CN-122026596-A - Topology identification system for overhauling power distribution network

CN122026596ACN 122026596 ACN122026596 ACN 122026596ACN-122026596-A

Abstract

The invention relates to a topology identification system for power distribution network maintenance, which is characterized in that a data acquisition module and a data preprocessing module are matched to acquire and preprocess operation data of a power distribution network so as to form a high-quality data set, a topology inference module is used for obtaining a power distribution network topology identification model through training, an optimization module is used for carrying out deep analysis on a mechanism of the power distribution network topology identification model, an algorithm confidence evaluation system is used for enhancing the accuracy and reliability of the power distribution network topology identification model, a topology generation module is used for generating an operation boundary topological graph related to a maintenance area based on the high-quality data set and the power distribution network topology identification model, and then the same of graph-model-site is realized through the operation boundary topological graph, so that misjudgment caused by inconsistent graph models is fundamentally avoided, and the safety of the maintenance process is improved.

Inventors

GU GUOZHI
MA XIAOQING
SONG XIAOBO
WANG YABING
CHEN BIN
SHEN CHAOQUN
BI LIANG
JIANG ZIYU
SUN YILONG
WANG WEIMIN
XU MING
YANG ZHONGLI
SU TAO

Assignees

国网河南省电力公司商丘供电公司

Dates

Publication Date: 20260512
Application Date: 20251204

Claims (6)

1. The utility model provides a distribution network overhauls uses topology identification system, includes data acquisition module, data preprocessing module, feature extraction module, topology reasoning module, optimization module, topology generation module and visual output module, its characterized in that: The data acquisition module consists of a plurality of wireless sensors, an intelligent monitoring unit and a time marking unit, wherein the wireless sensors are deployed at the nodes of a power distribution network line and a transformer and are used for acquiring voltage, current, power factor and phase data of power distribution network equipment, the intelligent monitoring unit is used for acquiring switching state data of a circuit breaker and a disconnecting switch in real time, and the time marking unit is used for time marking the data acquired by each data acquisition module, so that consistency of the data time stamps is ensured; The data preprocessing module is connected with the data acquisition module through a wireless communication assembly and is used for cleaning, filtering, normalizing and fusing the data acquired by the data acquisition module so as to form a high-quality data set; The characteristic extraction module is used for extracting characteristic parameters of the operation of the power distribution network nodes; The topology reasoning module can construct an initial topology model of the power distribution network based on the operation characteristic parameters of the power distribution network nodes extracted by the characteristic extraction module, and then receives perception data from the scene, takes the perception data as a topology verification reference with the highest priority, trains the initial topology model of the power distribution network, and further obtains a topology identification model of the power distribution network; The optimization module can carry out deep analysis on the mechanism of the power distribution network topology identification model based on a transfer learning technology, proposes a high-availability parameter identification algorithm suitable for complex sampling conditions, and then constructs an algorithm confidence assessment system through verification set performance monitoring, model correction and hidden mode extraction technologies in a supervision training process, so that the accuracy and reliability of the power distribution network topology identification model are enhanced; The topology generation module can input the high-quality data set into a power distribution network topology identification model so as to generate the current real topology of the power distribution network, and then automatically generate an operation boundary topology map related to the maintenance area according to a maintenance work plan ticket and combining the current real topology; The visual output module can be used for displaying the operation boundary topological graph and simultaneously planning and marking the access path of the maintenance area.
2. The topology identification system for power distribution network maintenance according to claim 1, wherein in the data preprocessing process, abnormal values and null values are removed by data cleaning, the influence of electromagnetic interference and impulse noise on data is removed by data filtering, the incoordination caused by different dimensions and sizes among data characteristics is eliminated by data normalization, and multi-source data fusion is realized by adopting a feature fusion algorithm based on an attention mechanism by data fusion.
3. A topology identification system for power distribution network maintenance as recited in claim 1, wherein said operational boundary topology map is capable of clearly marking maintenance equipment, isolation points, ground wire attachment points, and boundaries of live and power outage ranges.
4. The topology identification system for power distribution network maintenance according to claim 1, wherein an editing unit is arranged in the visual output module, and scaling and translation operations can be performed on the operation boundary topology map by the aid of the editing unit.
5. The topology identification system for power distribution network maintenance as recited in claim 1, wherein the data preprocessing module is internally provided with a storage unit, and the storage unit is utilized to store and backup a high-quality data set formed after preprocessing.
6. The topology identification system for power distribution network maintenance as recited in claim 1, wherein the visual output module is internally provided with a calibration unit, and the calibration unit is utilized to calibrate and recheck the position information of the power distribution network equipment in the operation boundary topology graph.

Description

Topology identification system for overhauling power distribution network Technical Field The invention belongs to the technical field of power distribution network overhaul, and particularly relates to a topology identification system for power distribution network overhaul. Background The power distribution network is used as an important component of a power system, the structure is complex, the nodes are numerous, the topological structure is used as the basis of power grid dispatching, fault processing, maintenance operation and load transfer, therefore, the accurate identification of the topological structure of the power distribution network is a premise of guaranteeing the safety and high-efficiency development of maintenance work, currently, the existing power distribution network topology identification mode mainly depends on manual inspection records or SCADA system data analysis, and has the defects that 1, the manual inspection efficiency is low, the error is large, the topology structure change is difficult to capture in real time, 2, the traditional topology identification algorithm is poor in dynamic adaptability and cannot respond to the power distribution network node state change in time based on static data, 3, the problem of 'figure-model inconsistency' is easy to occur due to lack of accurate positioning of related nodes in the maintenance process, and the situation of mistaken power transmission and power failure is caused by maintenance personnel, and a large potential safety hazard exists. Disclosure of Invention The invention aims to overcome the defects of the prior art and provides a topology identification system for power distribution network maintenance, which is accurate, efficient, safe and convenient. The invention aims to realize that the topology identification system for power distribution network overhaul comprises a data acquisition module, a data preprocessing module, a characteristic extraction module, a topology reasoning module, an optimization module, a topology generation module and a visual output module, wherein the data acquisition module consists of a plurality of wireless sensors, an intelligent monitoring unit and a time marking unit, wherein the wireless sensors are deployed at power distribution network lines and transformer nodes and are used for acquiring voltage, current, power factors and phase data of power distribution network equipment, the intelligent monitoring unit is used for acquiring switching state data of a circuit breaker and an isolating switch in real time, and the time marking unit is used for time marking the data acquired by each data acquisition module so as to ensure consistency of data time marks; The data preprocessing module is connected with the data acquisition module through a wireless communication assembly and is used for cleaning, filtering, normalizing and fusing the data acquired by the data acquisition module so as to form a high-quality data set; The characteristic extraction module is used for extracting characteristic parameters of the operation of the power distribution network nodes; The topology reasoning module can construct an initial topology model of the power distribution network based on the operation characteristic parameters of the power distribution network nodes extracted by the characteristic extraction module, and then receives perception data from the scene, takes the perception data as a topology verification reference with the highest priority, trains the initial topology model of the power distribution network, and further obtains a topology identification model of the power distribution network; The optimization module can carry out deep analysis on the mechanism of the power distribution network topology identification model based on a transfer learning technology, proposes a high-availability parameter identification algorithm suitable for complex sampling conditions, and then constructs an algorithm confidence assessment system through verification set performance monitoring, model correction and hidden mode extraction technologies in a supervision training process, so that the accuracy and reliability of the power distribution network topology identification model are enhanced; The topology generation module can input the high-quality data set into a power distribution network topology identification model so as to generate the current real topology of the power distribution network, and then automatically generate an operation boundary topology map related to the maintenance area according to a maintenance work plan ticket and combining the current real topology; The visual output module can be used for displaying the operation boundary topological graph and simultaneously planning and marking the access path of the maintenance area. In the data preprocessing process, abnormal values and null values are removed, the influence of electromagnetic interference and impulse noise on data is removed