Resource management method and device, equipment, storage medium and program product

Abstract

The invention relates to a resource management method and device, equipment, a storage medium and a program product. The method comprises the following steps: in response to a slice creation request of a target power terminal, determining an optimal resource node group of the target power terminal according to node information of each resource node in a power system and communication information between each resource node and the target power terminal; creating a new slice network in the network of the power system by adopting the resources of the optimal resource node group, and accessing the target power terminal to the new slice network; in the operation process of the new slice network, predicting resource demand change information of the target power terminal according to the monitoring state data of the target power terminal; and adjusting the resource node of the target power terminal according to the resource demand change information to complete resource management of the target power terminal. By adopting the method, the flexibility of resource scheduling can be improved.

Description

Technical Field

[0001] This application relates to the technical field of resource management, and in particular, to a resource management method, apparatus, device, storage medium, and program product. Background Art

[0002] The current power system faces many problems such as low resource utilization efficiency and high communication latency in dealing with the increasing data processing requirements and real-time computing tasks.

[0003] Network slicing is an emerging technology that supports different application requirements by creating multiple virtual networks on a shared physical network. In related technologies, in order to improve the flexibility and isolation of resources, network slicing technology is usually used in power systems. For example, network slicing technology is combined with an edge computing architecture for resource management and task processing.

[0004] However, the resource management method in related technologies has the technical problem of inflexible resource scheduling. Summary of the Invention

[0005] Based on this, it is necessary to provide a resource management method, apparatus, device, storage medium, and program product for the above technical problems, which can improve the flexibility of resource scheduling.

[0006] In a first aspect, an embodiment of this application provides a resource management method, including:

[0007] In response to a slice creation request of a target power terminal, determine an optimal resource node group for the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal;

[0008] Use the resources of the optimal resource node group to create a new slice network in the network of the power system, and connect the target power terminal to the new slice network;

[0009] During the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitored status data of the target power terminal;

[0010] Adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

[0011] In one embodiment, the communication information includes communication latency; determining an optimal resource node group for the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal includes:

[0012] Randomly generate a group of candidate solutions among the resource nodes, and each candidate solution represents a candidate resource node group;

[0013] Determine the fitness value of each candidate solution according to the node information of each resource node and the communication delay between each resource node and the target power terminal;

[0014] Determine the optimal resource node group of the target power terminal according to the fitness value of each candidate solution.

[0015] In one embodiment, determining the optimal resource node group of the target power terminal according to the fitness value of each candidate solution includes:

[0016] Determine the candidate solutions whose fitness values meet the preset conditions according to the fitness values of each candidate solution, and perform crossover and mutation on the candidate solutions that meet the conditions to generate new solutions;

[0017] Iteratively execute the step of generating new solutions until the preset iteration stop condition is met, and use the new solution with the highest fitness value as the optimal resource node group of the target power terminal.

[0018] In one embodiment, predicting the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal includes:

[0019] Input the monitoring status data of the target power terminal into a pre-trained resource demand prediction model to obtain the resource demand change information of the target power terminal output by the resource demand prediction model; the resource demand prediction model is trained using the monitoring status data and resource quantity data of multiple sample power terminals at different time periods.

[0020] In one embodiment, the resource demand change information includes a decrease in the resource quantity; according to the resource demand change information, adjust the resource nodes of the target power terminal to complete the resource management of the target power terminal, including:

[0021] Obtain the load rate and energy efficiency ratio of each optimal resource node in the optimal resource node group, and determine the resource status quantization value of each optimal resource node;

[0022] Adjust the resource nodes of the target power terminal according to each resource status quantization value to achieve the resource management of the target power terminal.

[0023] In one embodiment, adjusting the resource nodes of the target power terminal according to each resource status quantization value includes:

[0024] Determine the node priority of each optimal resource node according to each resource status quantization value;

[0025] Migrate the optimal resource node with the lowest node priority from the optimal resource node group according to each node priority to complete the resource management of the target power terminal.

[0026] In a second aspect, an embodiment of the present application further provides a resource management device, including:

[0027] A node group determination module, configured to, in response to a slice creation request of a target power terminal, determine an optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal;

[0028] A slice network creation module, configured to use the resources of the optimal resource node group to create a new slice network in the network of the power system and connect the target power terminal to the new slice network;

[0029] A resource demand prediction module, configured to, during the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitored status data of the target power terminal;

[0030] A resource adjustment module, configured to adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

[0031] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the method in any one of the embodiments of the first aspect are implemented.

[0032] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method in any one of the embodiments of the first aspect are implemented.

[0033] In a fifth aspect, the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the method in any one of the embodiments of the first aspect are implemented.

[0034] The resource management method, device, equipment, storage medium, and program product provided by the embodiments of the present application. The resource management method provided by the embodiments of the present application responds to a slice creation request of a target power terminal, determines an optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal, and then uses the resources of the optimal resource node group to create a new slice network in the network of the power system, and connects the target power terminal to the new slice network. After that, during the operation of the new slice network, according to the monitored status data of the target power terminal, the resource demand change information of the target power terminal is predicted, and finally, according to the resource demand change information, the resource nodes of the target power terminal are adjusted to complete the resource management of the target power terminal. In this method, after receiving the slice creation request of the target power interruption, the optimal resource node group of the target power terminal is screened from multiple resource nodes in the power system, and the resources of the optimal resource node group are used to create a new slice network. During the operation of the new slice network, based on the predicted resource demand information of the target power terminal, the resource nodes of the target power terminal are adjusted to realize the resource management of the target power terminal. Equivalently, it always responds to the slice creation demand of the target power terminal, dynamically adjusts the resources of the target power terminal according to the real-time slice creation demand, and further adjusts the resources through the predicted resource demand information, thereby improving the scheduling flexibility of the resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0036] Figure 1 It is the internal structure diagram of a computer device in an embodiment;

[0037] Figure 2 It is the flowchart of the resource management method in an embodiment;

[0038] Figure 3 It is the flowchart of determining the optimal resource node group in an embodiment;

[0039] Figure 4 It is the flowchart of adjusting the resource nodes of the target power terminal in an embodiment;

[0040] Figure 5 It is the architecture diagram of a dynamic resource management system in an embodiment;

[0041] Figure 6 Schematic structural diagram of a resource management device in an embodiment. Detailed implementation manners

[0042] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0043] The technical background of the embodiments of the present application will be described below first.

[0044] The current power system faces many problems such as low resource utilization efficiency, high communication latency, and low integration of network slicing and edge computing technologies in dealing with the increasing data processing requirements and real-time computing tasks. Existing technologies such as traditional network slicing technologies and edge computing architectures, although improving the resource management and task processing capabilities in some aspects, still have many deficiencies. Traditional network slicing technologies are usually pre-configured and lack dynamic response capabilities, and cannot adapt to the changing computing task requirements in the power grid. The existing edge computing architectures are not yet mature in combination with network slicing, resulting in inflexible resource scheduling and inability to fully respond to the dynamic load requirements of the power system. In addition, the complexity and lag of resource management also make it difficult for existing solutions to quickly respond to load changes, affecting the overall performance and service quality of the system.

[0045] Although network slicing technology can improve the flexibility and isolation of resources, there are still some limitations in the application of existing network slicing technology in the power system. First, existing slicing networks are usually pre-configured and cannot be dynamically adjusted according to real-time requirements, failing to meet the diverse computing task requirements in the power grid. Second, the service types of slicing networks are usually fixed and it is difficult to adapt to the diverse application scenarios in the power grid, such as different load forecasting, intelligent scheduling, energy management, etc. The computing and storage resources in the power grid are usually distributed in different regions and nodes, and traditional resource allocation methods are difficult to dynamically adjust, resulting in low resource utilization efficiency. In some cases, the computing power of some nodes is over-consumed while the resources of other nodes are idle. In addition, the task requirements of the power grid are dynamic, and the static allocation method of resources is difficult to cope with load changes, leading to waste of overall resources. The tasks of the power system usually have high requirements for real-time performance, such as power monitoring, fault detection, load balancing, etc. These tasks need to be completed within strict time constraints. Therefore, communication delay becomes a key factor affecting system performance. However, due to the distributed nature of resources, the physical distance between power terminals and computing nodes may be relatively long, resulting in increased communication delay and affecting the timely processing of tasks. The load demand in the power system is highly dynamic, including both daily load fluctuations and large-scale demand changes under emergencies. Existing network architectures and resource allocation mechanisms are difficult to quickly respond to these dynamic demands, often resulting in a decline in service quality or waste of resources. In addition, the terminal devices and application types in the power grid are diverse, requiring the network to be able to flexibly adapt to different computing task requirements, while existing solutions cannot effectively meet this requirement.

[0046] Based on this, the embodiment of the present application provides a resource management method. After receiving a slice creation request for a target power outage, the optimal resource node group of the target power terminal is screened from multiple resource nodes in the power system, and the resources of the optimal resource node group are used to create a new slice network. During the operation of the new slice network, based on the predicted resource demand information of the target power terminal, the resource nodes of the target power terminal are adjusted to achieve resource management of the target power terminal. That is, it responds to the slice creation demand of the target power terminal at all times, dynamically adjusts the resources of the target power terminal according to the real-time slice creation demand, and further adjusts the resources through the predicted resource demand information, thereby improving the scheduling flexibility of resources. Of course, the technical solution provided in the embodiment of the present application is not limited to only solving the above problems, and there are other technical effects. For specific details, please refer to the following embodiments.

[0047] It should be noted that the beneficial effects or technical problems solved by the embodiment of the present application are not limited to this one, and there may be other implicit or related problems. For specific details, please refer to the description of the following embodiments.

[0048] The resource management method provided by the embodiments of the present application can be applied to a computer device. The computer device can be a server, and its internal structure diagram can be as shown in Figure 1 . The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a resource management method. Those skilled in the art can understand that Figure 1 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component layout.

[0049] The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0050] In an exemplary embodiment, as shown in Figure 2 , a resource management method is provided. Taking the computer device in as an example, the method includes the following steps 201 to step 204. Among them: Figure 1 S201, in response to a slice creation request of a target power terminal, determine an optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal.

[0051]

[0052] ​The slice network creation request refers to a request sent by a target power terminal to a network infrastructure for creating a specific network slice in the application scenario of network slicing technology. The node information of a resource node includes the computing power of the resource node, the current load rate of the resource node, the energy efficiency ratio of the resource node, etc. The communication information between each resource node and the target power terminal includes information such as the communication delay between each resource node and the target power terminal.

[0053] Exemplarily, the method for determining the optimal resource node group of the target power terminal can be to determine the resource demand matching degree between each resource node and the target power terminal based on the node information of each resource node and the communication information between each resource node and the target power terminal, and then, based on the resource demand matching degree between each resource node and the target power terminal, screen out multiple resource nodes from each resource node to form a group, so as to obtain the optimal resource node group of the target power terminal.

[0054] Among them, the method for screening nodes from each resource node can be to sort each resource node based on the resource demand matching degree between each resource node and the target power terminal, and then select multiple resource nodes with higher rankings from the sorted resource nodes, and form a group with the multiple resource nodes to obtain the optimal resource node group of the target power terminal.

[0055] S202, use the resources of the optimal resource node group to create a new slice network in the network of the power system, and connect the target power terminal to the new slice network.

[0056] After screening out the optimal resource node group of the target power terminal, use the resources of the optimal resource node group to create a slice network, so that the target power terminal can operate in the newly created slice network.

[0057] Exemplarily, use the resources of the optimal resource node group to create a new slice network in the network of the power system, and after the new slice network is created, notify the target power terminal to connect the target power terminal to the newly created slice network.

[0058] S203, during the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal.

[0059] The monitoring status data includes real-time status data such as voltage, current, active power, reactive power, frequency, power factor, equipment temperature, and operation duration.

[0060] After obtaining the monitoring data, data cleaning and preprocessing are performed on the monitoring status data, including missing value handling, outlier detection, and data normalization / standardization. Among them, missing value handling can be achieved through interpolation methods (linear interpolation, time series interpolation), moving window mean / median filling, or model-based prediction of missing values.

[0061] Furthermore, the processed monitoring status data is used to train a prediction model, and the trained prediction model is used to process the monitoring status data of the target power terminal, thereby obtaining the resource demand change information of the target power terminal.

[0062] S204. According to the resource demand change information, the resource nodes of the target power terminal are adjusted to complete the resource management of the target power terminal.

[0063] After obtaining the resource demand change information, the resource nodes of the target power terminal are adjusted so that the adjusted resource nodes can match the current resource demand.

[0064] For example, when the resource demand change information indicates a decrease in the resource quantity, the resource nodes of the target power terminal can be reduced. At this time, a preset number of resource nodes can be deleted from the optimal resource node group corresponding to the target power terminal; when the resource demand change information indicates an increase in the resource quantity, the resource nodes of the target power terminal can be increased. At this time, a preset number of resource nodes can be added to the optimal resource node group corresponding to the target power terminal.

[0065] The resource management method provided by the embodiments of the present application, in response to a slice creation request of a target power terminal, determines an optimal resource node group for the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal. Then, it uses the resources of the optimal resource node group to create a new slice network in the network of the power system, and connects the target power terminal to the new slice network. After that, during the operation of the new slice network, it predicts the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal. Finally, according to the resource demand change information, it adjusts the resource nodes of the target power terminal to complete the resource management of the target power terminal. In this method, after receiving the slice creation request of the target power interruption, it screens the optimal resource node group of the target power terminal from multiple resource nodes in the power system, uses the resources of the optimal resource node group to create a new slice network, and during the operation of the new slice network, based on the predicted resource demand information of the target power terminal, it adjusts the resource nodes of the target power terminal to realize the resource management of the target power terminal. Equivalently, it always responds to the slice creation demand of the target power terminal, dynamically adjusts the resources of the target power terminal according to the real-time slice creation demand, and further adjusts the resources through the predicted resource demand information, thereby improving the scheduling flexibility of the resources.

[0066] Based on the above embodiments, an embodiment is provided to illustrate the process of determining the optimal resource node group.

[0067] In an exemplary embodiment, as Figure 3 shown, the communication information includes communication delay; determining the optimal resource node group for the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal includes:

[0068] S301, randomly generate a group of candidate solutions among the resource nodes, and each candidate solution represents a candidate resource node group.

[0069] First, generate an initial population, and randomly generate a group of candidate solutions among all the resource nodes in the power system. Each candidate solution represents a possible resource node combination.

[0070] S302, determine the fitness value of each candidate solution according to the node information of each resource node and the communication delay between each resource node and the target power terminal.

[0071] In the embodiments of the present application, based on information such as computing resource utilization rate and communication delay, calculate the fitness value Fitness( ) of each candidate solution in the population. Among them, the calculation formula for the fitness value of the candidate solution is:

[0072]

[0073] Among them, is the weight of the computing resource utilization rate; is the weight of the communication delay; is the weight of the energy efficiency ratio; is the weight of the load balancing; represents the j-th resource node and belongs to the candidate solution ; represents the resource node 's computing power (such as the number of CPU cores or the computing power unit); represents the resource node 's current load rate (resource utilization rate), ranging from [0, 1]; represents the communication delay between the resource node and the power terminal, usually in milliseconds (ms); the resource node 's energy efficiency ratio, representing the energy consumption efficiency of the node, usually expressed in computing unit power consumption (such as FLOPS / W per watt of computing power); is the candidate solution 's maximum load rate of all resource nodes; is the candidate solution 's minimum load rate of all resource nodes.

[0074] S303. Determine the optimal resource node group of the target power terminal according to the fitness values of each candidate solution.

[0075] Exemplarily, according to the fitness values of each candidate solution, determine the candidate solutions whose fitness values meet the preset conditions, and perform crossover and mutation on the candidate solutions that meet the conditions to generate new solutions; iteratively execute the steps of generating new solutions until the preset iteration stop condition is met, and use the new solution with the highest fitness value as the optimal resource node group of the target power terminal.

[0076] In the embodiments of the present application, selection is performed from the current population using the tournament selection method to select candidate solutions with higher fitness values, and then crossover (such as single-point crossover) and mutation (such as a mutation probability of 0.01) are performed on the selected candidate solutions to generate new solutions. Then, the above steps are iteratively executed until the predetermined number of iterations or fitness convergence is met, and finally, the candidate solution with the highest fitness value is selected as the optimal resource node combination, so as to obtain the optimal resource node group of the target power terminal.

[0077] The resource management method provided by the embodiments of the present application first randomly generates a set of candidate solutions in each resource node, where each candidate solution represents a candidate resource node group. Then, according to the node information of each resource node and the communication delay between each resource node and the target power terminal, the fitness value of each candidate solution is determined. After that, according to the fitness values of each candidate solution, the optimal resource node group of the target power terminal is determined. In this method, when screening the optimal resource nodes, a set of candidate resource nodes is first generated among all the resource nodes in the network, and by calculating the fitness values of each candidate resource node, a resource node combination close to the optimal solution can be quickly found among multiple resource nodes based on each fitness value, that is, the resource node group that optimizes the performance of the power terminal can be found more efficiently, reducing the search time and calculation cost.

[0078] Based on the above embodiments, an embodiment is provided to illustrate the process of predicting the resource demand change information of the above-mentioned target power terminal.

[0079] In an exemplary embodiment, predicting the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal includes:

[0080] Inputting the monitoring status data of the target power terminal into a pre-trained resource demand prediction model to obtain the resource demand change information of the target power terminal output by the resource demand prediction model; the resource demand prediction model is trained using the monitoring status data and resource quantity data of multiple sample power terminals at different time periods.

[0081] In the embodiments of the present application, historical data and real-time monitoring data are used to train a machine learning model to obtain a resource demand prediction model, so as to predict future resource demand information through the trained model.

[0082] Exemplarily, when training the resource demand prediction model, the monitoring status data of multiple sample power terminals at different time periods is used as input data, and the resource quantity data corresponding to different time periods is used as standard data. The monitoring status data of multiple sample power terminals at different time periods is input into a neural network model, and the neural network model analyzes and processes the monitoring status data to output a predicted resource quantity. Then, based on the predicted resource quantity and the resource quantity data, a loss calculation is performed to obtain the prediction loss value of the neural network model. Thus, based on the prediction loss value, the model parameters of the neural network model are continuously adjusted until the training is completed, and the trained model, that is, the resource demand prediction model, is obtained. After the resource demand prediction model is trained, the monitoring status data of the target power terminal is input into the trained resource demand prediction model, and the resource demand prediction model processes the monitoring status data of the target power terminal to output the resource demand change information of the target power terminal.

[0083] The resource management method provided by the embodiment of the present application inputs the monitoring status data of the target power terminal into a pre-trained resource demand prediction model to obtain the resource demand change information of the target power terminal output by the resource demand prediction model. The resource demand prediction model is trained using the monitoring status data and resource quantity data of multiple sample power terminals in different time periods. In this method, the prediction model is trained through historical data to predict the resource demand change information through the prediction model, enabling early resource planning and scheduling, ensuring that the resource allocation matches the actual demand, and reducing the overall waste of resources.

[0084] Based on the above embodiment, an embodiment for the process of adjusting the resource nodes of the target power terminal is provided for description.

[0085] In an exemplary embodiment, as Figure 4 shown, the resource demand change information includes a reduction in the resource quantity; according to the resource demand change information, the resource nodes of the target power terminal are adjusted to complete the resource management of the target power terminal, including:

[0086] S401. Determine the resource status quantization value of each optimal resource node according to the load rate and energy efficiency ratio of each optimal resource node in the optimal resource node group.

[0087] In the embodiment of the present application, the load rate and energy efficiency ratio of each resource node are stored in the database. Based on this, the load rate and energy efficiency ratio of each optimal resource node can be obtained from the database. Exemplarily, the node identifier of each optimal resource node in the optimal resource node group can be obtained, and then according to the node identifiers of each optimal resource node, the node information matching each node identifier can be obtained from the database, and then the load rate and energy efficiency ratio of each optimal resource node can be obtained from each node information.

[0088] After obtaining the load rate and energy efficiency ratio of each optimal resource node, based on each load rate and energy efficiency ratio, calculate the resource status quantization value of each optimal resource node. The calculation formula of the resource status quantization value is , is the resource status quantization value of the nth optimal resource node, is the load rate of the nth optimal resource node, is the energy efficiency ratio of the nth optimal resource node.

[0089] S402. Adjust the resource nodes of the target power terminal according to each resource status quantization value to achieve the resource management of the target power terminal.

[0090] Exemplarily, according to the quantization values of each resource state, determine the node priorities of each optimal resource node; according to each node priority, migrate the optimal resource node with the lowest node priority from the optimal resource node group to complete the resource management of the target power terminal.

[0091] In the embodiment of the present application, if the resource demand change information is a decrease in the resource amount, then at this time, it is necessary to reduce the resource nodes from the resource nodes of the target power terminal. Based on this, select the resource nodes to be reduced from the multiple optimal resource nodes of the target power terminal, so as to migrate the resource nodes to be reduced from the optimal resource group of the target power terminal to achieve the resource management of the target power terminal.

[0092] Sort each optimal resource node according to the quantization values of each resource state to obtain the node priorities of each optimal resource node. Then, based on the node priorities of each optimal resource node, migrate the optimal resource node with the lowest node priority from the optimal resource node group. That is, the resource migration strategy can be to select the node with the lowest load rate and the lowest energy efficiency ratio as the migration option, migrating from the idle slice to the busy slice to complete the resource management of the target power terminal.

[0093] On the premise of meeting the resource migration strategy, start to gradually recycle resources from the node with the lowest load rate and the lowest energy efficiency ratio. The resource recycling stop conditions include that the total amount of recycled resources reaches a predetermined target and further recycling will affect the normal operation of the current task.

[0094] The resource management method provided by the embodiment of the present application first obtains the load rate and energy efficiency ratio of each optimal resource node in the optimal resource node group, determines the quantization values of each resource state of each optimal resource node, and then adjusts the resource nodes of the target power terminal according to the quantization values of each resource state to achieve the resource management of the target power terminal. In this method, by calculating the quantization values of each optimal resource node, based on this quantization value, select and schedule the resource nodes of the target power terminal, maximize the utilization rate of computing resources, and achieve load balancing. This resource optimization method can prevent the excessive consumption of computing power of some nodes, and at the same time avoid the idleness of other node resources, significantly improving the resource utilization efficiency.

[0095] In addition, in an exemplary embodiment, an embodiment of the dynamic resource management system in the embodiment of the present application is described.

[0096] As Figure 5 shown, it is a schematic diagram of the architecture of the dynamic resource management system. The dynamic resource management system includes:

[0097] The Network Slice Management Function (NSMF) has the function of network slice management, responsible for the management and configuration of task groups, including receiving task group information, generating slice configuration requests, and sending the generated slice configuration information to the slice subnet management module; and real-time monitoring of the resource usage of each slice. This module is located in the core network of the mobile network.

[0098] The Network Slice Subnet Management Function (NSSMF) has the function of network slice subnet management. According to the slice configuration request of the NSMF, it allocates and configures subnet slices; and feeds back the subnet slice configuration information to the NSMF. This module is located in the core network of the mobile network.

[0099] The demand awareness module: This module real-time monitors the computing task demands of power terminals and feeds back the demand changes to the slice management module. This module is responsible for capturing the task characteristics of power terminals, such as computing demands, storage demands, bandwidth demands, and latency requirements; when new demands or changes in existing task demands are detected, it triggers a resource adjustment request. This module is located in the Radio Access Network (RAN) part of the mobile network and is usually deployed on edge computing nodes. These nodes are close to power terminals and are responsible for real-time monitoring of the computing task demands of power terminals.

[0100] The resource management module: This module is responsible for real-time monitoring and managing all resource nodes in the system, recording status information such as the computing power, bandwidth, storage, and current load of each node. At the same time, it uses multi-objective optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.) to select resource nodes across the network to build an optimal resource combination that meets the demands of power terminals.

[0101] The load prediction module uses machine learning models to predict future resource demand changes through historical data and real-time monitoring data for advance resource adjustment planning. This module is located on the edge nodes of the data analysis center of the power grid.

[0102] In the embodiments of this application, the process of resource management using a dynamic resource management system includes the following steps:

[0103] (1) Report slice requests. When a power terminal initiates a new slice request, the demand awareness module in the radio access network of the mobile network will capture these requests and transfer them to the slice management module in the core network.

[0104] (2) Slice planning. The slice management module then invokes the resource management module to screen all available resources, select a combination of resource nodes that meet the conditions from the network-wide resource pool, and use a multi-objective optimization algorithm to select the optimal subset of nodes. This resource selection process will consider the resource distribution in both the 5G network and the power grid simultaneously to determine the resource selection for the slice network.

[0105] (3) Slice information distribution. After the resource selection is completed, the resource management module feeds back the result to the slice management module. The slice management module configures the selected resource nodes based on the feedback and distributes the slice configuration information to the slice subnet management module.

[0106] (4) Slice creation. The slice subnet management module synchronously creates a new slice network in the network and feeds back the information of successful creation to the slice management module. At the same time, the slice subnet management module feeds back the information of successful creation to the demand awareness module and notifies the power terminal to access the newly created slice network. For example, Figure 5 Slice 1 in

[0107] (5) Load prediction. During the operation of the slice network, the load prediction module continuously monitors the status of power terminals, monitors the joining of new power terminals or the leaving of existing terminals, predicts the changes in resources based on this, conducts load prediction, further updates the demand information, and transfers the updated demand to the slice management module.

[0108] (6) Resource adjustment. The slice management module invokes the resource management module to evaluate the resource usage of the current slice and, if necessary, increase or reduce resource nodes through the resource management module. After the resource management module executes the resource adjustment, it feeds back the result to the slice management module.

[0109] (7) Slice reconfiguration. The slice management module distributes the slice reconfiguration information to the slice subnet management module. The slice subnet management module synchronously modifies the slice network configuration in the power grid and the 5G network and feeds back the information of successful creation to the slice management module. At the same time, the slice subnet management module feeds back the information of successful creation to the demand awareness module and notifies the power terminal to access the newly created slice network.

[0110] (8) Modification feedback. After the resource adjustment is successful, the slice management module sends a confirmation feedback to the demand awareness module and the power terminal to ensure that the impact of the resource adjustment on the terminal tasks is minimized.

[0111] In this embodiment, a demand awareness module and a resource management module are introduced into the dynamic resource management framework, enabling the network slice to no longer be a pre-configured static structure but to dynamically adjust resources according to real-time demands. This solution can monitor the computing task demands of power terminals in real time and dynamically allocate or reclaim resource nodes through the slice management module to ensure that the network slice can flexibly respond to the changing computing task demands in the power system. This dynamic adjustment ability significantly improves the applicability of network slicing technology in the power system and overcomes the limitations of the prior art that cannot adapt to real-time changes. Through the combination of the load prediction module and the resource management module, support for multiple application scenarios (such as load prediction, intelligent scheduling, energy management, etc.) is achieved. The service type of the slice network is no longer fixed but is flexibly adjusted according to real-time demands, capable of meeting the diverse application scenarios in the power grid. This technical effect greatly improves the flexibility and adaptability of network slicing and can effectively address different types of power system tasks and service demands. Through the multi-objective optimization algorithm, the selection and scheduling of resource nodes are carried out to maximize the utilization rate of computing resources and achieve load balancing. This resource optimization method can prevent the excessive consumption of computing power of some nodes while avoiding the idle of other node resources, significantly improving the resource utilization efficiency. In addition, the resource management module combined with the load prediction model can perform resource planning and scheduling in advance to ensure that the resource configuration matches the actual demands and reduce the overall waste of resources. The architecture design supports the adaptation to multiple types of power terminal devices and application scenarios and can perform real-time resource allocation and scheduling according to the specific demands of different computing tasks (such as computing intensity, bandwidth requirements, storage capacity, etc.). This flexibility ensures that the system can handle diverse dynamic demands and effectively solves the challenges that are difficult to meet by the prior art.

[0112] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same time but can be executed at different times. The execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0113] Based on the same inventive concept, an embodiment of the present application further provides a resource management device for implementing the above-mentioned resource management method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the resource management device provided below can refer to the limitations on the resource management method in the above text and will not be elaborated here.

[0114] In an exemplary embodiment, as Figure 6 shown, a resource management device 1 is provided, including: a node group determination module 10, a slice network creation module 20, a resource demand prediction module 30, and a resource adjustment module 40, where:

[0115] The node group determination module 10 is configured to, in response to a slice creation request of a target power terminal, determine an optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal;

[0116] The slice network creation module 20 is configured to use the resources of the optimal resource node group to create a new slice network in the power system network and connect the target power terminal to the new slice network;

[0117] The resource demand prediction module 30 is configured to, during the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal;

[0118] The resource adjustment module 40 is configured to adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

[0119] In one embodiment, the above node group determination module 10 is further configured to:

[0120] Randomly generate a group of candidate solutions among the resource nodes, where each candidate solution represents a candidate resource node group; determine the fitness value of each candidate solution according to the node information of each resource node and the communication delay between each resource node and the target power terminal; determine the optimal resource node group of the target power terminal according to the fitness values of each candidate solution.

[0121] In one embodiment, the above node group determination module 10 is further configured to:

[0122] Determine the candidate solutions whose fitness values meet the preset conditions according to the fitness values of each candidate solution, and perform crossover and mutation on the candidate solutions that meet the conditions to generate new solutions; iteratively execute the step of generating new solutions until the preset iteration stop condition is met, and use the new solution with the highest fitness value as the optimal resource node group of the target power terminal.

[0123] In one embodiment, the above-mentioned resource demand prediction module 30 is further configured to:

[0124] Input the monitoring status data of the target power terminal into a pre-trained resource demand prediction model to obtain the resource demand change information of the target power terminal output by the resource demand prediction model; the resource demand prediction model is trained using the monitoring status data and resource quantity data of multiple sample power terminals at different time periods.

[0125] In one embodiment, the above-mentioned resource adjustment module 40 is further configured to:

[0126] Obtain the load rate and energy efficiency ratio of each optimal resource node in the optimal resource node group, and determine the resource status quantization value of each optimal resource node; according to each resource status quantization value, adjust the resource nodes of the target power terminal to achieve resource management of the target power terminal.

[0127] In one embodiment, the above-mentioned resource adjustment module 40 is further configured to:

[0128] According to each resource status quantization value, determine the node priority of each optimal resource node; according to each node priority, migrate the optimal resource node with the lowest node priority from the optimal resource node group to complete the resource management of the target power terminal.

[0129] Each module in the above-mentioned resource management device can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0130] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented:

[0131] In response to a slice creation request of the target power terminal, determine the optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal;

[0132] Use the resources of the optimal resource node group to create a new slice network in the network of the power system, and connect the target power terminal to the new slice network;

[0133] During the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal;

[0134] Adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

[0135] The implementation principles and technical effects of the steps implemented by the processor in the embodiments of this application are similar to those of the above resource management method, and will not be elaborated here.

[0136] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0137] In response to a slice creation request of the target power terminal, determine the optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal;

[0138] Use the resources of the optimal resource node group to create a new slice network in the network of the power system and connect the target power terminal to the new slice network;

[0139] During the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal;

[0140] Adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

[0141] The implementation principles and technical effects of the steps implemented when the computer program in the embodiments of this application is executed by a processor are similar to those of the above resource management method, and will not be elaborated here.

[0142] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the following steps are implemented:

[0143] In response to a slice creation request of the target power terminal, determine the optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal;

[0144] Use the resources of the optimal resource node group to create a new slice network in the network of the power system and connect the target power terminal to the new slice network;

[0145] During the operation of the new slice network, predict the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal;

[0146] Adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

[0147] In the embodiments of the present application, the steps implemented when the computer program is executed by the processor have similar principles and technical effects to those of the above resource management method, which will not be elaborated here.

[0148] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0149] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., and are not limited thereto.

[0150] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0151] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A resource management method, characterized in that The method includes: In response to a slice creation request of a target power terminal, determining an optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each of the resource nodes and the target power terminal; Using the resources of the optimal resource node group to create a new slice network in the network of the power system, and connecting the target power terminal to the new slice network; During the operation of the new slice network, predicting the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal; Adjusting the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

2. The method according to claim 1, wherein The communication information includes communication delay; The determining the optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each of the resource nodes and the target power terminal includes: Randomly generating a group of candidate solutions among the resource nodes, and each candidate solution represents a candidate resource node group; Determining the fitness value of each candidate solution according to the node information of each resource node and the communication delay between each resource node and the target power terminal; Determining the optimal resource node group of the target power terminal according to the fitness values of the candidate solutions.

3. The method according to claim 2, characterized in that, The determining the optimal resource node group of the target power terminal according to the fitness values of the candidate solutions includes: Determining candidate solutions whose fitness values meet the preset conditions according to the fitness values of the candidate solutions, and performing crossover and mutation on the candidate solutions that meet the conditions to generate new solutions; Iteratively executing the step of generating new solutions until the preset iteration stop condition is met, and taking the new solution with the highest fitness value as the optimal resource node group of the target power terminal.

4. The method according to any one of claims 1 to 3, characterized in that, The predicting the resource demand change information of the target power terminal according to the monitoring status data of the target power terminal includes: Inputting the monitoring status data of the target power terminal into a pre-trained resource demand prediction model to obtain the resource demand change information of the target power terminal output by the resource demand prediction model; the resource demand prediction model is trained using the monitoring status data and resource quantity data of multiple sample power terminals at different time periods.

5. The method according to any one of claims 1 to 3, characterized in that, The resource demand change information includes a reduction in resource quantity; the adjusting the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal includes: Determining the resource status quantization value of each optimal resource node according to the load rate and energy efficiency ratio of each optimal resource node in the optimal resource node group; Adjusting the resource nodes of the target power terminal according to the resource status quantization values to achieve the resource management of the target power terminal.

6. The method according to claim 5, characterized in that The adjusting the resource nodes of the target power terminal according to the resource status quantization values includes: Determining the node priority of each optimal resource node according to the resource status quantization values; According to each of the node priorities, migrate the optimal resource node with the lowest node priority from the optimal resource node group to complete the resource management of the target power terminal.

7. A resource management device, characterized in that, The device includes: A node group determination module, configured to, in response to a slice creation request of a target power terminal, determine an optimal resource node group of the target power terminal according to the node information of each resource node in the power system and the communication information between each resource node and the target power terminal; A slice network creation module, configured to use the resources of the optimal resource node group to create a new slice network in the network of the power system and connect the target power terminal to the new slice network; A resource demand prediction module, configured to predict the resource demand change information of the target power terminal according to the monitored status data of the target power terminal during the operation of the new slice network; A resource adjustment module, configured to adjust the resource nodes of the target power terminal according to the resource demand change information to complete the resource management of the target power terminal.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

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

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

Images