Dynamic resource optimal configuration method for electric power communication network

Abstract

The invention discloses a dynamic resource optimal configuration method for an electric power communication network, and belongs to the technical field of electric power system communication. The problems of low communication resource utilization rate and poor real-time response in the existing electric power communication are solved, the communication resources of physical network equipment are acquired through a network management system, and network operation state data and traffic flow characteristics are acquired in real time by utilizing a network probe or NETCONF / YANG; the method comprises the following steps: identifying service flows flowing into a communication network by adopting a deep packet inspection technology, classifying the service flows according to security partitions of power services, and mapping different types of service flows to different service levels; a quantitative QoS demand template is established for each type of services; and on the basis of the network operation state data, the service flow characteristics and the QoS demand template, constructing a multi-objective optimization model which takes guarantee of the service quality of key services as a highest priority objective, and solving by adopting an optimization algorithm to obtain an optimal resource allocation strategy. The method is suitable for optimal configuration of dynamic resources of the power communication network.

Description

Technical Field

[0001] This invention belongs to the field of power system communication technology. Background Technology

[0002] The power communication network is the nerve center ensuring the safe, stable, and efficient operation of the power grid. With the development of smart grids and the energy internet, power communication services are exhibiting significant differentiation characteristics: on the one hand, production control services (such as relay protection and safety and stability control) have extremely high requirements for communication reliability and real-time performance (reliability > 99.999%, latency < 10ms); on the other hand, management information and internet-related services have high bandwidth requirements. Traditional power communication network resource allocation methods are mostly static or semi-static, employing a pre-planned, fixed bandwidth allocation model. This fails to perceive the dynamic changes in network status and service demands in real time, resulting in the following drawbacks:

[0003] Low resource utilization: To ensure high-priority business, resources are usually reserved based on peak demand, resulting in a large amount of idle resources during off-peak hours.

[0004] Poor flexibility: When network congestion or failure occurs, it is difficult to quickly and accurately reallocate the optimal path and resources for critical services, resulting in slow response speed.

[0005] Coarse Quality of Service (QoS) assurance: It is impossible to achieve fine-grained QoS assurance based on business granularity, making it difficult to meet the needs of "business-driven" intelligent power grids.

[0006] Therefore, there is an urgent need for a resource allocation method that can dynamically sense, make intelligent decisions, and respond in real time to resolve the contradiction between limited communication resources and differentiated service needs. Summary of the Invention

[0007] This invention aims to address the problems of low utilization rate of communication resources and poor real-time response in existing power communication systems, and provides a method for dynamic resource optimization and allocation in power communication networks.

[0008] The present invention provides a method for dynamic resource optimization and allocation in power communication networks, comprising:

[0009] Step 1: Obtain the communication resources of physical network devices through the network management system, and collect network operation status data and service traffic characteristics in real time using network probes or the NETCONF / YANG protocol;

[0010] Step 2: Using deep packet inspection technology, identify the service flows flowing into the communication network, classify the service flows according to the security partition of the power service, map different categories of service flows to different service levels, and establish a quantified QoS requirement template for each type of service.

[0011] Step 3: Based on the network operation status data, service traffic characteristics, and QoS requirement template, construct a multi-objective optimization model with the highest priority of ensuring the quality of critical service, and solve it using an optimization algorithm to obtain the optimal resource allocation strategy.

[0012] Furthermore, in this invention, in step one, the physical device includes a router, a switch, an optical transmission device, or a wireless private network base station;

[0013] The resources include: bandwidth, latency, port or computing resources;

[0014] Network operation status data includes link bandwidth utilization, end-to-end latency, jitter, packet loss rate, device load, and characteristic data of service traffic;

[0015] The characteristic data of the service flow includes: traffic data, source / destination address, and service type.

[0016] Furthermore, in this invention, in step two, the security zones for power services include: a production control zone, a management information zone, and an internet zone;

[0017] The QoS requirement template shall include at least: latency limit, reliability metric and priority.

[0018] Furthermore, in this invention, in step three, the decision variables of the multi-objective optimization model include: route selection, bandwidth allocation, queue scheduling priority, spectrum allocation, or power control;

[0019] The constraints include service reliability requirements, security isolation requirements, and network capacity constraints.

[0020] Furthermore, in this invention, in step three, the optimization algorithm employs a mixed integer linear programming (MILP) algorithm, a heuristic algorithm, or a reinforcement learning (RL) algorithm.

[0021] This invention, through service-driven, fine-grained QoS management, can absolutely guarantee the high reliability and low latency transmission of critical services such as relay protection, greatly improving the inherent safety level of the power grid. Simultaneously, it transforms static configuration into dynamic intelligent allocation, achieving peak shaving and valley filling of network resources, significantly improving the utilization rate of valuable communication resources such as spectrum and bandwidth. The SDN / NFV-based architecture gives the network high programmability and flexibility, enabling rapid response to power grid topology changes, fault recovery, and new service deployment needs. Furthermore, the introduction of artificial intelligence algorithms (especially reinforcement learning) gives the system self-learning, self-optimization, and predictive capabilities, realizing an evolution from passive response to proactive protection. Attached Figure Description

[0022] Figure 1This is a flowchart of the method described in this invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0024] Specific implementation method one: Refer to Figure 1 This embodiment describes a dynamic resource optimization allocation method for power communication networks, comprising:

[0025] Step 1: Obtain the communication resources of physical network devices through the network management system, and collect network operation status data and service traffic characteristics in real time using network probes or the NETCONF / YANG protocol;

[0026] Step 2: Using deep packet inspection technology, identify the service flows flowing into the communication network, classify the service flows according to the security partition of the power service, map different categories of service flows to different service levels, and establish a quantified QoS requirement template for each type of service.

[0027] Step 3: Based on the network operation status data, service traffic characteristics, and QoS requirement template, construct a multi-objective optimization model with the highest priority of ensuring the quality of critical service, and solve it using an optimization algorithm to obtain the optimal resource allocation strategy.

[0028] Furthermore, in this invention, in step one, the physical device includes a router, a switch, an optical transmission device, or a wireless private network base station;

[0029] The resources include bandwidth, latency, port or computing resources;

[0030] Network operation status data includes link bandwidth utilization, end-to-end latency, jitter, packet loss rate, device load, and characteristic data of service traffic;

[0031] The characteristic data of the service flow includes: traffic data, source / destination address, and service type.

[0032] Furthermore, in this invention, in step two, the security zones for power services include: a production control zone, a management information zone, and an internet zone;

[0033] The QoS requirement template shall include at least: latency limit, reliability metric and priority.

[0034] Furthermore, in this invention, in step three, the decision variables of the multi-objective optimization model include: route selection, bandwidth allocation, queue scheduling priority, spectrum allocation, or power control;

[0035] The constraints include service reliability requirements, security isolation requirements, and network capacity constraints.

[0036] Furthermore, in this invention, in step three, the optimization algorithm employs a mixed integer linear programming (MILP) algorithm, a heuristic algorithm, or a reinforcement learning (RL) algorithm.

[0037] The method described in this invention is a dynamic resource optimization and allocation method for power communication networks aimed at ensuring the reliability and real-time performance of power services. Based on the core principles of "service-driven, dynamic sensing, and intelligent allocation," this method employs a three-layer dynamic optimization model, including an infrastructure and sensing layer for infrastructure data acquisition, an intelligent optimization engine control layer using intelligent algorithms, and an application and strategy layer. The method includes the following steps:

[0038] Step S1: Resource abstraction and status monitoring (infrastructure and perception layer).

[0039] S101: Resource Abstraction: Through the SDN controller or network management system, the communication resources (including ports, bandwidth, latency, and computing resources) of physical network devices (including routers, switches, optical transmission equipment SDH / OTN, and wireless private network base stations) are uniformly abstracted to form a globally visible and schedulable resource pool.

[0040] S102: Status Monitoring: Utilize network probes or management protocols such as NETCONF / YANG to collect real-time operational status data of the entire network, including link bandwidth utilization, end-to-end latency, jitter, packet loss rate, device load, and characteristic data of service traffic (including traffic size, source / destination address, and service type).

[0041] Step S2: Business requirement perception and modeling (application and strategy layer);

[0042] S201: Service Identification and Classification: Based on pre-configured policies or deep packet inspection (DPI) technology, identify service flows entering the network and classify them according to the security zones of power services (production control zone, management information zone, and Internet zone), mapping them to different service levels.

[0043] S202: QoS Policy Templates: Establish quantified QoS requirement templates for each type of service. For example, define a template for "Relay Protection Service": latency limit <10ms, reliability >99.999%, and assign it the highest priority.

[0044] Step S3: Intelligent Optimization Decision (Intelligent Optimization Engine Control Layer);

[0045] This step is the core of the method. The intelligent optimization engine receives data from S1 and S2 and performs the following operations:

[0046] S301: Constructing a multi-objective optimization model:

[0047] Objective function: Minimizing end-to-end latency and packet loss rate for critical power services is the highest priority objective, while maximizing overall network resource utilization is the secondary objective.

[0048] Constraints include business reliability constraints (requiring physical separation of primary and backup paths), security isolation constraints (isolation of business traffic in different regions), and resource capacity constraints.

[0049] Decision variables include traffic routing, bandwidth allocation, queue scheduling priority, spectrum resource allocation (for optical networks), and transmit power control (for wireless networks).

[0050] S302: Solve the optimization model by running the optimization algorithm: Use one or more of the following algorithms in combination to solve the above optimization model:

[0051] For small-scale networks or offline planning, mixed-integer linear programming (MILP) is used to obtain an exact solution.

[0052] For large-scale real-time optimization, heuristic algorithms (such as genetic algorithms and particle swarm optimization) are used to find satisfactory solutions within a feasible time. When reinforcement learning (RL) algorithms are used, the agent learns the optimal resource allocation strategy autonomously through continuous interaction with the network environment, and has the ability to predict events such as power grid failures and pre-allocate resources.

[0053] Step S4: Strategy issuance and execution;

[0054] S401: Policy Generation and Translation: Transforms the optimization decisions obtained in S3 (such as optimal path, bandwidth value, priority) into specific network device configuration instructions.

[0055] S402: Policy Deployment: Through the southbound interface of the SDN controller (such as the OpenFlow protocol), configuration commands are deployed to the corresponding network devices (such as switches and routers) to dynamically establish or modify the forwarding paths and QoS policies of service flows.

[0056] Step S5: Dynamic feedback and adjustment;

[0057] Continuously monitor the actual transmission quality (QoS) of service flows and changes in network status. When network congestion, faults, or changes in service requirements are detected, feedback is sent to step S1, triggering a new round of optimization decision-making process (S2-S4), achieving closed-loop control and dynamic adjustment of resources.

[0058] Taking the differential protection service of smart distribution networks as an example, the specific application and implementation process of this invention is explained as follows:

[0059] Event Trigger (S1 / S2): A momentary fault occurs on a line in the distribution network. The differential protection device is activated, generating a communication request. The application layer identifies this as the highest priority "differential protection service" through pre-configured policies, with QoS requirements of latency <10ms and reliability >99.999%.

[0060] Status Acquisition (S1): The monitoring system at the infrastructure layer collects the latency and load information of each link in the current network in real time.

[0061] Intelligent Decision (S3): The intelligent optimization engine (using a genetic algorithm in this example) receives the service request and network status and performs calculations within milliseconds. It evaluates all feasible paths from protection device A to protection device B, selects the path with the shortest latency and lightest load as the primary path, calculates a physically separate path as a backup path, and reserves the highest priority queues and guaranteed bandwidth for the service flow on all network devices it passes through.

[0062] Policy Enforcement (S4): The SDN controller translates the above decisions into specific OpenFlow flow table entries and distributes them to all switches along the path. Based on this, the switches establish a "virtual leased line" to serve the differential protection service.

[0063] Service transmission: Differential protection data packets are transmitted quickly and reliably through this dedicated line, enabling the protection device to accurately determine the fault location and operate correctly based on the data.

[0064] Dynamic Adjustment (S5): After the fault is cleared, the differential protection service traffic returns to normal levels. The optimization engine detects this change, triggers the resource reclamation process, releases the previously reserved dedicated resources, and re-includes them in the public resource pool for use by other services.

[0065] The above embodiments demonstrate the effectiveness and superiority of the method of the present invention in ensuring critical power grid operations and improving resource efficiency. Although the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for dynamic resource optimization allocation in a power communication network, characterized in that, include: Step 1: Obtain the communication resources of physical network devices through the network management system, and collect network operation status data and service traffic characteristics in real time using network probes or the NETCONF / YANG protocol; Step 2: Using deep packet inspection technology, identify the service flows flowing into the communication network, classify the service flows according to the security partition of the power service, and map different categories of service flows to different service levels; And establish quantified QoS requirement templates for each type of service; Step 3: Based on the network operation status data, service traffic characteristics, and QoS requirement template, construct a multi-objective optimization model with the highest priority of ensuring the quality of critical service, and solve it using an optimization algorithm to obtain the optimal resource allocation strategy.

2. The method for dynamic resource optimization allocation in a power communication network according to claim 1, characterized in that, In step one, the physical devices include routers, switches, optical transmission equipment, or wireless private network base stations; The resources include bandwidth, latency, port or computing resources; Network operation status data includes link bandwidth utilization, end-to-end latency, jitter, packet loss rate, device load, and characteristic data of service traffic; The characteristic data of the service flow includes: traffic data, source / destination address, and service type.

3. The method for dynamic resource optimization allocation in a power communication network according to claim 1, characterized in that, In step two, the security zones for power operations include: the production control zone, the management information zone, and the internet zone. The QoS requirement template shall include at least: latency limit, reliability metric and priority.

4. The method for dynamic resource optimization allocation in a power communication network according to claim 1, characterized in that, In step three, the decision variables of the multi-objective optimization model include: route selection, bandwidth allocation, queue scheduling priority, spectrum allocation, or power control; The constraints include service reliability requirements, security isolation requirements, and network capacity constraints.

5. The method for dynamic resource optimization allocation in a power communication network according to claim 1, characterized in that, In step three, the optimization algorithm employs mixed integer linear programming (MILP), heuristic algorithms, or reinforcement learning (RL) algorithms.

Images