Method and device for determining sink node of power data communication network, equipment, medium and product

Abstract

The invention discloses a method, a device, equipment, a medium and a product for determining an aggregation node of an electric power data communication network, and the method comprises the steps: obtaining the current link state information, the current environment perception information and the current electric power business data when the link state changes; determining current link multi-dimensional state evaluation information according to the current link state information; determining a business demand classification result according to the current power business data and a pre-constructed improved deep forest model; and according to the current link multi-dimensional state evaluation information, the current environment perception information and the service demand classification result, determining a sink node. And when the link state changes, determining the multi-dimensional state evaluation information of the multi-dimensional current link, and further determining the aggregation node in combination with the sensing condition of the current environment and the classification result of the current service demand. More comprehensive and accurate link state information is provided, and dynamic selection of aggregation nodes is realized to adapt to constantly changing service requirements and network conditions.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular, to a method, apparatus, device, medium, and product for determining a convergence node of a power data communication network. Background Art

[0002] The convergence node of the power data communication network is an important part of the power data communication network. It is located between the access layer and the core layer and is responsible for summarizing the data transmissions of multiple access layer devices and forwarding them to the core layer. The convergence node of the power data communication network plays a crucial role in the power system. It not only ensures the effective convergence and efficient forwarding of data, but also enhances the stability and security of the network through functions such as protocol conversion and providing high availability.

[0003] The prior art usually forms the fitness of the nodes in the power communication network according to the network attributes, power characteristics, computing attributes, and data characteristics of each node in the power communication network. On the basis of this evaluation, nodes ranked among the top in the set of task selection orders issued for the task nodes are selected as working nodes. After the node selection is completed, the paths between the selected working nodes and the task nodes are calculated to form a working layer topology, and the construction of federated learning is completed, considering multiple attributes of the nodes for node selection.

[0004] However, this method only focuses on the attributes of the nodes themselves. When facing sudden network failures or a sharp increase in traffic, the network performance will drop significantly, affecting the safe and stable operation of the power system. In a complex and changing power communication environment, it cannot adapt to the diverse and dynamically changing characteristics of power services and cannot effectively cope with the complex and changing environment of the power data communication network. Summary of the Invention

[0005] The present invention provides a method, apparatus, device, medium, and product for determining a convergence node of a power data communication network to achieve dynamic determination of the convergence node to adapt to changing service requirements and network conditions.

[0006] According to a first aspect of the present invention, there is provided a method for determining a convergence node of a power data communication network, including:

[0007] When the link state changes, obtain the current link state information, the current environmental perception information, and the current power service data;

[0008] Determine the current link multi-dimensional state evaluation information according to the current link state information;

[0009] Determine the service requirement classification result according to the current power service data and a pre-constructed improved deep forest model;

[0010] Determine the aggregation node according to the current link multi-dimensional state evaluation information, the current environment perception information, and the classification result of service requirements.

[0011] According to a second aspect of the present invention, there is provided a device for determining an aggregation node of a power data communication network, including:

[0012] An information acquisition module, configured to acquire current link state information, current environment perception information, and current power service data when the link state changes;

[0013] An information determination module, configured to determine current link multi-dimensional state evaluation information according to the current link state information;

[0014] A result determination module, configured to determine a service requirement classification result according to the current power service data and a pre-constructed improved deep forest model;

[0015] A node determination module, configured to determine an aggregation node according to the current link multi-dimensional state evaluation information, the current environment perception information, and the classification result of service requirements.

[0016] According to a third aspect of the present invention, there is provided an electronic device, the electronic device includes:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining an aggregation node of a power data communication network according to any embodiment of the present invention.

[0020] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method for determining an aggregation node of a power data communication network according to any embodiment of the present invention when executed.

[0021] According to a fifth aspect of the present invention, an embodiment of the present invention further provides a computer program product, the computer program product includes a computer program, and the computer program realizes the method for determining an aggregation node of a power data communication network according to any embodiment of the present invention when executed by a processor.

[0022] In the technical solution of the embodiment of the present invention, when the link state changes, the current link state information, the current environment perception information, and the current power service data are acquired; according to the current link state information, the current link multi-dimensional state evaluation information is determined; according to the current power service data and the pre-constructed improved deep forest model, the service requirement classification result is determined; according to the current link multi-dimensional state evaluation information, the current environment perception information, and the service requirement classification result, the aggregation node is determined. When the link state changes, the multi-dimensional current link multi-dimensional state evaluation information is determined, and then, in combination with the perception of the current environment and the current service requirement classification result, the aggregation node is determined. More comprehensive and accurate link state information is provided, providing a more reliable basis for the selection of the aggregation node, realizing the accurate classification of service requirements, and realizing the dynamic selection of the aggregation node to adapt to the changing service requirements and network conditions.

[0023] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0025] Figure 1 is a flowchart of a method for determining an aggregation node of a power data communication network according to Embodiment 1 of the present invention;

[0026] Figure 2 is a flowchart of a method for determining an aggregation node of a power data communication network according to Embodiment 2 of the present invention;

[0027] Figure 3 is an example flowchart of a method for determining an aggregation node of a power data communication network according to Embodiment 2 of the present invention;

[0028] Figure 4 is a schematic structural diagram of a device for determining an aggregation node of a power data communication network according to Embodiment 3 of the present invention;

[0029] Figure 5 is a schematic structural diagram of an electronic device for implementing the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0032] Embodiment 1

[0033] Figure 1 This is a flowchart of a method for determining a convergence node of a power data communication network provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of dynamically determining the convergence node of the power data communication network. This method can be executed by a determining device for the convergence node of the power data communication network. The determining device for the convergence node of the power data communication network can be implemented in the form of hardware and / or software, and the determining device for the convergence node of the power data communication network can be configured in an electronic device. As Figure 1 shown, the method includes:

[0034] S110. When the link state changes, obtain the current link state information, the current environmental perception information, and the current power service data.

[0035] In this embodiment, the link state can be understood as an index characterizing the link state, such as indexes including link bandwidth and delay. The current link state information can be understood as including indexes characterizing the link state in multiple dimensions. The current environmental perception information can be understood as monitoring information used to characterize changes in the network environment. The current power service data can be understood as the service data that needs to be transmitted currently.

[0036] Specifically, the processor can detect the link state. When it detects a change in the link state, the processor can obtain the current link state information, and obtain the current environmental perception information based on the monitoring of the current network environment. The processor can also obtain the current power service data that needs to be transmitted currently.

[0037] S120. Determine the current link multi-dimensional state evaluation information according to the current link state information.

[0038] In this embodiment, the current link multi-dimensional state evaluation information can be understood as the information used to evaluate the current link state.

[0039] Specifically, the processor can evaluate through a multi-dimensional evaluation system according to the actual link situation included in the current link state information, and obtain the current link multi-dimensional state evaluation information in multiple dimensions. For example, it can include the traditional link bandwidth, delay, and packet loss rate indicators, and can also introduce link stability and interference situation indicators.

[0040] S130. Determine the service demand classification result according to the current power service data and the pre-constructed improved deep forest model.

[0041] In this embodiment, the pre-constructed improved deep forest model can be understood as a deep forest model constructed based on historical data and with improvements in the decision-making link. The service demand classification result can be understood as the result used to represent the category to which the current power service data belongs.

[0042] Specifically, the processor can input the current power service data into the pre-constructed improved deep forest model, and the improved deep forest model can decide the service demand classification result to which the current power service data belongs.

[0043] S140. Determine the aggregation node according to the current link multi-dimensional state evaluation information, the current environmental perception information, and the service demand classification result.

[0044] In this embodiment, the aggregation node can be understood as the aggregation node adapted to the current link and service situation.

[0045] Specifically, the processor can determine the aggregation node through the deep reinforcement learning network algorithm. The processor can first determine whether the environment has changed according to the current environmental perception information, adjust the parameters of the deep reinforcement learning network algorithm when the environment changes, use the current link multi-dimensional state evaluation information and the service demand classification result together as the state space in the deep reinforcement learning network algorithm, set the actions and rewards of the deep reinforcement learning network algorithm, and the agent continuously optimizes its decision-making strategy through business transmission until the algorithm ends, so as to take the node with the highest score as the aggregation node.

[0046] The technical solution of the embodiment of the present invention is as follows: when the link state changes, obtain the current link state information, the current environmental perception information, and the current power service data; determine the current link multi-dimensional state evaluation information according to the current link state information; determine the service requirement classification result according to the current power service data and the pre-constructed improved deep forest model; determine the aggregation node according to the current link multi-dimensional state evaluation information, the current environmental perception information, and the service requirement classification result. When the link state changes, determine the multi-dimensional current link multi-dimensional state evaluation information, and then combine the perception of the current environment and the current service requirement classification result to determine the aggregation node. Provide more comprehensive and accurate link state information, provide a more reliable basis for the selection of the aggregation node, realize the accurate classification of service requirements, and realize the dynamic selection of the aggregation node to adapt to the changing service requirements and network conditions.

[0047] Further, on the basis of the above embodiment, the determination method of the current environmental perception information includes at least one of the following:

[0048] Determine the current environmental perception information by monitoring the network topology structure;

[0049] Determine the current environmental perception information by monitoring the access of new services and the withdrawal of existing services.

[0050] In this embodiment, the network topology structure can be understood as a structure used to represent the connection situation of network links.

[0051] Specifically, the processor can regularly collect the status information of network devices (such as routers and switches) through a network management protocol (such as the Simple Network Management Protocol, SNMP), including port status, link connection relationships, etc. When a new link is established, for example, a new high-speed fiber optic link is added, the processor will record the relevant parameters of the link, such as bandwidth and latency, and generate the current environment perception parameters. If a fault is detected at a certain node, resulting in a link interruption, the processor will promptly update the network topology, recalculate the reachable paths between each node, and form these changes into the current environment perception parameters. When docking with the service management system, the processor can obtain the access information of new services and the withdrawal information of existing services in real time. When a new service is added, the processor will analyze key information such as the type, traffic characteristics, and real-time requirements of the new service. For example, when a high-definition video live broadcast service is newly accessed, the processor will identify its high requirements for bandwidth and real-time performance and incorporate this information into the current environment perception parameters. For the withdrawal of existing services, the processor will also adjust the service demand distribution information accordingly, reducing the estimated demand for relevant link resources. It realizes the comprehensive and real-time monitoring of network environment changes, so as to be able to re-learn and make decisions based on the new environmental conditions in the future to adapt to the dynamically changing network environment.

[0052] Embodiment 2

[0053] Figure 2 The flowchart of a method for determining the aggregation node of a power data communication network provided by Embodiment 2 of the present invention. This embodiment is a further refinement of the above embodiment. As Figure 2 shown, the method includes:

[0054] S201. When the link state changes, obtain the current link state information, the current environment perception information, and the current power service data.

[0055] S202. Determine the current bandwidth evaluation value according to the link transmission sub-information in the current link state information.

[0056] In this embodiment, the link transmission sub-information can be understood as information related to link transmission, such as the total size of data packets transmitted by the link within a time window, etc. The current bandwidth evaluation value can be understood as a value used to reflect the data transmission capacity of the link within a specific time period.

[0057] Specifically, the processor can use the sliding window method to obtain the link transmission sub-information and determine the current bandwidth evaluation value based on the traffic statistics of the link transmission sub-information within the time window.

[0058] Exemplarily, within the time window T, the sum S of the sizes of the data packets transmitted by the link is statistically calculated, and the calculation formula for the current bandwidth evaluation value B is This model can reflect the data transmission capacity of the link within a specific time period. The selection of the time window T needs to be adjusted according to the characteristics of the network service. For services with high real-time requirements, T takes a smaller value to more timely reflect the bandwidth change; for services with large data volume transmission, T can be appropriately increased to obtain a more stable bandwidth evaluation.

[0059] S203. Determine the current delay evaluation value according to the link sending and receiving times in the current link status information.

[0060] In this embodiment, the link sending and receiving times can be understood as the moments when the probe packets are received and sent in the link. The current delay evaluation value can be understood as a value used to characterize the situation of the link sending data packets with delay.

[0061] Specifically, the processor can send probe packets with timestamps at both ends of the link and record the link sending and receiving times, including the sending time t send and the receiving time t recv , then the current delay evaluation value D is D = t recv -t send . Multiple measurements are averaged to improve accuracy. The purpose of averaging multiple measurements is to reduce the influence of accidental factors on delay measurement, such as momentary network congestion, device processing delay fluctuations, etc. In practical applications, the number of measurements needs to be determined according to the stability and accuracy requirements of the network. A network with poor stability requires more measurement times to ensure the reliability of the results.

[0062] S204. Determine the current packet loss rate according to the data packet sub-information in the current link status information.

[0063] In this embodiment, the data packet sub-information can be understood as information related to the data packets sent by the link. For example, it can include the total number of data packets sent and the total number of successfully received data packets. The current packet loss rate can be understood as a value used to reflect the transmission reliability of the link.

[0064] Specifically, the processor can statistically calculate that within a certain time, the total number of data packets sent by the link is N send , and the number of successfully received data packets is N recv . These are jointly used as the data packet sub-information, then the current packet loss rate L is The packet loss rate directly reflects the transmission reliability of the link. When the packet loss rate is too high, it indicates that there may be problems such as link failures, interference, or insufficient bandwidth. By monitoring the packet loss rate, link anomalies can be detected in a timely manner, providing a key basis for the selection of the aggregation node.

[0065] S205. Determine the current stability evaluation value according to the delay sub - information in the current link - state information.

[0066] In this embodiment, the delay sub - information can be understood as the analysis of the delay situation within a period of time, such as the delay and average delay in each measurement, etc. The current stability evaluation value can be understood as being used to characterize the fluctuation of delay or bandwidth in the link.

[0067] Specifically, the processor can use time - series analysis methods to analyze the delay or bandwidth data within a period of time to determine the delay sub - information, and then determine the current stability evaluation value based on the delay sub - information.

[0068] Exemplarily, the processor can calculate the current stability evaluation value of the delay data through the following formula, that is, the following standard deviation σ:

[0069]

[0070] where D i is the delay of the i - th measurement, is the average delay, n is the number of measurements. The smaller the standard deviation, the more stable the link. The standard deviation measures the degree of dispersion of the data. In the link - stability evaluation, it can reflect the fluctuation of delay or bandwidth. When the standard deviation is large, it indicates that the link state is unstable and may have an adverse impact on service transmission.

[0071] S206. Determine the current stability evaluation value according to the delay sub - information in the current link - state information.

[0072] In this embodiment, the interference - signal sub - information can be understood as being used to characterize the situation of interference signals around the link, such as including the interference - signal strength and frequency. The current interference evaluation value is used to reflect the interference intensity.

[0073] Specifically, the processor can obtain the interference - signal sub - information by monitoring the interference - signal strength I and frequency f around the link, and determine the current interference evaluation value in a weighted manner.

[0074] Exemplarily, the current interference evaluation value I eval can be calculated through the following formula:

[0075] I eval = w 1 I + w 2 f

[0076] where w 1 and w 2 are weights determined according to the actual situation. The determination of the weights needs to consider the characteristics of interference signals in the network environment. For example, in an environment where the interference - signal strength changes greatly, w 1can be appropriately increased; in areas where the interference frequency changes frequently, the weight of w 2 can be increased to more accurately evaluate the impact of interference on the link.

[0077] S207. Use the current bandwidth evaluation value, current delay evaluation value, current packet loss rate, current stability evaluation value, and current interference evaluation value as the current link multi-dimensional state evaluation information.

[0078] Specifically, the processor can use the current bandwidth evaluation value, current delay evaluation value, current packet loss rate, current stability evaluation value, and current interference evaluation value together as the current link multi-dimensional state evaluation information.

[0079] S208. Input the current power service data into a pre-constructed improved deep forest model to obtain a preliminary classification decision result.

[0080] In this embodiment, the preliminary classification decision result can be understood as the classification result determined by the classification decision rules in the deep forest model.

[0081] Specifically, the processor can input the current power service data into a pre-constructed improved deep forest model to obtain a preliminary classification decision result.

[0082] Based on the above embodiments, the construction of the improved deep forest model can be further elaborated as follows:

[0083] First, the processor can first collect service historical data, clean and preprocess the data, remove noise data, and perform missing value filling and data normalization. Perform feature fusion on the processed historical data. For features such as the real-time traffic, priority, and real-time requirements of the service in the input layer, use a weighted fusion method. Assign different weights according to the importance of the features for service classification. For example, in a service scenario with high real-time requirements, the weight of the real-time feature can be set to 0.5, the traffic weight to 0.3, and the priority weight to 0.2. The calculation formula is:

[0084]

[0085] where, F merged is the fused feature vector, w i is the weight of the i-th feature, f i is the value of the i-th feature, and n is the number of features.

[0086] During the hierarchical construction process, a business scenario analysis mechanism is introduced. According to different business types, such as video conferencing, file transfer, etc., targeted hierarchical structures are constructed respectively. For example, for the video conferencing business, due to its high requirements for real-time performance and bandwidth, feature branches related to real-time performance and bandwidth are given priority in the hierarchical structure. When constructing each layer of the forest, an adaptive tree number adjustment strategy is adopted. According to the classification accuracy rate of the previous layer, the number of decision trees in the forest of this layer is dynamically adjusted. If the accuracy rate of the previous layer is lower than the set threshold (such as 70%), the number of decision trees in this layer is increased to improve the learning ability of the model. Finally, the hierarchical construction is completed to form a complete improved deep forest structure. In the decision-making stage, in addition to the traditional majority voting mechanism, business expert knowledge is introduced to assist in decision-making. A business expert knowledge base is constructed to store classification decision rules under common business scenarios. When the decision result of the deep forest model is uncertain (such as the number of votes for different categories is close), the rules in the expert knowledge base are referred to for the final decision. The improved deep forest after construction is obtained.

[0087] S209. If there is uncertainty in the preliminary classification decision result, the final decision is made through the business expert knowledge base in the improved deep forest model to obtain the business requirement classification result.

[0088] In this embodiment, uncertainty can be understood as a situation where the number of votes for different categories is close. The business expert knowledge base can be understood as a rule base composed of human historical experience.

[0089] Specifically, when the number of votes for different categories in the preliminary classification decision result is close, it is considered that there is uncertainty. The processor can make the final decision by referring to the rules in the expert knowledge base to obtain the business requirement classification result.

[0090] S210. Otherwise, the preliminary classification decision result is used as the business requirement classification result.

[0091] Specifically, when there is no uncertainty in the preliminary classification decision result, that is, the number of votes for one type is far higher than other types, the processor can directly use the preliminary classification decision result as the business requirement classification result.

[0092] S211. When the current environmental perception information changes, the state space and expected value are adjusted based on the current environmental perception information.

[0093] In this embodiment, the state space can be understood as a set describing all possible states in the network environment decision-making process. The expected value can be understood as a convergence node used to predict a future time point or time period.

[0094] Specifically, first, the processor can initialize the weight vector, state space, and prediction value table. When the current environmental perception information changes, the processor can redefine the value range and initial values of each state variable in the state space according to the new network topology and service demand distribution in the current environmental perception information; for the prediction value table, the processor can reset it to the initial random value or the initial value set according to experience, so as to be able to re-learn and make decisions based on the new environmental conditions.

[0095] S212. Update the state space according to the current multi-dimensional link state evaluation information and the service demand classification result to obtain the target state space.

[0096] In this embodiment, the target state space can be understood as a state space that is more in line with the current actual link situation.

[0097] Specifically, the processor can update the state space according to the current multi-dimensional link state evaluation information and the service demand classification result to obtain the target state space.

[0098] Exemplarily, the target state space can be S = [B, D, L, S stab , I, C], where B is the bandwidth, representing the amount of data that the link can transmit per unit time, which directly affects the transmission efficiency of large data volume services. D is the delay, which refers to the time it takes for data to travel from the sender to the receiver. For services with high real-time requirements, such as video conferencing and online games, the size of the delay is crucial for the quality of the service experience. L is the packet loss rate, which reflects the proportion of lost data packets during transmission. An excessively high packet loss rate will seriously affect the reliability of the service. S stab is the stability, which is obtained by analyzing the fluctuation of link state parameters over a period of time. A stable link is crucial for continuous and stable service transmission. I is the interference situation, which reflects the degree of interference of the surrounding environment of the link on signal transmission. The stronger the interference, the more easily the effective transmission capacity of the link is affected. C is the service demand category. Different service demands, such as real-time communication, file transfer, data backup, etc., have different emphases on link state indicators.

[0099] S213. Take the adjustment of the link state indicator weights as the action adjustment strategy.

[0100] In this embodiment, the link state indicator weights can be understood as the weights set for each item in the above current multi-dimensional link state evaluation information. The action adjustment strategy can be understood as the basis for the action adjustment of the deep reinforcement learning network algorithm.

[0101] Specifically, the processor can take the adjustment of the link state indicator weights as the action adjustment strategy.

[0102] Exemplarily, the action is an adjustment strategy for the weights of link state metrics. Let the weight vector W of the link state metric weights be W = [w 1 , w 2 , w 3 , w 4 , w 5 , corresponding to the weights of bandwidth, latency, packet loss rate, stability, and interference respectively. The action is manifested as a fine-tuning of the weight vector, such as ΔW = [Δw 1 , Δw 2 , Δw 3 , Δw 4 , Δw 5 , and the adjusted weight W' = W + ΔW.

[0103] S214. Determine the aggregation node through the deep reinforcement learning network algorithm according to the action adjustment strategy, the target state space, and the expected value.

[0104] Specifically, by simulating the decision-making process of the intelligent agent in the network environment, the expected values corresponding to different actions can be determined based on the target state space through the deep reinforcement learning network algorithm, so as to achieve precise and dynamic adjustment of the weights of the link state metrics to adapt to the changing business requirements and network conditions. Until the algorithm ends, the node with the highest comprehensive score is taken as the aggregation node.

[0105] Further, on the basis of the above embodiments, the step of determining the aggregation node through the deep reinforcement learning network algorithm according to the action adjustment strategy, the target state space, and the expected value can be refined as follows:

[0106] Select the action adjustment strategy through the deep reinforcement learning network algorithm and adjust the weights of the link state metrics; perform service transmission with the updated weights of the link state metrics and determine the current reward signal; update the expected value with the current reward signal by the intelligent agent and perform iterative loops to continuously optimize the weights of the link state metrics until the algorithm ends and calculate the comprehensive scores of each node; take the node with the highest comprehensive score as the aggregation node.

[0107] In this embodiment, the current reward signal can be understood as reflecting the quality of this weight adjustment and service transmission effect. The intelligent agent can be understood as an entity that can perceive the environment and take autonomous actions to achieve specific goals. A node can be understood as a device that performs information transmission, processing, exchange, or control in the power data communication network. The comprehensive score can be understood as characterizing the transmission effect of the node.

[0108] Specifically, the processor can select an action adjustment strategy and adjust the link state metric weights through a deep reinforcement learning network algorithm; perform service transmission with the updated link state metric weights and determine the current reward signal; update the expected value with the current reward signal by the agent, and perform iterative loops to continuously optimize the link state metric weights until the algorithm ends and calculates the comprehensive scores of each node; select the node with the highest comprehensive score as the aggregation node.

[0109] Exemplarily, after obtaining the state space S, according to the deep reinforcement learning network algorithm, by calculating the predicted Q values corresponding to different actions, select the action ΔW from the action space, and this action represents the adjustment strategy for the weight vector W. After determining the action, actually adjust the weight vector and update W to W + ΔW. After adjusting the weights, perform service transmission based on the new weights and apply the adjusted weights to the actual service data transmission process. After the transmission is completed, evaluate the service transmission effect, and comprehensively calculate the current reward signal R based on key indicators such as service transmission success rate, average delay, and average packet loss rate. R reflects the quality of this weight adjustment and service transmission. Finally, update the Q value according to the obtained R and the update formula of the deep reinforcement learning network. The agent continuously optimizes its decision-making strategy based on this current reward signal to prepare for the next weight adjustment decision. The entire process is in continuous circulation, continuously optimizing the weights to adapt to the dynamically changing network environment and service requirements until the algorithm ends, calculating the comprehensive scores of each node, and selecting the node with the highest comprehensive score as the aggregation node.

[0110] The technical solution of the embodiment of the present invention combines traditional link bandwidth, delay, and packet loss rate indicators with newly introduced link stability and interference situation indicators by constructing a multi-dimensional evaluation system to determine the current multi-dimensional link state evaluation information. Compared with traditional single-index evaluation, multi-dimensional link state evaluation can provide more comprehensive and accurate link state information, provide a more reliable basis for the selection of the aggregation node, and effectively avoid service interruption or performance degradation caused by sudden link conditions. Through the improved deep forest algorithm, it classifies according to the characteristics of service data rather than simply based on functions, optimizes the feature fusion and hierarchical construction methods, and the dynamic classification can adapt to the changes in service requirements in real time, making the selection of the aggregation node more in line with the actual situation of the service, improving the rationality of network resource allocation, and ensuring the quality of service of different types of services. By dynamically perceiving the environmental situation, generating the current environmental perception information, and combining it with the deep reinforcement learning network algorithm, it realizes the accurate and dynamic adjustment of the link state metric weights to adapt to the changing service requirements and network conditions, and realizes the dynamic selection of the aggregation node of the power data communication network.

[0111] Exemplarily, a specific example is used to demonstrate the overall method of the present invention.Figure 3 The following is an example flowchart of a method for determining a convergence node of a power data communication network provided in the second embodiment of the present invention. As Figure 3 shown, the steps may be as follows:

[0112] S301. Initialize system parameters, including weight vector W, state space S, and Q-value table;

[0113] S302. Determine the current bandwidth evaluation value according to the link transmission sub-information in the current link state information;

[0114] S303. Determine the current delay evaluation value according to the link transceiver time in the current link state information;

[0115] S304. Determine the current packet loss rate according to the data packet sub-information in the current link state information;

[0116] S305. Determine the current stability evaluation value according to the delay sub-information in the current link state information;

[0117] S306. Determine the current stability evaluation value according to the delay sub-information in the current link state information;

[0118] S307. Obtain the current link multi-dimensional state evaluation information;

[0119] S308. Input the current power service data into a pre-constructed improved deep forest model to obtain a preliminary classification decision result;

[0120] S309. If there is uncertainty in the preliminary classification decision result, perform a final decision through the business expert knowledge base in the improved deep forest model to obtain a business requirement classification result;

[0121] S310. Otherwise, use the preliminary classification decision result as the business requirement classification result;

[0122] S311. Obtain the current environmental perception information;

[0123] S312. Determine whether the environment has changed; if so, jump to S313; if not, jump to S314;

[0124] S313. Adjust the state space S and Q-value table based on the current environmental perception information;

[0125] S314. Use the current link multi-dimensional state evaluation information and business requirement classification result as the target ratio state space S to provide a basis for subsequent decisions;

[0126] S315. Select an action ΔW according to the deep reinforcement learning network algorithm, and determine the optimal adjustment strategy by calculating the Q-values of different actions;

[0127] S316. Adjust the weight vector W’ = W + ΔW, actually adjust the weights according to the selected actions, determine the reward signal, and continuously optimize its own decision-making strategy until the algorithm ends;

[0128] S317. Calculate the comprehensive scores of each node, and select the node with the highest comprehensive score as the aggregation node.

[0129] Embodiment III

[0130] Figure 4 It is a schematic structural diagram of a device for determining an aggregation node of a power data communication network provided in Embodiment III of the present invention. As Figure 4 shown, the device includes:

[0131] An information acquisition module 41, configured to acquire the current link status information, the current environment perception information, and the current power service data when the link status changes;

[0132] An information determination module 42, configured to determine the current link multi-dimensional state evaluation information according to the current link status information;

[0133] A result determination module 43, configured to determine the service demand classification result according to the current power service data and a pre-constructed improved deep forest model;

[0134] A node determination module 44, configured to determine the aggregation node according to the current link multi-dimensional state evaluation information, the current environment perception information, and the service demand classification result.

[0135] The technical solution of the embodiment of the present invention is as follows: when the link status changes, acquire the current link status information, the current environment perception information, and the current power service data; determine the current link multi-dimensional state evaluation information according to the current link status information; determine the service demand classification result according to the current power service data and a pre-constructed improved deep forest model; determine the aggregation node according to the current link multi-dimensional state evaluation information, the current environment perception information, and the service demand classification result. When the link status changes, determine the multi-dimensional current link multi-dimensional state evaluation information, and then combine the perception of the current environment and the current service demand classification result to determine the aggregation node. Provide more comprehensive and accurate link status information, provide a more reliable basis for the selection of the aggregation node, realize the accurate classification of service demands, realize the dynamic selection of the aggregation node, so as to adapt to the changing service demands and network conditions.

[0136] Further, the information determination module 42 is specifically configured to:

[0137] Determine the current bandwidth evaluation value according to the link transmission sub-information in the current link status information;

[0138] Determine the current delay evaluation value according to the link transceiver time in the current link state information;

[0139] Determine the current packet loss rate according to the data packet sub - information in the current link state information;

[0140] Determine the current stability evaluation value according to the delay sub - information in the current link state information;

[0141] Determine the current interference evaluation value according to the interference signal sub - information in the current link state information;

[0142] Take the current bandwidth evaluation value, the current delay evaluation value, the current packet loss rate, the current stability evaluation value and the current interference evaluation value as the current link multi - dimensional state evaluation information.

[0143] Further, the result determination module 43 is specifically used for:

[0144] Input the current power service data into a pre - constructed improved deep forest model to obtain a preliminary classification decision result;

[0145] If there is uncertainty in the preliminary classification decision result, perform a final decision through the business expert knowledge base in the improved deep forest model to obtain a business requirement classification result;

[0146] Otherwise, take the preliminary classification decision result as the business requirement classification result.

[0147] Further, the node determination module 44 includes:

[0148] A state adjustment unit, configured to adjust the state space and the expected value based on the current environment perception information when the current environment perception information changes;

[0149] A space update unit, configured to update the state space according to the current link multi - dimensional state evaluation information and the business requirement classification result to obtain a target state space;

[0150] A policy determination unit, configured to take the adjustment of the link state index weight as an action adjustment policy;

[0151] A node determination unit, configured to determine the aggregation node through a deep reinforcement learning network algorithm according to the action adjustment policy, the target state space and the expected value.

[0152] Among them, the node determination unit is specifically used for:

[0153] Through the deep reinforcement learning network algorithm, select the action adjustment policy and adjust the link state index weight;

[0154] Perform service transmission with the updated link state metric weights and determine the current reward signal;

[0155] The agent updates the expected value with the current reward signal and performs iterative loops to continuously optimize the link state metric weights until the algorithm ends and calculates the comprehensive scores of each node;

[0156] Take the node with the highest comprehensive score as the aggregation node.

[0157] Optionally, the device further includes: an environment perception module.

[0158] The environment perception module is used to determine the current environment perception information, and the determination method of the current environment perception information includes at least one of the following:

[0159] Determine the current environment perception information by monitoring the network topology structure;

[0160] Determine the current environment perception information by monitoring the access of new services and the withdrawal of existing services.

[0161] The device for determining the aggregation node of the power data communication network provided by the embodiments of the present invention can execute the method for determining the aggregation node of the power data communication network provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0162] Embodiment 4

[0163] Figure 5 FIG. shows a schematic structural diagram of an electronic device 50 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0164] As Figure 5As shown, the electronic device 50 includes at least one processor 51 and a memory communicatively connected to the at least one processor 51, such as a read-only memory (ROM) 52, a random access memory (RAM) 53, etc. The memory stores a computer program executable by the at least one processor. The processor 51 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 52 or the computer program loaded from the storage unit 58 into the random access memory (RAM) 53. In the RAM 53, various programs and data required for the operation of the electronic device 50 can also be stored. The processor 51, the ROM 52, and the RAM 53 are connected to each other via a bus 54. An input / output (I / O) interface 55 is also connected to the bus 54.

[0165] Multiple components in the electronic device 50 are connected to the I / O interface 55, including: an input unit 56, such as a keyboard, a mouse, etc.; an output unit 57, such as various types of displays, speakers, etc.; a storage unit 58, such as a magnetic disk, an optical disc, etc.; and a communication unit 59, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 59 allows the electronic device 50 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0166] The processor 51 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 51 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 51 executes the various methods and processes described above, such as the method for determining a convergence node of a power data communication network.

[0167] In some embodiments, the method for determining a convergence node of a power data communication network can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 58. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 50 via the ROM 52 and / or the communication unit 59. When the computer program is loaded into the RAM 53 and executed by the processor 51, one or more steps of the method for determining a convergence node of a power data communication network described above can be executed. Alternatively, in other embodiments, the processor 51 can be configured to execute the method for determining a convergence node of a power data communication network by any other appropriate means (e.g., by means of firmware).

[0168] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0169] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0170] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0171] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0172] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0173] The computing system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0174] In one embodiment, the embodiment of the present invention further includes a computer program product, which includes a computer program that, when executed by a processor, implements the method for determining a convergence node of a power data communication network according to any embodiment of the present invention.

[0175] In the process of implementing the computer program product, computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any kind of network - including a local area network (LAN) or a wide area network (WAN) - or, it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0176] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0177] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for determining a convergence node of a power data communication network, characterized in that: include: When the link status changes, obtain the current link status information, current environment perception information and current power business data; Determine current link multi-dimensional state evaluation information according to the current link state information; Determine a business demand classification result according to the current power business data and the pre-built improved deep forest model; A convergence node is determined according to the current link multi-dimensional state evaluation information, the current environment perception information and the business demand classification result.

2. The method according to claim 1, characterized in that The step of determining the current link multi-dimensional state evaluation information according to the current link state information includes: Determine a current bandwidth evaluation value according to the link transmission sub-information in the current link state information; Determine a current delay evaluation value according to the link receiving and sending time in the current link state information; Determining a current packet loss rate according to the data packet sub-information in the current link state information; Determining a current stability evaluation value according to the delay sub-information in the current link state information; Determining a current interference assessment value according to the interference signal sub-information in the current link state information; The current bandwidth evaluation value, the current delay evaluation value, the current packet loss rate, the current stability evaluation value and the current interference evaluation value are used as current link multi-dimensional state evaluation information.

3. The method according to claim 1, characterized in that Determining the business demand classification result according to the current power business data and the pre-built improved deep forest model includes: Inputting the current power business data into a pre-built improved deep forest model to obtain a preliminary classification decision result; If there is uncertainty in the preliminary classification decision result, a final decision is made through the business expert knowledge base in the improved deep forest model to obtain a business demand classification result; Otherwise, the preliminary classification decision result is used as the business demand classification result.

4. The method according to claim 1, characterized in that: The current environment perception information is determined in a manner including at least one of the following: By monitoring the network topology, the current environmental perception information is determined; By monitoring the entry of new services and the exit of existing services, the current environmental perception information is determined.

5. The method according to claim 1, characterized in that The determining of the aggregation node according to the current link multi-dimensional state evaluation information, the current environment perception information and the business demand classification result includes: When the current environmental perception information changes, adjusting the state space and the expected value based on the current environmental perception information; According to the current link multi-dimensional state evaluation information and the business demand classification result, the state space is updated to obtain a target state space; The adjustment of the link status indicator weight is used as an action adjustment strategy; According to the action adjustment strategy, the target state space and the expected value, a convergence node is determined through a deep reinforcement learning network algorithm.

6. The method according to claim 5, characterized in that The step of determining a convergence node by a deep reinforcement learning network algorithm according to the action adjustment strategy, the target state space, and the expected value includes: Selecting the action adjustment strategy and adjusting the link status indicator weight through a deep reinforcement learning network algorithm; The service is transmitted through the updated link status indicator weight, and the reward signal is determined; The expected value is updated by the agent using the current reward signal, and the link status indicator weight is continuously optimized in an iterative cycle until the algorithm finishes calculating the comprehensive score of each node; The node with the highest comprehensive score is taken as the sink node.

7. A device for determining a convergence node of a power data communication network, characterized in that: include: An information acquisition module is used to obtain current link status information, current environment perception information and current power business data when the link status changes; An information determination module, used to determine current link multi-dimensional state evaluation information according to the current link state information; A result determination module, used to determine the business demand classification result according to the current power business data and the pre-built improved deep forest model; The node determination module is used to determine the aggregation node according to the current link multi-dimensional state evaluation information, the current environment perception information and the business demand classification result.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining an aggregation node of a power data communication network according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining a power data communication network aggregation node according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements the method for determining a power data communication network aggregation node according to any one of claims 1 to 6.

Images