowledge graph, obtain the transmission delay, transmission bandwidth, and packet loss rate of the power service; If at least one of the following conditions is met: the transmission delay is less than or equal to a delay threshold, the transmission bandwidth is less than a bandwidth threshold, or the packet loss rate is less than a packet loss threshold, then the network resource requirements of the power service are determined to be unmet.

3. The method according to claim 1, characterized in that, The resource requirements also include computing power resource requirements; the process of filtering power services whose resource requirements are not met from the latest power private network knowledge graph includes: For each power service in the latest power private network knowledge graph, obtain the computing power and memory of the power service; If at least one of the following conditions is met: the computing power is below the capacity threshold or the memory is below the memory threshold, it is determined that the computing power resource requirements of the power service are not being met.

4. The method according to claim 1, characterized in that, The resource scheduling strategy includes the network resource scheduling strategy for the base station area where the power terminal carrying the power service is located. The updating of the resource scheduling strategy corresponding to the power service includes: Extract the set of power terminals that are in the same base station area as the power service from the latest power private network knowledge graph; According to the respective service weights of each power terminal in the power terminal set, the network resources of each power terminal are configured to update the network resource scheduling strategy of the base station area.

5. The method according to claim 1, characterized in that, The resource scheduling strategy includes the computing resource scheduling strategy related to the power service; updating the resource scheduling strategy corresponding to the power service includes: Extract the set of edge servers from the latest power private network knowledge graph, and obtain the available computing power resources corresponding to each edge server in the set, as well as the transmission path between the edge server and the power service. Based on the available computing resources and the transmission path, the target weight corresponding to the edge server is determined; According to the target weight, the edge servers are arranged in descending order to obtain the computing power resource queue; The computing resources in the computing resource queue are allocated to the power service in sequence to update the computing resource scheduling strategy related to the power service until the computing resource requirements of the power service are met.

6. The method according to claim 1, characterized in that, The changes in the status of the power service include power service migration; updating the resource scheduling strategy corresponding to the power service includes: Release the resources currently allocated to the migrating power service; Identify the target power terminals corresponding to other power services that are located in the same base station area as the migrated power service and whose resource requirements have not been met; The released resources are allocated to the target power terminal according to the service weight corresponding to the target power terminal; The target base station area after the migration of the power service is determined, and resources are reallocated according to the service carrying weight of the power terminals already carrying the service in the target base station area and the service carrying weight of the power terminals carrying the migrated power service.

7. The method according to claim 1, characterized in that, The changes in resource status include the addition of new resources; The updating of the resource scheduling strategy corresponding to the power service includes: Determine the base station area to which the power service belongs. If there is a closed power service in the base station area, release the resources allocated to the closed power service to obtain the new resources. The new resources are allocated to the power service according to the service weight corresponding to the power terminal carrying the power service.

8. A resource management device for private power grids, characterized in that, The device includes: The information acquisition module is used to collect power business information, network information, and computing power information from the dedicated power grid. The knowledge graph construction and update module is used to construct a power private network knowledge graph based on the power business information, the network information, and the computing power information. The monitoring module is used to periodically monitor changes in the power service status and resource status in the dedicated power grid; wherein, the resource status includes the status of network resources and the status of computing resources; The knowledge graph construction and update module is also used to update the power private network knowledge graph according to the changes in the power business status and the changes in the resource status; The strategy update module is used to filter out power services whose resource demands are not being met from the latest power private network knowledge graph, and update the resource scheduling strategy corresponding to the power service. The policy update module is also used to distribute the resource scheduling policy to the network nodes corresponding to the power service for execution.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

Images