CN-121984217-A - Intelligent operation and maintenance monitoring method and system for power distribution network

Abstract

The invention provides an intelligent operation and maintenance monitoring method and system of a power distribution network, and relates to the technical field of operation and maintenance of power distribution networks, wherein the method comprises the steps of continuously collecting operation characteristic data of all branches of the power distribution network in a preset time window, and predicting whether all branches of the power distribution network are overloaded or not based on the operation characteristic data; if the power distribution network is predicted to have the branch overload, judging whether the overload branch can carry out bypass branch power distribution compensation or not; and if bypass branch distribution compensation can be performed, performing branch distribution decision. The method and the system realize the dynamic pre-judgment of the running state of the power distribution network, avoid the problem of timely predicting and finding the overload of the branch of the power distribution network under the condition of saving the operation and maintenance monitoring resources of the power distribution network and avoid the problem of repairing the east wall and the west wall.

Inventors

• ZHU YING
• SUN CHANGWEN
• WU MINGFENG
• WU XIAOJUN
• ZHANG GUANGWEI
• XIA YANHUI
• ZHANG YING
• SHEN DONGMING
• WU MING
• XU BINGYAN
• GUO LEI
• YAO WEI
• ZHAO YINGYING
• GAO SONGYAO

Assignees

• 国网山西省电力有限公司太原供电分公司
• 国网山西省电力有限公司电力科学研究院
• 国网上海市电力公司金山供电公司
• 华东电力试验研究院有限公司

Dates

Publication Date

20260505

Application Date

20260409

Claims (10)

1. 1. An intelligent operation and maintenance monitoring method for a power distribution network is characterized by comprising the following steps: Continuously collecting operation characteristic data of each branch of a power distribution network in a preset time window, and predicting whether each branch of the power distribution network is overloaded or not based on the operation characteristic data; If the power distribution network is predicted to have the branch overload, judging whether the overload branch can carry out bypass branch power distribution compensation or not; And if bypass branch distribution compensation can be performed, performing branch distribution decision.
2. 2. The intelligent operation and maintenance monitoring method of a power distribution network according to claim 1, wherein continuously collecting operation characteristic data of each branch of the power distribution network within a preset time window, predicting whether each branch of the power distribution network is overloaded based on the operation characteristic data, comprises: Acquiring historical operation characteristic data of each branch of the power distribution network, wherein the historical operation characteristic data is [ time stamp, branch ID, current value, active power, temperature and voltage ]; If the current value in the historical operation characteristic data is continuously greater than or equal to a rated value in a preset time window, marking the classification label of the operation characteristic data of the branch at the moment as overload=1, otherwise, marking the classification label as overload=0, and generating the historical operation characteristic data [ time stamp, branch ID, current value, active power, temperature, voltage, classification label and overload ] with the classification label; Extracting samples from the tagged historical operation characteristic data of each branch of the power distribution network, and constructing a sample set based on the samples, wherein each sample in the sample set comprises two output targets, the output targets comprise classification tags and overload amounts, a multi-task neural network model is trained according to the sample set to obtain an overload prediction model, and the sample set is divided into a training set, a verification set and a test set; And acquiring real-time operation characteristic data of each branch of the power distribution network, and inputting the real-time operation characteristic data into the overload prediction model to obtain whether each branch of the power distribution network is overloaded and the overload amount.
3. 3. The intelligent operation and maintenance monitoring method for the power distribution network according to claim 2, wherein the multi-task neural network model comprises a common feature extraction layer, a classification branch for outputting classification labels and a regression branch for outputting overload, and the performance weights of the classification branch and the regression branch are adjusted simultaneously according to the verification set.
4. 4. The intelligent operation and maintenance monitoring method for a power distribution network according to claim 2, wherein real-time operation characteristic data is input into the overload prediction model for prediction based on a prediction period to obtain whether each branch of the power distribution network is overloaded and the overload amount, the prediction period is calculated according to the following manner, ; Wherein, the method comprises the steps of, The prediction period is represented by a time period, Representing a preset maximum predicted period of time, Representing a preset minimum prediction period of time, A weight coefficient representing the load fluctuation index, Representing a load volatility index; Representing the real-time load of the branch at the current moment, Representing the historical load of the last sample period branch, Indicating the rated load of the branch circuit, The standard deviation of the load forward direction is represented, and K represents the load approximation coefficient.
5. 5. The intelligent operation and maintenance monitoring method of a power distribution network according to claim 4, wherein the load approximation coefficients are calculated according to the following manner, ; Wherein K represents a load approximation coefficient, Indicating that a low risk threshold is preset, Representing a preset high risk threshold value, Representing a preset risk enlargement factor threshold.
6. 6. The intelligent operation and maintenance monitoring method of a power distribution network according to claim 1, wherein if it is predicted that the power distribution network has a branch overload, determining whether the overloaded branch can perform bypass branch power distribution compensation comprises: If the power distribution network is predicted to have branch overload, judging whether the overloaded branch is physically communicated with the adjacent branch or not by checking whether a closed tie switch exists in the SCADA topological graph of the power distribution network; If adjacent branches are physically communicated, judging that the overloaded branch can perform bypass branch power distribution compensation; And if no adjacent branches are physically communicated, reporting real-time operation characteristic data of the overload branch of the power distribution network to a dispatching center.
7. 7. The intelligent operation and maintenance monitoring method of a power distribution network according to claim 2, wherein if bypass branch power distribution compensation is enabled, making a branch power distribution decision comprises: if the adjacent branches are in physical communication, predicting whether the adjacent branches are overloaded or not through the overload prediction model, and if the adjacent branches are not overloaded, dividing the adjacent branches into qualified compensation branch pools; judging whether the total safety margin of all qualified compensation branches in the qualified compensation branch pool is larger than or equal to an overload notch of the overload branch; If the total safety margin is greater than or equal to the overload notch of the overload branch, calculating the unit transfer cost of each branch in the qualified compensation branch pool, preferentially distributing the overload notch to a first branch with the lowest cost, if the safety margin of the first branch is smaller than the overload notch of the overload branch, continuously distributing the overload notch to a second branch when the safety margin of the first branch is consumed, wherein the unit transfer cost of the second branch is only greater than the unit transfer cost of the first branch; and if the total safety margin is smaller than the overload notch of the overload branch, carrying out comprehensive compensation through the qualified compensation branch and the dispatching center.
8. 8. The intelligent operation and maintenance monitoring method of a power distribution network according to claim 7, wherein if the total safety margin is smaller than an overload gap of an overload branch, comprehensive compensation is performed by a qualified compensation branch and a dispatching center, comprising: sorting qualified compensation branches according to unit cost from low to high, and enabling each branch to bear the maximum safety margin; And then, reporting overload branch information and qualified compensation branch compensation records to a dispatching center, wherein the overload branch information comprises a branch ID and a current overload gap.
9. 9. The intelligent operation and maintenance monitoring method of a power distribution network according to claim 1 or 7, wherein if bypass branch distribution compensation is enabled, making a branch distribution decision, further comprising: if a certain branch is simultaneously distributed to a plurality of overload branches for compensation, calculating the resource fitness score of the overload branches and the branch; And allocating the branch to the branch with the highest resource adaptation degree score.
10. 10. An intelligent operation and maintenance monitoring system for a power distribution network, the system comprising: the prediction overload module is used for continuously collecting operation characteristic data of each branch of the power distribution network in a preset time window and predicting whether each branch of the power distribution network is overloaded or not based on the operation characteristic data; the judging and compensating module is used for judging whether the overloaded branch can carry out bypass branch distribution compensation if the power distribution network is predicted to have the branch overloaded; and the decision module is used for making branch distribution decisions if bypass branch distribution compensation can be performed.

Description

Intelligent operation and maintenance monitoring method and system for power distribution network Technical Field The invention relates to the technical field of operation and maintenance of power distribution networks, in particular to an intelligent operation and maintenance monitoring method and system of a power distribution network. Background In the conventional power distribution network operation scheduling process, when a certain branch is close to an overload threshold value, an immediate load transfer strategy is generally adopted for scheduling decision. According to the strategy, the load level of each branch is monitored in real time, and the load of an overload risk branch is temporarily transferred to a current light-load branch, so that the local overload pressure is rapidly relieved. However, the scheduling method based on static load distribution has the remarkable limitation that a decision mechanism only considers the load state at the current moment, and the dynamic evolution pre-judging capability of the running situation of the power distribution network is lacking. If the selected light load branch enters an overload state in a future period due to the increase of local load, overload risks are transferred among the branches, a linkage overload event is caused, and the selected light load branch needs other line rescue, so that frequent operation of a switch, shortened service life of equipment and surge of cost are caused. The problems of frequent action of the circuit breaker, aggravation of the loss of the mechanical life of the equipment, rising of the operation and maintenance cost and the like are caused. In order to solve the above problems, the prior art attempts to introduce a load prediction technique to optimize the scheduling decision by pre-judging the future load variation trend of the light load branch. However, the existing load prediction scheme cannot be adaptively adjusted according to actual operation characteristics of the power distribution network. Disclosure of Invention Therefore, the invention aims to solve the problems that in the prior art, the operation and maintenance monitoring resources are wasted and the overload of a branch of a power distribution network cannot be predicted in time due to the fact that the power distribution network scheduling lacks the dynamic pre-judging capability of the operation state of the power distribution network and the prediction period cannot be dynamically adjusted according to the operation condition of the power distribution network, and provides an intelligent operation and maintenance monitoring method and system for the power distribution network, so that the dynamic pre-judging capability of the operation state of the power distribution network is realized, the problem that the overload of the branch of the power distribution network is predicted and found in time under the condition of saving operation and maintenance monitoring resources of the power distribution network is avoided. In order to solve the technical problems, the invention provides an intelligent operation and maintenance monitoring method for a power distribution network, which comprises the following steps: Continuously collecting operation characteristic data of each branch of a power distribution network in a preset time window, and predicting whether each branch of the power distribution network is overloaded or not based on the operation characteristic data; If the power distribution network is predicted to have the branch overload, judging whether the overload branch can carry out bypass branch power distribution compensation or not; And if bypass branch distribution compensation can be performed, performing branch distribution decision. Preferably, continuously collecting operation characteristic data of each branch of the power distribution network within a preset time window, predicting whether each branch of the power distribution network is overloaded based on the operation characteristic data, and including: Acquiring historical operation characteristic data of each branch of the power distribution network, wherein the historical operation characteristic data is [ time stamp, branch ID, current value, active power, temperature and voltage ]; If the current value in the historical operation characteristic data is continuously greater than or equal to a rated value in a preset time window, marking the classification label of the operation characteristic data of the branch at the moment as overload=1, otherwise, generating the historical operation characteristic data with the classification label [ time stamp, branch ID, current value, active power, temperature, voltage, classification label and overload ] with the classification label; Extracting samples from the tagged historical operation characteristic data of each branch of the power distribution network, and constructing a sample set based on the samples, wherein each sample in t