CN-122001082-A - Intelligent monitoring method, system, equipment and medium for power distribution network with multipath adaptive communication switching

CN122001082ACN 122001082 ACN122001082 ACN 122001082ACN-122001082-A

Abstract

The invention relates to the technical field of power system power distribution network monitoring, and discloses a method, a system, equipment and a medium for intelligent monitoring of a power distribution network with multipath adaptive communication switching. The system monitors the state of each link in real time, triggers millisecond-level fast switching based on a Markov decision process when quality degradation or prediction fault risk is detected, and avoids multi-node switching conflict by utilizing a distributed cooperative mechanism. Through resource pooling management and service priority scheduling, the reliability of high-priority services such as remote control and remote signaling is ensured, and meanwhile, the safe and stable operation of the power distribution network is effectively ensured.

Inventors

LI QINGSHENG
ZHANG YAN
Fan Junqiu
ZHANG ZHAOFENG
YANG JIERUI
YU WANRONG
LUO NING
LI ZHEN
LONG JIAHUAN
CHEN JULONG
ZHANG YU
WANG XIANG
TANG XUEYONG
Luo Zhongke
GAO HUA

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260508
Application Date: 20251203

Claims (10)

1. The intelligent monitoring method for the power distribution network with the multipath adaptive communication switching is characterized by comprising the following steps of, Configuring parameters of a multipath communication link and priority of monitoring service of a power distribution network, and initializing a communication link state and an evaluation model; the method comprises the steps of collecting performance parameters of communication links in parallel, and predicting availability, failure rate and residual life of each communication link based on historical data and current performance parameters; Selecting an optimal communication path through a path selection algorithm according to the current state, the prediction reliability and the service priority of the communication link; Triggering and executing fast switching to the standby path when the communication link quality is lower than a threshold value or the reliability prediction indicates a fault risk; and coordinating switching actions among the plurality of power distribution monitoring nodes, and dynamically distributing and optimizing communication resources according to the service load and the link state.
2. The intelligent monitoring method for the power distribution network with multipath adaptive communication switching as claimed in claim 1, wherein the step of parallel acquisition of the performance parameters of each communication link comprises the steps of carrying out normalization processing and weighting calculation on the acquired performance parameters through a multidimensional parameter comprehensive evaluation method to obtain the comprehensive quality score of each communication link.
3. The intelligent monitoring method for the power distribution network with multipath adaptive communication switching as claimed in claim 2, wherein the predicting the availability, the failure rate and the residual life of each communication link comprises calculating the availability and the failure rate of the link based on historical link state data through a sliding window method and an exponential weighted moving average algorithm, and combining a residual life prediction model to obtain a link health evaluation result.
4. The intelligent monitoring method for the power distribution network for switching multi-path adaptive communication according to claim 3, wherein said selecting an optimal communication path comprises modeling a path selection problem through a Markov decision process, defining a state space, an action space and a reward function, and solving an optimal path selection strategy by using a reinforcement learning algorithm to obtain an optimal communication path suitable for service requirements.
5. The intelligent monitoring method for the power distribution network with multipath adaptive communication switching as claimed in claim 4, wherein the reinforcement learning algorithm is a deep Q network, wherein a Q value function is approximated by a neural network, training is performed by using an empirical playback buffer zone and a target network, a loss function is defined as a mean square error, and network parameters are updated by a gradient descent method to optimize a path selection strategy; adopting a depth Q network DQN optimization path selection strategy, wherein the Q value function is approximately as follows: in the formula, A neural network approximation of the Q function, As a function of the parameters of the network, As a matrix of weights, the weight matrix, Is a state-motion feature vector; The loss function is defined as: in the formula, For the loss function, D is the empirical playback buffer, Is a target network parameter; The parameter updating adopts a gradient descent method: in the formula, In order to update the parameters of the data, In order for the rate of learning to be high, Is a gradient operator; exploration strategy adoption -Greedy approach: in the formula, For the action selected at the moment t, Is the exploration rate.
6. The intelligent monitoring method for the power distribution network of the multipath adaptive communication switching system of claim 5, wherein the switching to the standby path comprises the steps of judging switching conditions based on the quality and the health degree of a current link, estimating the total time delay and the required data buffer capacity of the switching process, estimating the switching success rate, and adopting a data retransmission mechanism and forward error correction coding in the switching process; The handover decision triggering condition is defined as: in the formula, In order to trigger the signal, For the current link quality, As a quality threshold value, For the current link health level, Is the minimum health requirement; the switching delay estimation model is as follows: in the formula, For the total switching time delay to be the same, In order to provide a time delay for the fault detection, For the decision to calculate the time delay, In order to switch the execution time delay of the switch, For the handover acknowledgement delay.
7. The intelligent monitoring method for the power distribution network of the multipath adaptive communication switching system of claim 6, wherein the dynamically allocating and optimizing the communication resources comprises realizing the state synchronization of each node through a distributed consensus algorithm, ensuring the state consistency of the multi-node cooperative switching system through the distributed consensus algorithm, wherein the state synchronization equation is that, In the formula, Is the state vector at the moment of node i, As a step-size parameter, For the set of neighbors of node i, Is an adjacency matrix element; The time synchronization precision requirement of cooperative switching is as follows: in the formula, For the time offset of node i and node j, And The local times of the two nodes respectively, Is the maximum allowable synchronization error; The switching coordination mechanism adopts a token passing algorithm, and the token holding time is limited as follows: in the formula, The time for which the token is held for node i, For a maximum allowable holding time, For the period of the token cycle, Is the total number of nodes.
8. The intelligent monitoring system for the power distribution network with the multipath adaptive communication switching, which is applied to the intelligent monitoring method for the power distribution network with the multipath adaptive communication switching according to any one of claims 1-7, is characterized by comprising a link monitoring module, a health prediction module, an intelligent decision module, a switching execution module and a collaborative management module; the link monitoring module is used for collecting and evaluating the performance parameters of each communication link in real time and calculating a unified link quality score; The health prediction module predicts the future reliability of the link based on the history and real-time data, generates link health assessment and realizes early warning of faults; the intelligent decision module synthesizes the link quality, the health and the service requirement, and decides the current optimal communication path by using an intelligent algorithm; the switching execution module accurately triggers and efficiently executes path switching operation, manages data buffering and retransmission, and ensures continuity and integrity of data transmission in the switching process; the collaborative management module coordinates switching actions among multiple nodes of the system, avoids conflict and globally manages all communication resources.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the intelligent monitoring method for a power distribution network for multi-path adaptive communication switching according to any one of claims 1 to 7.
10. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor performs the steps of the intelligent monitoring method for a power distribution network for multi-path adaptive communication switching according to any one of claims 1 to 7.

Description

Intelligent monitoring method, system, equipment and medium for power distribution network with multipath adaptive communication switching Technical Field The invention relates to the technical field of monitoring of power distribution networks of power systems, in particular to an intelligent monitoring method, system, equipment and medium for a power distribution network with multipath adaptive communication switching. Background With the deep promotion of intelligent power distribution network construction, the requirement of a power distribution network monitoring system on communication reliability is increasingly improved. The traditional power distribution network communication system mainly adopts a single communication mode, such as optical fiber private network or power carrier communication, and has a plurality of technical defects. When a single communication link faces fiber breakage, base station fault or carrier interference, the continuity of data transmission cannot be guaranteed, so that a monitoring system is invalid, and the safe operation of a power distribution network is seriously affected. The existing system lacks the capability of real-time monitoring and dynamic evaluation of communication quality, cannot early warn the performance degradation of a communication link, often carries out manual intervention after the occurrence of a fault, and delays the time of fault processing. Disclosure of Invention In view of the existing problems, the invention provides a method, a system, equipment and a medium for intelligently monitoring a power distribution network by switching multi-path adaptive communication. Therefore, the intelligent monitoring system aims at the technical problems of high single-point fault risk, slow switching response, difficult quality evaluation and the like of the power distribution network communication system, integrates various communication modes and has self-adaptive switching capability. The method comprises the steps of establishing a multidimensional communication quality evaluation model, realizing real-time monitoring and prediction of the performance of each communication link, and adopting a deep reinforcement learning algorithm to optimize a path selection strategy so as to realize millisecond-level rapid switching response. The intelligent monitoring method for the power distribution network comprises the steps of configuring parameters of multipath communication links and priority of monitoring service of the power distribution network, initializing a communication link state and an evaluation model, collecting performance parameters of all communication links in parallel, predicting availability, failure rate and residual service life of all communication links based on historical data and current performance parameters, selecting an optimal communication path through a path selection algorithm according to the current state, prediction reliability and service priority of the communication links, triggering and executing rapid switching to a standby path when the quality of the communication links is lower than a threshold value or the reliability prediction indicates failure risk, coordinating switching actions among a plurality of power distribution monitoring nodes, and dynamically distributing and optimizing communication resources according to service loads and link states. The intelligent monitoring method for the power distribution network with the multipath adaptive communication switching is used for realizing the optimal scheme, wherein the parallel acquisition of the performance parameters of each communication link comprises the steps of carrying out normalization processing and weighting calculation on the acquired performance parameters through a multidimensional parameter comprehensive evaluation method to obtain the comprehensive quality score of each communication link. The intelligent monitoring method for the power distribution network with the multipath adaptive communication switching comprises the steps of predicting the availability, the failure rate and the residual life of each communication link, calculating the availability and the failure rate of the link based on historical link state data through a sliding window method and an exponential weighted moving average algorithm, and combining a residual life prediction model to obtain a health evaluation result of the link. The optimal communication path selection method comprises the steps of modeling a path selection problem through a Markov decision process, defining a state space, an action space and a reward function, and solving an optimal path selection strategy by using a reinforcement learning algorithm to obtain an optimal communication path adapting to business requirements. The invention relates to a method for intelligently monitoring a power distribution network with multipath adaptive communication switching, which is a preferable scheme, wherein the reinfor