CN-122001085-A - Transmission line fault detection and positioning method and system based on sensor network

CN122001085ACN 122001085 ACN122001085 ACN 122001085ACN-122001085-A

Abstract

The invention is suitable for the field of automatic monitoring of power systems, and provides a transmission line fault detection and positioning method and system based on a sensor network, wherein the method comprises the steps that a sensor continuously collects multi-dimensional monitoring data; the method comprises the steps of independently judging abnormality by each sensor, marking abnormal nodes and generating an abnormal information packet, broadcasting information by the abnormal nodes, verifying and generating verification information containing coordinates, voting types and confidence weights by neighbor nodes, integrating the verification information, achieving fault consensus if the sum of the confidence weights exceeds a dynamic threshold value, calculating and correcting fault positions if the fault consensus is achieved, and discarding data and processing false alarm if the fault position is not achieved. The system comprises corresponding acquisition, judgment, broadcasting, consensus judgment and positioning processing modules. The invention improves the reliability and positioning accuracy of fault detection, reduces false alarm and adapts to complex environment.

Inventors

YANG YUQIANG
ZHOU ZHIMING
LI GANG
ZHOU JIE
LIU GUOFENG
SONG LEI
JIANG XIANGWEI
WANG MENGHUA
WANG WENHUI
LIU NAN

Assignees

国网山东省电力公司武城县供电公司

Dates

Publication Date: 20260508
Application Date: 20260109

Claims (8)

1. The transmission line fault detection and positioning method based on the sensor network is characterized by comprising the following steps of: continuously collecting multi-dimensional monitoring data of the power transmission line through a sensor, wherein the multi-dimensional monitoring data comprises electric quantity data and physical quantity data; Based on the multidimensional monitoring data, each sensor independently judges whether line abnormality occurs in real time, marks abnormal nodes of the sensor nodes when any sensor detects a line abnormality event, and simultaneously generates an abnormal information packet containing a node identifier, an abnormality type and an abnormality intensity value; Broadcasting an abnormal information packet to all neighbor nodes in a communication range by using the abnormal node as an initiating node, and using the neighbor node which receives the abnormal information packet as a verification node, carrying out local verification on the abnormal information packet based on local multi-dimensional monitoring data acquired by the neighbor node, calculating the correlation between the multi-dimensional monitoring data and the abnormal information packet, and generating verification information comprising node coordinates, voting types and confidence weights; Integrating verification information of all verification nodes in a preset range by taking an initiating node as a center, and judging that fault consensus is achieved when the total of the confidence weights exceeds a dynamic consensus threshold value based on the confidence weights in the verification information; And when the fault consensus is not achieved, automatically discarding all relevant data of the line abnormal event, and executing false alarm suppression processing.
2. The method of claim 1, wherein continuously collecting multi-dimensional monitoring data of the transmission line by the sensor comprises: Continuously collecting current and voltage waveform data, vibration acceleration data and sound wave signal data of the power transmission line through a sensor to serve as multi-dimensional monitoring data; Preprocessing the collected multi-dimensional monitoring data, storing the preprocessed multi-dimensional monitoring data in a local cache according to a time sequence, and adding a time stamp mark for each data block.
3. The method according to claim 2, wherein generating an anomaly packet comprising a node identifier, an anomaly type and an anomaly strength value, comprises: Calculating time domain characteristic parameters and frequency domain characteristic parameters of the multi-dimensional monitoring data in real time, wherein the time domain characteristic parameters comprise a current effective value, a voltage peak value and a vibration root mean square value, and the frequency domain characteristic parameters comprise current harmonic content, a sound wave main frequency amplitude value and vibration spectrum energy; Comparing the calculated time domain characteristic parameters and frequency domain characteristic parameters with a preset fault threshold value item by item, judging that a line abnormal event of a type corresponding to the characteristic parameters occurs when any characteristic parameter continuously exceeds the corresponding fault threshold value and the duration exceeds the safety judging duration, and calculating an abnormal intensity value; marking the sensor node with the detected line abnormal event as an abnormal node, and setting an abnormal state mark in a node state register; based on the determination result of the line anomaly event, an anomaly packet including a node identifier, an anomaly type, and an anomaly strength value is generated.
4. A method according to claim 3, characterized in that said generating verification information comprising node coordinates, voting types and confidence weights, in particular comprises: using the abnormal node as an initiating node, broadcasting and transmitting the abnormal information packet to all neighbor nodes in a communication range, and starting a timer to record broadcasting time; The neighbor node receiving the abnormal information packet is used as a verification node, and multi-dimensional monitoring data corresponding to the time stamp of the abnormal information packet is retrieved from a local cache, wherein the multi-dimensional monitoring data comprise current waveform data, voltage waveform data, vibration acceleration data and sound wave signal data; the verification node extracts the characteristic parameters corresponding to the abnormal type, calculates the time domain correlation coefficient of the characteristic parameters and the abnormal intensity value, simultaneously analyzes the frequency spectrum characteristic, and calculates the frequency domain similarity of the typical fault frequency spectrum corresponding to the abnormal type; Generating a verification conclusion based on the weighted average value of the time domain correlation coefficient and the frequency domain similarity, generating passing verification information when the weighted average value is larger than 0.75, or generating non-passing verification information, and simultaneously generating confidence weights in the range of 0-1 according to the size quantification of the weighted average value, wherein the confidence weights of the passing verification information are reset to positive values, the confidence weights of the non-passing verification information are reset to negative values, and packaging to generate a complete verification information packet.
5. The method of claim 4, wherein the determining achieves fault consensus, in particular comprising: Setting a maximum propagation range, establishing a monitoring area which takes an initiating node as a center and covers all reachable nodes in a large propagation range, transferring received verification information in a communication range by a verification node, and counting down the monitoring area range in the forwarding process; the initiating node collects verification information sent by all verification nodes in a monitoring area in a preset time window, wherein the verification information comprises node coordinates, voting types and confidence weights; screening the collected verification information, and reserving effective verification information containing complete node coordinates, voting types and confidence weights; And calculating the confidence weight sum of all the effective verification information, comparing the confidence weight sum with a consensus threshold dynamically adjusted based on the historical data, judging that the fault consensus is achieved when the confidence weight sum exceeds the dynamic consensus threshold, and judging that the fault consensus is not achieved if the confidence weight sum does not exceed the dynamic consensus threshold.
6. The method of claim 5, wherein when a failure consensus is reached: extracting node coordinates and confidence weights of verification nodes from all verification information, and carrying out weighted calculation on the node coordinates and the confidence weights of each verification node to obtain initial fault position coordinates; And correcting the fault position coordinates according to the fault type, and outputting final fault position coordinates.
7. The method of claim 5, wherein when a failure consensus is not reached: When failure consensus is not achieved, automatically clearing abnormal information packets and verification information stored in the abnormal nodes and the verification nodes; Updating the false alarm statistical record, and self-adaptively adjusting fault threshold parameters in the abnormal detection algorithm according to the false alarm statistical record.
8. Transmission line fault detection and positioning system based on sensor network, its characterized in that, the system includes: The multi-dimensional monitoring data acquisition module is used for continuously acquiring multi-dimensional monitoring data of the power transmission line through the sensor, wherein the multi-dimensional monitoring data comprise electric quantity data and physical quantity data; The abnormal event independent judging module is used for independently judging whether line abnormality occurs or not in real time by each sensor based on the multidimensional monitoring data, marking the sensor node with an abnormal node when any sensor detects the line abnormal event, and generating an abnormal information packet containing a node identifier, an abnormal type and an abnormal intensity value; The abnormal information broadcasting module is used for broadcasting an abnormal information packet to all neighbor nodes in a communication range by taking the abnormal node as an initiating node, and the neighbor nodes receiving the abnormal information packet serve as verification nodes, carrying out local verification on the abnormal information packet based on local multi-dimensional monitoring data acquired by the abnormal information broadcasting module, calculating the correlation between the multi-dimensional monitoring data and the abnormal information packet, and generating verification information comprising node coordinates, voting types and confidence weights; The fault consensus judging module is used for integrating verification information of all verification nodes in a preset range by taking the initiating node as a center, and judging that fault consensus is achieved when the total of the confidence weights exceeds a dynamic consensus threshold value based on the confidence weights in the verification information; the fault position calculating and false alarm processing module is used for calculating the fault position coordinates based on the node coordinates and the confidence weights of all verification nodes when the fault consensus is achieved, automatically discarding all relevant data of the line abnormal event when the fault consensus is not achieved, and executing false alarm suppression processing.

Description

Transmission line fault detection and positioning method and system based on sensor network Technical Field The invention belongs to the field of automatic monitoring of power systems, and particularly relates to a power transmission line fault detection and positioning method and system based on a sensor network. Background The power transmission line is a core carrier for the power system to realize the remote transmission of electric energy, and the running state of the power transmission line directly determines the continuity and reliability of power supply. With the development of the power network to the high-voltage, long-distance and complex terrain coverage direction, risks of natural interference, mechanical loss, external damage and the like of the power transmission line are obviously increased, and the requirements of the power system on real-time performance and positioning accuracy of fault detection are difficult to meet in the traditional fault management mode relying on manual inspection and fixed-point monitoring. In recent years, the sensor network technology gradually becomes the core technical direction of power transmission line monitoring by virtue of the advantages of distributed deployment, multi-node cooperation, real-time data interaction and the like, the industry generally promotes the fusion of the sensor network and the power monitoring, and the rapid identification and the accurate positioning of faults are expected to be realized through multi-dimensional data acquisition and analysis, so that the operation and maintenance efficiency and the power supply stability of a power system are improved. The problems of insufficient fault detection reliability, limited positioning precision and poor environmental adaptability of the whole prior art are particularly represented by single data monitoring dimension, failure in comprehensively capturing the differentiation characteristics of electrical faults and mechanical faults, failure omission, lack of a multi-node collaborative verification and dynamic consensus mechanism, easy false alarm caused by single-node misjudgment or instantaneous interference, fixed fault judgment threshold, incapability of adapting to the running environment changes of different regions and different seasons, failure positioning, non-combination of data reliability weight and failure type characteristics, larger positioning deviation, no effective misalarm suppression and self-optimization mechanism, easy decline of system performance after long-term running, and difficulty in meeting the fault management requirements of a power transmission line in a complex environment. Disclosure of Invention The invention aims to provide a transmission line fault detection and positioning method based on a sensor network, and aims to solve the technical problems in the prior art determined in the background art. The invention is realized in such a way that the transmission line fault detection and positioning method based on the sensor network comprises the following steps: continuously collecting multi-dimensional monitoring data of the power transmission line through a sensor, wherein the multi-dimensional monitoring data comprises electric quantity data and physical quantity data; Based on the multidimensional monitoring data, each sensor independently judges whether line abnormality occurs in real time, marks abnormal nodes of the sensor nodes when any sensor detects a line abnormality event, and simultaneously generates an abnormal information packet containing a node identifier, an abnormality type and an abnormality intensity value; Broadcasting an abnormal information packet to all neighbor nodes in a communication range by using the abnormal node as an initiating node, and using the neighbor node which receives the abnormal information packet as a verification node, carrying out local verification on the abnormal information packet based on local multi-dimensional monitoring data acquired by the neighbor node, calculating the correlation between the multi-dimensional monitoring data and the abnormal information packet, and generating verification information comprising node coordinates, voting types and confidence weights; Integrating verification information of all verification nodes in a preset range by taking an initiating node as a center, and judging that fault consensus is achieved when the total of the confidence weights exceeds a dynamic consensus threshold value based on the confidence weights in the verification information; And when the fault consensus is not achieved, automatically discarding all relevant data of the line abnormal event, and executing false alarm suppression processing. As a further aspect of the present invention, the continuously collecting the multi-dimensional monitoring data of the power transmission line through the sensor includes: Continuously collecting current and voltage waveform data, vibration acceleration data and