Electric power communication network route intelligent recommendation system

Abstract

The invention relates to the technical field of electric power automation, and discloses an electric power communication network route intelligent recommendation system which comprises a sensing module, a strategy module, a decision module and a communication module. According to the application, the sensing module, the strategy module, the decision module and the communication module are utilized, the sensing module senses the operation states of the intelligent agent and the neighbor intelligent agent in real time, the strategy module provides a routing recommendation strategy based on the states, and the decision module performs routing intelligent recommendation. The communication module realizes information exchange and cooperation among the modules and with other communication modules, realizes routing intelligent recommendation of the electric power communication network, perceives states based on different triggering conditions, provides various routing strategies, adapts to dynamic changes of the network, and improves reliability, adaptability, load balancing and collaboration of the network.

Description

Technical Field

[0001] This application relates to the field of power automation technology, and specifically to an intelligent routing recommendation system for power communication networks. Background Art

[0002] In power communication networks, efficient and reliable routing selection is crucial. However, the inventors found the following problems in the research on existing power communication network routing technologies:

[0003] 1. Limitations of centralized control: Currently, many power communication network routings adopt a centralized control method, where a central node is responsible for routing decisions; this method has a risk of single-point failure. Once the central node has problems, the routing function of the entire network may be severely affected; moreover, when dealing with large-scale networks, centralized control may face problems such as high computational pressure and long decision response times, and it is difficult to meet the requirements of the power system for real-time performance and reliability;

[0004] 2. Lack of dynamic adaptability: Existing routing schemes often cannot well adapt to the dynamic changes of the network environment. For example, when the network load changes, a fault occurs, or there are new service requirements, the routing system cannot adjust the routing strategy in a timely manner, resulting in a decline in network performance; in power communication networks, the change of service traffic is relatively frequent, and existing routing technologies are difficult to respond quickly, which may cause network congestion or resource waste;

[0005] 3. Insufficient load balancing ability: In power communication networks, the load conditions of different nodes and links may vary greatly; existing routing technologies have poor effects in achieving load balancing, and it is easy to cause some nodes or links to be overloaded, while other parts of the resources are not fully utilized; this not only affects network performance but also may reduce the reliability and stability of the network;

[0006] 4. Lack of coordination: There is no effective coordination mechanism between network devices; in the face of complex network situations, the decisions of individual devices may not achieve global optimality; for example, when a network fault occurs, different devices may independently search for alternative paths without coordination, resulting in low routing efficiency and even possible conflicts.

[0007] In summary, there is an urgent need for a new technical solution for intelligent routing recommendation in power communication networks to improve the reliability, adaptability, load balancing, and coordination of power communication network routing recommendations. Summary of the Invention

[0008] The purpose of this application is to provide an intelligent routing recommendation system for power communication networks to solve the technical problems raised in the above background art.

[0009] To achieve the above object, the present application discloses the following technical solutions: A power communication network routing intelligent recommendation system, comprising: a sensing module, a policy module, a decision-making module, and a communication module; the sensing module, the policy module, and the decision-making module are all communicatively connected to the communication module;

[0010] The sensing module is used to sense the first operating state of the agent itself and the second operating state of the neighbor agent in real time; wherein, the agent is a digital mapping of the network node of the corresponding power communication network, and the operating state at least includes the load condition, connection status, available bandwidth, the operating state includes the first operating state and the second operating state, and the sensing module includes multiple agents; the triggering conditions for the operation of the sensing module include periodic triggering and event triggering, the periodic triggering is to periodically trigger the operation of the sensing module, and the event triggering is to temporarily trigger the operation of the sensing module;

[0011] The policy module is used to provide corresponding routing recommendation policies based on the operating state for the decision-making module to perform routing intelligent recommendation; the routing recommendation policies at least include a distributed routing protocol, a load balancing protocol, a fault recovery protocol, and a dynamic adaptation protocol, and the routing intelligent recommendation is used to recommend the routing of the power communication network based on the obtained routing recommendation policies;

[0012] The decision-making module is used to perform the routing intelligent recommendation;

[0013] The communication module is used to exchange information and cooperate among the sensing module, the policy module, and the decision-making module, and the communication module is also used to communicatively connect with other communication modules in the power communication network and exchange information and cooperate.

[0014] Preferably, the agent is a digital mapping of the network node of the corresponding power communication network, specifically:

[0015] Collect the operation data of each network node of the power communication network, the network nodes at least include routers, switches, and servers, and the operation data at least includes load data, connection data, and bandwidth data;

[0016] Extract features from the network nodes and the corresponding operation data to obtain the corresponding agents and the corresponding operating states.

[0017] Preferably, the periodic triggering is to periodically collect the operation data based on a preset sensing period to obtain the corresponding operating state;

[0018] The event trigger is to temporarily collect the operation data when there is an event in the preset event library to obtain the corresponding operation status; wherein, the event library is used to store the events for determining whether the event trigger is satisfied, and the events at least include load events, fault events, and service events.

[0019] Preferably, the operation of the policy module is specifically as follows:

[0020] A1: When the operation status is obtained based on the periodic trigger, execute step A2; when the operation status is obtained based on the event trigger, execute step A3;

[0021] A2: Output the distributed routing protocol;

[0022] A3: Judge the event corresponding to the event trigger, and the events at least include load events, fault events, and service events;

[0023] When the event is the load event, output the load balancing protocol; when the event is the fault event, output the fault recovery protocol; when the event is the service event, output the dynamic adaptation protocol.

[0024] Preferably, the distributed routing protocol is specifically as follows:

[0025] Set a corresponding routing table for the agent and train it using a machine learning algorithm. The routing table stores the next-hop nodes and estimated distances for the agent to reach different destinations;

[0026] After being trained by the machine learning algorithm, when the agent is called to execute the distributed routing protocol, it performs the following steps:

[0027] B1: Obtain the first operation status and the second operation status;

[0028] B2: Update the routing table using the first operation status and the second operation status obtained in step B1;

[0029] B3: Perform routing intelligent recommendation based on the routing table updated in step B2.

[0030] Preferably, the load balancing protocol is specifically as follows:

[0031] Train the agent using a load balancing algorithm;

[0032] After being trained by the load balancing algorithm, when the agent is called to execute the load balancing protocol, it performs the following steps:

[0033] C1: Monitor the load conditions of the agent itself and its neighboring agents. When the load event occurs, execute step C2. The load event specifically refers to the event that the load of the agent itself is too high.

[0034] C2: Send a request to the neighboring agent with a lower load to negotiate the transfer of some traffic.

[0035] Preferably, the fault recovery protocol is specifically as follows:

[0036] Train the agent using network fault recovery technology.

[0037] After being trained by the network fault recovery technology, when the agent is called to execute the fault recovery protocol, it performs the following steps:

[0038] D1: Monitor the link and device status in the power communication network where the agent is located. When the fault event occurs, execute step D2. The fault event specifically refers to the event that a fault is detected in the power communication network where the agent is located.

[0039] D2: Start the preset fault recovery process and call the backup routing path.

[0040] Preferably, the dynamic adaptation protocol is specifically as follows:

[0041] Train the agent using dynamic programming technology.

[0042] After being trained by the dynamic programming technology, when the agent is called to execute the dynamic adaptation protocol, it performs the following steps:

[0043] E1: Monitor the service requirements and environmental changes in the power communication network where the agent is located. When the service event occurs, execute step E2. The service event specifically refers to the change situation of the service requirements.

[0044] E2: Analyze the service event and call the distributed routing protocol to execute steps B1 to B3.

[0045] Preferably, the operation of the decision-making module is specifically as follows:

[0046] F1: Receive the routing recommendation strategy.

[0047] F2: Run the routing recommendation strategy obtained in step F1 and perform routing intelligent recommendation to obtain the recommendation of the routing of the corresponding power communication network.

[0048] Preferably, the decision-making module also communicates with multiple policy modules through the communication module.

[0049] When the decision-making module is communicatively connected to a plurality of the policy modules, it receives a plurality of the routing recommendation policies and integrates them to obtain a recommendation for a coordinated routing of the power communication network.

[0050] Beneficial effects: The intelligent routing recommendation system for a power communication network of the present application... BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0052] Figure 1 It is a structural diagram of the intelligent routing recommendation system for a power communication network provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0054] In this article, the term "including" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without more limitations, the elements defined by the statement "including..." do not exclude the existence of additional identical elements in the process, method, article or device including the said elements.

[0055] This embodiment discloses a kind of intelligent routing recommendation system for a power communication network as Figure 1 shown, including: a sensing module, a policy module, a decision-making module and a communication module; the sensing module, the policy module and the decision-making module are all communicatively connected to the communication module;

[0056] The perception module is used to perceive the first operating state of the agent itself and the second operating state of neighboring agents in real time; wherein, the agent is a digital mapping of the network nodes of the corresponding power communication network, and the operating state at least includes the load condition, connection status, and available bandwidth. The operating state includes the first operating state and the second operating state, and the perception module includes multiple agents; the triggering conditions for the operation of the perception module include periodic triggering and event triggering. Periodic triggering is to periodically trigger the operation of the perception module, and event triggering is to temporarily trigger the operation of the perception module;

[0057] The policy module is used to provide corresponding routing recommendation policies based on the operating state for the decision-making module to perform intelligent routing recommendation; the routing recommendation policies at least include distributed routing protocols, load balancing protocols, fault recovery protocols, and dynamic adaptation protocols. Intelligent routing recommendation is used to recommend the routing of the power communication network based on the obtained routing recommendation policies;

[0058] The decision-making module is used to perform intelligent routing recommendation;

[0059] The communication module is used to exchange information and cooperate among the perception module, policy module, and decision-making module. The communication module is also used to communicate and connect with other communication modules in the power communication network and exchange information and cooperate.

[0060] In a simple example, the first operating state of this embodiment can be that the router agent can determine its own load by monitoring its own port traffic and processing capabilities, and the second operating state of this embodiment can be to obtain their connection status and available bandwidth information by communicating with neighboring routers.

[0061] By the above, this embodiment utilizes the perception module, policy module, decision-making module, and communication module. The perception module perceives the operating states of the agent itself and neighboring agents in real time. The policy module provides routing recommendation policies based on the states. The decision-making module performs intelligent routing recommendation. The communication module realizes information exchange and cooperation among the modules and with other communication modules, realizes intelligent routing recommendation of the power communication network, perceives states based on different triggering conditions, provides multiple routing policies, adapts to network dynamic changes, and improves the reliability, adaptability, load balancing, and coordination of the network.

[0062] Specifically, the agent is a digital mapping of the network nodes of the corresponding power communication network, specifically:

[0063] Collect the operation data of each network node of the power communication network. The network nodes at least include routers, switches, and servers, and the operation data at least includes load data, connection data, and bandwidth data;

[0064] Extract features from the network nodes and the corresponding operation data to obtain the corresponding agents and the corresponding operating states.

[0065] It should be noted that in this embodiment, existing feature extraction technologies are used to extract features from network nodes and corresponding operation data, obtaining corresponding agents and corresponding operation states.

[0066] Through the above, this embodiment collects and extracts features from the operation data of power communication network nodes. By mapping network nodes to agents and corresponding operation states, it provides a data basis and a technical basis for the creation of agents, and provides data support for accurately perceiving the network state and subsequent intelligent routing recommendations.

[0067] Specifically, periodic triggering is to collect periodic operation data based on a preset perception period to obtain corresponding operation states;

[0068] Event triggering is to collect temporary operation data when there are events in a preset event library to obtain corresponding operation states; among them, the event library is used to store events for determining whether event triggering is satisfied, and the events at least include load events, fault events, and service events.

[0069] In a specific application, the perception period is an empirical value well-known to those skilled in the art, and status information is exchanged regularly to maintain an understanding of the overall network situation; the events are empirical events well-known to those skilled in the art. Exemplarily, when a certain agent detects a major change such as a network fault or a drastic change in load, it immediately sends a message to surrounding agents. Through different triggering conditions, the agents can share information in a timely manner to jointly handle various situations in the network. Exemplarily, when a link fails, adjacent agents can quickly spread the fault information through the communication module so that other agents can adjust their routing strategies.

[0070] In a simple example, the load event in this embodiment can be that the load of a certain node exceeds a preset threshold. For example, when the CPU utilization rate of a router reaches more than 80% or the port traffic reaches more than 70% of the bandwidth, it can also be that the average load of multiple nodes in a specific area continues to rise by more than a certain amplitude. For example, the average load of multiple communication devices in a substation increases by 30% within a certain period of time; the fault event in this embodiment can be that a certain key network device, such as a core router or a switch, issues a fault alarm, or it can also be that the number of consecutive restarts of the device exceeds a preset number. Exemplarily, a server automatically restarts three times within one hour; the service event in this embodiment can be that a new high-priority service request accesses the network, such as an urgent power dispatching instruction needs to be transmitted quickly, or it can also be that the traffic of a certain specific service suddenly increases significantly. Exemplarily, the traffic of the video surveillance service doubles within a short period of time.

[0071] Through the above, this embodiment utilizes a preset sensing period and an event library, and collects operation data through periodic triggering and event triggering, achieving flexible acquisition of the operation status of network nodes, timely sensing of network changes based on different situations, and providing the latest data basis for route recommendation.

[0072] Specifically, the operation of the policy module is as follows:

[0073] A1: When the operation status is obtained based on periodic triggering, execute step A2; when the operation status is obtained based on event triggering, execute step A3;

[0074] A2: Output a distributed routing protocol;

[0075] A3: Judge the event corresponding to the event triggering. The event includes at least a load event, a fault event, and a service event;

[0076] When the event is a load event, output a load balancing protocol; when the event is a fault event, output a fault recovery protocol; when the event is a service event, output a dynamic adaptation protocol.

[0077] Through the above, this embodiment judges the event type based on the triggering method, and realizes providing a suitable strategy for the decision-making module according to different situations by outputting the corresponding route recommendation strategy, ensuring effective route intelligent recommendation in different scenarios.

[0078] Specifically, the distributed routing protocol is as follows:

[0079] Set a corresponding routing table for the agent and train it using a machine learning algorithm. The routing table stores the next-hop node and the estimated distance for the agent to reach different destinations;

[0080] The agent trained by the machine learning algorithm executes the following steps when the distributed routing protocol is called:

[0081] B1: Obtain the first operation status and the second operation status;

[0082] B2: Update the routing table using the first operation status and the second operation status obtained in step B1;

[0083] B3: Perform route intelligent recommendation based on the routing table updated in step B2.

[0084] It should be noted that in this embodiment, existing machine learning algorithms are utilized. Exemplarily, such as using the ant colony algorithm, the agent can leave a certain amount of "pheromone" on the path according to the quality of the path during the data transmission process. The intensity of the pheromone represents the quality of the path. When other agents select a route, they tend to choose the path with a high pheromone intensity, so as to find the optimal routing path through mutual cooperation without centralized control.

[0085] By the above, in this embodiment, a routing table is set for the agent and trained using machine learning algorithms. By obtaining state updates to the routing table for intelligent routing recommendations, intelligent routing selection under the distributed routing protocol is achieved, thereby adapting to network dynamic changes and finding the optimal routing path.

[0086] Specifically, the load balancing protocol is as follows:

[0087] Train the agent using the load balancing algorithm;

[0088] The agent trained by the load balancing algorithm executes the following steps when the load balancing protocol is called:

[0089] C1: Monitor the load conditions of the agent itself and neighbor agents. When a load event occurs, execute step C2. The load event is specifically an event where the load of the agent itself is too high;

[0090] C2: Send a request to the neighbor agent with a lower load to negotiate the transfer of some traffic over.

[0091] It should be noted that in this embodiment, existing load balancing algorithms are used for agent training. In a specific application, this embodiment monitors the CPU utilization rate, port traffic, etc. to judge its own load. When it is too high, it sends a request to the neighbor agent with a lower load to negotiate the transfer of some traffic over. Algorithms such as weighted round-robin and least connections can be used to determine the traffic allocation.

[0092] In this embodiment, the agent is trained using the load balancing algorithm. By monitoring the load conditions and transferring traffic to neighbor agents with low load, network load balancing under the load balancing protocol is achieved, improving network resource utilization.

[0093] Specifically, the fault recovery protocol is as follows:

[0094] Train the agent using network fault recovery technology;

[0095] The agent trained by the network fault recovery technology executes the following steps when the fault recovery protocol is called:

[0096] D1: Monitor the link and device status in the power communication network where the agent is located. When a fault event occurs, execute step D2. The fault event specifically refers to an event where a fault is detected in the power communication network where the agent is located;

[0097] D2: Start the preset fault recovery process and call the backup routing path.

[0098] It should be noted that in this embodiment, the agent is trained using the existing backup path technology to find the pre-computed backup path, or cooperate with neighbor agents to jointly find a new available path.

[0099] Through the above, in this embodiment, the agent is trained using the network fault recovery technology. By monitoring the link and device status and calling the backup routing, a fast fault response under the fault recovery protocol is achieved, thus ensuring communication continuity.

[0100] Specifically, the dynamic adaptation protocol is specifically as follows:

[0101] Train the agent using dynamic programming technology;

[0102] The agent trained by the dynamic programming technology executes the following steps when the dynamic adaptation protocol is called:

[0103] E1: Monitor the service requirements and environmental changes in the power communication network where the agent is located. When a service event occurs, execute step E2. The service event specifically refers to the change situation of the service requirements;

[0104] E2: Analyze the service event and call the distributed routing protocol to execute steps B1 to B3.

[0105] It should be noted that in this embodiment, the agent is trained using the existing dynamic programming technology. Exemplarily, such as monitoring the spatio-temporal distribution of service traffic, network latency, etc. It can also be that during the peak service period, preferentially select links with high bandwidth and low latency; during the off-peak service period, select more energy-efficient routing paths.

[0106] Through the above, in this embodiment, the agent is trained using the dynamic programming technology. By monitoring the service requirements and environmental changes and calling the distributed routing protocol, an adaptive routing adjustment under the dynamic adaptation protocol is achieved, adapting to service changes and optimizing network performance.

[0107] Specifically, the operation of the decision-making module is specifically as follows:

[0108] F1: Receive the routing recommendation strategy;

[0109] F2: Run the routing recommendation strategy obtained in step F1 and perform intelligent routing recommendation to obtain the recommendation of the routing of the corresponding power communication network.

[0110] With the above, this embodiment utilizes the received routing recommendation strategy and runs it. By performing intelligent routing recommendation, it realizes the routing recommendation of the decision-making module, selects routes based on the routing recommendation strategy, and provides optimized routing recommendations for the network.

[0111] Specifically, the decision-making module also communicates with multiple policy modules through the communication module;

[0112] When the decision-making module communicates with multiple policy modules, it receives multiple routing recommendation strategies and integrates them to obtain recommendations for coordinated routes in the power communication network.

[0113] With the above, this embodiment utilizes communication connections with multiple policy modules. By receiving and integrating multiple routing recommendation strategies, it realizes coordinated routing recommendation, improves the accuracy and adaptability of routing recommendation, and better meets the requirements of the power communication network.

[0114] In summary, the intelligent routing recommendation system for the power communication network in this embodiment utilizes the sensing module, policy module, decision-making module, and communication module. Through the sensing module, it can continuously sense the operating states of the agent itself and its neighbor agents in real time. The policy module provides routing recommendation strategies based on the states, the decision-making module performs intelligent routing recommendation, and the communication module realizes information exchange and cooperation among the modules and with other communication modules, thus realizing intelligent routing recommendation for the power communication network. Based on different triggering conditions, it senses the states, provides multiple routing strategies, adapts to the dynamic changes of the network, and improves the reliability, adaptability, load balancing, and coordination of the network.

[0115] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transmission of a computer program from one place to another. The storage media can be any available medium that can be accessed by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disc storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0116] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent recommendation system for power communication network routing, characterized in that: include: A perception module, a strategy module, a decision module and a communication module; the perception module, the strategy module and the decision module are all connected to the communication module for communication; The perception module is used to perceive the first operating state of the intelligent agent itself and the second operating state of the neighboring intelligent agent in real time; wherein the intelligent agent is a digital mapping of the network node of the corresponding power communication network, the operating state includes at least load condition, connection state, and available bandwidth, the operating state includes the first operating state and the second operating state, and the perception module includes a plurality of intelligent agents; the triggering conditions for the operation of the perception module include periodic triggering and event triggering, the periodic triggering is to periodically trigger the operation of the perception module, and the event triggering is to temporarily trigger the operation of the perception module; The strategy module is used to provide a corresponding routing recommendation strategy based on the operating state for the decision module to perform intelligent routing recommendation; the routing recommendation strategy at least includes a distributed routing protocol, a load balancing protocol, a fault recovery protocol and a dynamic adaptation protocol, and the intelligent routing recommendation is used to recommend a route for the power communication network based on the obtained routing recommendation strategy; The decision module is used to perform the intelligent route recommendation; The communication module is used for information exchange and collaboration among the perception module, the strategy module and the decision module. The communication module is also used for communication connection with other communication modules in the power communication network and for information exchange and collaboration.

2. The intelligent recommendation system for power communication network routing according to claim 1, characterized in that: The intelligent agent is a digital mapping of the network nodes of the corresponding power communication network, specifically: Collecting operation data of each network node of the power communication network, wherein the network nodes include at least a router, a switch and a server, and the operation data includes at least load data, connection data and bandwidth data; Feature extraction is performed on the network nodes and the corresponding operating data to obtain the corresponding intelligent agents and the corresponding operating states.

3. The intelligent recommendation system for power communication network routing according to claim 2, characterized in that: The periodic trigger is to periodically collect the operating data based on a preset sensing period to obtain the corresponding operating status; The event trigger is to temporarily collect the operating data when there is an event in the preset event library to obtain the corresponding operating status; wherein the event library is used to store the events for determining whether the event trigger is met, and the events include at least load events, fault events and business events.

4. The power communication network routing intelligent recommendation system according to claim 1, characterized in that: The operation of the policy module is specifically as follows: A1: When the running state is obtained based on the periodic trigger, execute step A2; when the running state is obtained based on the event trigger, execute step A3; A2: output the distributed routing protocol; A3: Determine an event corresponding to the event triggered by the event, where the event at least includes a load event, a fault event, and a service event; When the event is the load event, outputting the load balancing protocol; When the event is the fault event, the fault recovery protocol is output; when the event is the service event, the dynamic adaptation protocol is output.

5. The power communication network routing intelligent recommendation system according to claim 1, characterized in that: The distributed routing protocol is specifically: Setting a corresponding routing table for the agent and training it using a machine learning algorithm, wherein the routing table stores the next hop nodes and estimated distances for the agent to reach different destinations; The agent trained by the machine learning algorithm performs the following steps when the distributed routing protocol is called: B1: Acquire the first operating state and the second operating state; B2: Update the routing table using the first operating state and the second operating state obtained in step B1; B3: Perform the intelligent route recommendation based on the routing table updated in step B2.

6. The power communication network routing intelligent recommendation system according to claim 1, characterized in that: The load balancing protocol is specifically: Training the agent using a load balancing algorithm; The agent trained by the load balancing algorithm performs the following steps when the load balancing protocol is called: C1: monitor the load of the agent itself and neighboring agents, and execute step C2 when the load event occurs, wherein the load event is specifically an event that the load of the agent itself is too high; C2: Send a request to the neighboring agent with lower load to negotiate to transfer part of the traffic to it.

7. The power communication network routing intelligent recommendation system according to claim 1, characterized in that: The fault recovery protocol is specifically as follows: Training the agent using network failure recovery techniques; The agent trained by the network fault recovery technology performs the following steps when the fault recovery protocol is called: D1: monitoring the link and device status in the power communication network where the intelligent agent is located, and executing step D2 when the fault event occurs, wherein the fault event is specifically an event in which a fault is detected in the power communication network where the intelligent agent is located; D2: Start the preset fault recovery process and call the backup routing path.

8. The power communication network routing intelligent recommendation system according to claim 5, characterized in that: The dynamic adaptation protocol is specifically: Training the agent using dynamic programming techniques; The agent trained by dynamic programming technology performs the following steps when the dynamic adaptation protocol is called: E1: monitoring the business demands and environmental changes in the electric power communication network where the intelligent agent is located, and executing step E2 when the business event occurs, wherein the business event is specifically a change in business demands; E2: Analyze the business event and call the distributed routing protocol to execute steps B1 to B3.

9. The power communication network routing intelligent recommendation system according to claim 1, characterized in that: The operation of the decision module is specifically as follows: F1: receiving the routing recommendation strategy; F2: Run the route recommendation strategy obtained in step F1 and perform the intelligent route recommendation to obtain the route recommendation of the corresponding power communication network.

10. The intelligent recommendation system for power communication network routing according to claim 9, characterized in that: The decision module also utilizes the communication module to communicate with the plurality of the strategy modules; When the decision module is in communication connection with the plurality of the strategy modules, it receives and integrates the plurality of route recommendation strategies to obtain recommendations for collaborative routes for the power communication network.

Images