CN-122017475-A - Distribution network ground fault information studying and judging method, system, computer equipment and medium

CN122017475ACN 122017475 ACN122017475 ACN 122017475ACN-122017475-A

Abstract

The invention relates to the technical field of analysis of power systems, in particular to a method, a system, computer equipment and a medium for studying and judging ground fault information of a distribution network; the method comprises the steps of obtaining multi-source fault data in a distribution network system, preprocessing the multi-source fault data to obtain a standardized fault data set, extracting fault characteristic information from the standardized fault data set through a time sequence pattern recognition algorithm, performing clustering analysis based on topological relation on the fault characteristic information through a multi-source data fusion and research algorithm based on the fault characteristic information to generate a fault section positioning result, and generating distribution network ground fault information according to topological relation identification and electrical parameter characteristics in the fault section positioning result. By the mode, the technical problem that the positioning accuracy is insufficient in the power distribution network with high reliability requirement in the existing ground fault judging technology is solved, and the automation and intelligence level of fault judging and the accuracy and usability of judging conclusion are improved.

Inventors

WU WEIBIN
YIN YUE
ZHENG LI
XU PENGHUI
XU HONGCHUAN
LIU JIANGTAO
CHEN SHANSHAN
CHEN WEI
He Liye
ZHOU YIJIAN
SHI ZHIJIA

Assignees

国网浙江省电力有限公司龙泉市供电公司

Dates

Publication Date: 20260512
Application Date: 20260415

Claims (10)

1. The method for judging the ground fault information of the distribution network is characterized by comprising the following steps of: the method comprises the steps of obtaining multi-source fault data in a distribution network system, and preprocessing the multi-source fault data to obtain a standardized fault data set; Extracting fault feature information from the standardized fault dataset through a time sequence pattern recognition algorithm configured to perform first half-wave feature extraction and waveform similarity measurement on zero sequence current waveforms in the standardized fault dataset; Performing a clustering analysis based on a topological relation on the fault characteristic information by a multi-source data fusion and research algorithm based on the fault characteristic information to generate a fault section positioning result, wherein the multi-source data fusion and research algorithm is configured to calculate an output section identifier by adopting a waveform correlation based on a time derivative Euclidean distance; and generating distribution network ground fault information according to the topological relation identification and the electrical parameter characteristics in the fault section positioning result.
2. The method of claim 1, wherein the standardized fault data set includes ground fault log data, wherein the extracting fault signature information from the standardized fault data set by a time sequence pattern recognition algorithm comprises: Performing first half-wave interception treatment on the zero-sequence current waveform in the ground fault wave recording data to obtain a first half-wave waveform; carrying out nonlinear fitting calculation on the first half-wave waveform by a nonlinear least square optimization algorithm to obtain a zero-sequence current fitting function; Calculating time derivative parameters of the fitting waveform according to the zero sequence current fitting function, and executing Euclidean distance calculation based on the time derivative parameters to obtain a waveform similarity measurement value; And constructing the fault characteristic information based on the zero sequence current fitting function, the time derivative parameter and the waveform similarity measurement value.
3. The method of claim 2, wherein the generating a fault section localization result by performing a topology-based cluster analysis on the fault signature by a multi-source data fusion and research algorithm based on the fault signature comprises: Based on the waveform similarity measurement value, constructing a multidimensional feature vector of the feeder line outgoing switches, and calculating waveform correlation coefficients among the feeder line outgoing switches through a multidimensional feature distance statistical algorithm; Based on the waveform correlation coefficient, determining a feeder line with the lowest waveform correlation coefficient as a fault line through feeder line level cluster analysis; Calculating waveform correlation coefficients between adjacent switch nodes in a topological graph through a section level graph analysis algorithm based on the switch topological relation of the fault line; and identifying the adjacent switch pair with the lowest waveform correlation coefficient through a minimum similarity node pair identification algorithm, and generating a fault section positioning result, wherein the fault section positioning result comprises identification information of the adjacent switch pair.
4. A method according to claim 3, wherein generating distribution network ground fault information according to the topology relation identification and the electrical parameter characteristics in the fault section location result comprises: generating a fault section boundary identifier through a boundary node coding algorithm based on the identification information of the adjacent switch pairs; Acquiring a line identifier and fault phase characteristics from the fault section positioning result through a characteristic extraction engine; constructing standardized fault event descriptions through an event description generator by combining the time stamp information; And injecting the fault section boundary identification, the line identifier, the fault phase characteristics and the standardized fault event description into an information fusion engine, performing format standardized encapsulation through a data serialization protocol, and outputting the distribution network ground fault information.
5. The method of claim 1, wherein the multi-source fault data comprises a distribution terminal ground alarm signal, bus ground information, low current ground line selection information, arc suppression coil information, and bus voltage information acquired from a distribution network IV system, the multi-source fault data being real-time fault data acquired from the distribution network IV system by a secure data acquisition agent configured to: Establishing a secure communication tunnel with a distribution network IV area system, and acquiring a data access token through a credential authentication mechanism; analyzing and executing a preset graphical interface operation script by adopting a human-computer interaction simulation engine, and driving an acquisition flow based on a predefined rule engine; performing hash check on the collected original data stream, and constructing an end-to-end data security pipeline by adopting a transmission layer security protocol; And inputting the encrypted and transmitted real-time fault data into a data preprocessing pipeline, and executing data pattern mapping and conversion and outlier filtering based on statistical distribution.
6. The method of claim 5, wherein the distribution network ground fault information is output through a human-machine interaction interface module configured to: constructing a graphical monitoring panel through a visual rendering engine, and dynamically rendering a fault line topological graph and an electrical parameter waveform sequence; Integrating an interactive editing workflow, supporting a rich text markup language editing and structured data confirmation protocol; Deploying a multichannel message distribution engine, packaging standardized fault information into a cross-platform message format, and adapting to protocol conversion of a short message gateway and a collaborative office system; An asynchronous auditing workflow engine is deployed, a multi-stage verification node is integrated in a message distribution pipeline, and an identity verification protocol and a digital signature algorithm are executed.
7. The method of claim 1, wherein fault tolerant control of the process flow is implemented by a system reliability assurance architecture configured to: Deploying a periodic polling scheduler, and starting a data acquisition and fault research and judgment task pipeline based on a time trigger mechanism; Executing a timeout detection algorithm, and establishing an independent execution time limit monitoring channel for each processing module; deploying a process state monitor, executing a timeout or abnormal exit event through an abnormal state identification engine detection module, calling a process to terminate a process related to closing of the reestablishing method and executing a system reinitialization sequence; And running an audit trail system, continuously recording the running indexes and the abnormal event log, and establishing an operation and maintenance audit data chain.
8. The utility model provides a join in marriage net ground fault information and judge system which characterized in that includes: the fault data set construction module is used for acquiring multi-source fault data in the distribution network system and preprocessing the multi-source fault data to obtain a standardized fault data set; The fault characteristic information extraction module is used for extracting fault characteristic information from the standardized fault data set through a time sequence pattern recognition algorithm, and the time sequence pattern recognition algorithm is configured to perform first half-wave characteristic extraction and waveform similarity measurement on zero-sequence current waveforms in the standardized fault data set; The fault section positioning module is used for performing clustering analysis based on a topological relation on the fault characteristic information through a multi-source data fusion and judgment algorithm based on the fault characteristic information to generate a fault section positioning result, and the multi-source data fusion and judgment algorithm is configured to calculate an output section identifier by adopting waveform correlation based on a time derivative Euclidean distance; and the ground fault information generation module is used for generating the ground fault information of the distribution network according to the topological relation identification and the electrical parameter characteristics in the fault section positioning result.
9. A computer device, comprising: and a memory communicatively coupled to the at least one processor; Wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
10. A non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are for causing a computer to perform the method of any one of claims 1-7.

Description

Distribution network ground fault information studying and judging method, system, computer equipment and medium Technical Field The invention relates to the technical field of analysis of power systems, in particular to a method, a system, computer equipment and a medium for studying and judging ground fault information of a distribution network. Background In operation of the power system, single-phase earth faults are one of the most common fault types in the power distribution network, and the occurrence rate of the single-phase earth faults accounts for more than 70% of the total fault rate of the system. The ground fault not only can cause overvoltage impact to cause insulation breakdown and equipment burning, but also causes that fault current continuously flows into the ground to influence the action logic of the protection device, and can induce override trip to cause large-area power failure and even cause chain reactions such as fire, explosion and the like. Particularly, long-time indirect faults, have serious influence on the service life of the system and the power supply reliability. Taking actual operation data as an example, the line of a power supply company in a region in 2024 has a ground fault for 40 times, and the average fault time reaches 1.5 hours, which highlights the obvious problems in the ground fault research and judgment process, namely that the research and judgment time is overlong, the fault range is positioned inaccurately, the fault line inspection and finding time is prolonged, and the fault treatment efficiency is seriously affected. Currently, the ground fault research and judgment of a power distribution network mainly depends on alarm information sent by a station line selection device and a feeder terminal, and is manually or simply automatically analyzed by combining a line topological graph. The prior art (application publication number CN 119291395A) discloses a single-phase earth fault section positioning method of a power distribution network based on multi-source data fusion, which is used for comprehensively studying and judging by acquiring power distribution cloud master station and dispatching remote signaling data and utilizing various information such as ground fault alarm information, zero sequence current telemetry value, ground fault wave recording and the like, and aims to improve the reliability and accuracy of positioning. However, the method in the prior art still has the limitations that firstly, for different neutral point grounding modes (such as an arc suppression coil grounding system or an ungrounded system), the adaptability of a research and judgment strategy is insufficient, the situations of failure in information uploading or data missing cannot be fully considered, so that the research and judgment accuracy in practical application is limited, secondly, the method has weaker global fault characteristic perceptibility on multiple feeder line working conditions under complex power distribution network topology, and is difficult to effectively cope with voltage abnormality caused by reverse power flow or intermittent fluctuation, and finally, the prior art generally relies on offline calculation or fixed threshold values, lacks a real-time dynamic adjustment mechanism, and is easy to cause research and judgment delay or misjudgment. Therefore, in the existing ground fault research and judgment technology, in the power distribution network with high reliability requirements, the technical problem of insufficient positioning precision exists, and the actual requirements of quick fault isolation and power supply recovery cannot be met. Disclosure of Invention The invention aims at the defects or shortcomings, provides a method, a system, computer equipment and a medium for researching and judging the ground fault information of a distribution network, and can solve the technical problem that the existing ground fault research and judgment technology has insufficient positioning precision in a power distribution network with high reliability requirements. The invention provides a method for studying and judging ground fault information of a distribution network, which comprises the following steps: and acquiring multi-source fault data in the distribution network system, and preprocessing the multi-source fault data to obtain a standardized fault data set. Fault characteristic information is extracted from the standardized fault data set by a time sequence pattern recognition algorithm configured to perform first half-wave characteristic extraction and waveform similarity measurement on the zero sequence current waveform in the standardized fault data set. Based on the fault feature information, performing a topology-based clustering analysis on the fault feature information by a multi-source data fusion and pestilence algorithm configured to calculate an output segment identifier using a time derivative euclidean distance-based waveform corr