CN-121981848-A - Intelligent monitoring method, system, equipment and medium for high-risk area on distribution network side

CN121981848ACN 121981848 ACN121981848 ACN 121981848ACN-121981848-A

Abstract

The invention discloses a distribution network side high risk area intelligent monitoring method, a system, equipment and a medium, which relate to the technical field of distribution networks and comprise the steps of constructing a distribution network side high risk area multi-dimensional intelligent monitoring system, collecting multi-source monitoring data, transmitting the multi-source monitoring data to a data processing center in real time in a dual-mode transmission mode, carrying out fusion processing and data standardization to generate a standardized monitoring data set, carrying out risk identification and dynamic assessment based on the standardized monitoring data set, combining an unsafe state cause process model and a risk identification method, identifying a risk type, assessing risk occurrence probability and consequence severity, generating grading early warning information according to a risk identification and dynamic assessment result, and carrying out real-time monitoring and advanced prevention and control of distribution network side high risk area risks. The invention realizes accurate identification, dynamic evaluation and advanced early warning of distribution network risks, upgrades the traditional passive rush repair into active defense, and remarkably improves the safety toughness of the power grid against extreme weather.

Inventors

LIU ANJIANG
ZUO HONGYU
WENG DI
ZHANG SONG
WU PENG
LI XINHAO
WANG ZHUOYUE
CAI YONGXIANG
XU YUTAO
CHEN YU
HE MINGJUN
ZHENG YOUZHUO
WANG YANG
YOU XINYU
ZHANG YANG
LI QIANMIN
Cheng Wengou
YANG JIE
GUO DAO
JIANG FUGUO
Jia Xianping
FAN KE
LI YUE
HU TIANSONG
LAI JINZHOU
CHEN KAILEI
SHI TONG
PAN FUXIANG
LI SIJIA
HE RUI
XIAO XIAOBING
MIAO YU
HAO SHUQING
ZHANG HENGRONG
LI HUAPENG
ZHANG WANCHENG

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260505
Application Date: 20251128

Claims (10)

1. The intelligent monitoring method for the high-risk area on the distribution network side is characterized by comprising the following steps of, Constructing a multi-dimensional intelligent monitoring system of a high risk area at the distribution network side, collecting multi-source monitoring data, transmitting the multi-source monitoring data to a data processing center in real time in a dual-mode transmission mode, carrying out fusion processing and data standardization, and generating a standardized monitoring data set; Based on a standardized monitoring data set, carrying out risk identification and dynamic evaluation, combining an unsafe state causation process model and a risk identification method, identifying a risk type, and evaluating risk occurrence probability and consequence severity; and generating grading early warning information according to the risk identification and dynamic evaluation results, pushing the grading early warning information to a distribution network regulation command center, and carrying out real-time monitoring and advanced prevention and control on the risk of the high-risk area at the distribution network side.
2. The intelligent monitoring method for the high risk area on the distribution network side according to claim 1, wherein the fusion processing comprises the steps of carrying out radiation normalization processing on the multi-source remote sensing image, constructing a multi-source heterogeneous data interaction channel, and converting the multi-source monitoring data into a standard structural format; And extracting risk association features in the standardized data, and establishing a mapping relation between the features and risk types to obtain a feature library for risk analysis.
3. The intelligent monitoring method for the high-risk area on the distribution network side according to claim 2, wherein the risk identification comprises the steps of identifying extreme weather and risk types caused by secondary disasters from three levels of an area level, a line level and a user level based on an unsafe state cause process model and combining historical fault data and disaster cases; And quantifying the cause weight of risk occurrence by adopting a risk identification method, and determining the association relation between the risk type and the disaster causing factor by analyzing the cause chain and the deviation state.
4. The intelligent monitoring method for the distribution network side high risk area of claim 3, wherein the dynamic evaluation comprises constructing a disaster remote sensing model, simulating a distribution network fault scene under extreme weather based on a standardized monitoring data set, and calculating fault occurrence probability; and introducing a risk assessment index, calculating a risk value based on the occurrence probability and the result severity, and quantifying the overall risk level by a weighted summation mode.
5. The intelligent monitoring method for the high risk area on the distribution network side is characterized in that the radiation normalization processing comprises the steps of calculating normalized pixel values by means of a method based on multi-scale histogram matching and by means of mean value and standard deviation of source data and mean value and standard deviation of target data, introducing a stabilization coefficient, fusing a satellite remote sensing image with an unmanned aerial vehicle aerial image, performing space alignment of the multi-source image by means of an image registration method, performing fusion processing on the registered image by means of a fusion method based on multi-resolution analysis, and verifying fusion effect by means of a quality evaluation index; the calculation of the normalized pixel value is formulated as: Wherein, the For the value of the original picture element, For the purpose of normalizing the pixel values, As the mean value of the source data, As a standard deviation of the source data, As the mean value of the data of the object, As a standard deviation of the target data, A stabilization factor to prevent denominator from approaching zero; Constructing a multi-source heterogeneous data interaction channel, namely constructing a data interaction channel by adopting a Redis publishing/subscribing data bus technology, establishing a unified data dictionary and metadata specifications, defining format standards of various monitoring data, designing a standardized data template, and converting heterogeneous data of microclimate data, equipment state data and video monitoring data into a unified structural format according to preset rules; The method comprises the steps of obtaining a feature library for risk analysis, extracting surface temperature, vegetation coverage, equipment load rate and disaster influence range from standardized data as key features, mapping features and risk types through a lookup table and rules, verifying the accuracy of a mapping relation by utilizing historical data, dynamically updating the feature library, and matching different disaster scenes.
6. The intelligent monitoring method for the high risk area on the distribution network side is characterized in that the risk identification method comprises the steps of dividing key nodes according to distribution network topology through HAZOP analysis by adopting a HAZOP and FTA combination method, defining equipment normal states and layering deviations, analyzing causes and results of the deviations, initially establishing association between risk types and disaster-causing factors, constructing an accident tree containing intermediate events and bottom events by taking the serious results identified by the HAZOP as top events analyzed by the FTA, marking historical occurrence probability of the bottom events, calculating importance of the bottom events through minimum cut sets and Boolean algebra, and quantifying the cause weight of the occurrence of the risks.
7. The intelligent monitoring method for the distribution network side high risk area of claim 6, wherein the constructing of the disaster remote sensing model comprises the steps of analyzing satellite remote sensing images and unmanned aerial vehicle aerial images by a remote sensing information processing area and a power grid fault mapping area to extract extreme weather space distribution characteristics, integrating distribution network equipment parameters and equipment vulnerability threshold values by the power grid fault mapping area, and establishing a correlation channel between remote sensing monitoring information and power grid equipment states; The disaster remote sensing model utilizes a decision tree algorithm, takes extreme weather parameters as root nodes, equipment states as internal nodes and failure occurrence probability as leaf nodes, adopts a radix coefficient as a splitting criterion, calculates the leaf node probability along a decision path after new data is input, and carries out dynamic simulation of a distribution network failure scene under extreme weather; the splitting criterion is expressed as a coefficient of kunity formula: Wherein, the For the probability of occurrence of a class i fault, Is the coefficient of the foundation; introducing a risk assessment index, calculating a risk value based on the occurrence probability and the result severity, and quantifying the overall risk level in a weighted summation mode, wherein the formula is as follows: Wherein, the In order to achieve a high degree of severity of the consequences, As a value of the risk, Is a risk class weight coefficient; the outcome severity level criteria may include, Is in the color of blue and has the advantages of high color, Is in the form of yellow color, and the color is, Is in the form of an orange color, Is red; The generation of the grading early warning information comprises four early warning grades of red, orange, yellow and blue, and four response strategies of emergency treatment, enhanced prevention and control, conventional monitoring and attention early warning are respectively carried out; The early warning information is based on multiple risk coupling early warning indexes, wherein the indexes comprise the micro weather parameter exceeding time length, the equipment state abnormal accumulated value and the risk diffusion speed, and the prevention and control advice is visually displayed through an auxiliary decision platform to support SVG graphics, tables and risk distribution map output.
8. An intelligent monitoring system for a high-risk area on a distribution network side, which is applied to the intelligent monitoring method for the high-risk area on the distribution network side according to any one of claims 1-7, and is characterized by comprising a three-dimensional monitoring network module, a multi-source data fusion processing module, a multiple risk identification module, a dynamic risk assessment module and a hierarchical early warning and prevention module; The three-dimensional monitoring network module is used for forming a three-dimensional monitoring network covering an environment, an equipment body and a channel by deploying various monitoring equipment at disaster hidden danger points, lines and user sides of a high-risk area at the distribution network side and collecting multi-source monitoring data; The multi-source data fusion processing module is used for solving the problem of inconsistent radiometric calibration of the satellite and the unmanned aerial vehicle image by adopting a multi-source remote sensing image fusion technology and a radiometric normalization method based on multi-scale histogram matching, constructing a multi-source heterogeneous data interaction channel based on a Redis publishing/subscribing data bus technology, establishing a unified data specification, converting microclimate data, equipment state data and video monitoring data into a standard structural format, extracting risk association characteristics of ground surface temperature, vegetation coverage and equipment load rate, establishing a mapping relation with risk types, and generating a standardized monitoring data set; The multiple risk identification module is used for dividing key nodes according to distribution network topology through HAZOP analysis, defining equipment normal state and layering deviation, analyzing causes and results of the deviations, constructing an accident tree by taking the serious results identified by the HAZOP as top events of FTA analysis, calculating the importance of the events through a minimum cut set and Boolean algebra, and quantifying the cause weight of risk occurrence; The dynamic risk assessment module is used for simulating a distribution network fault scene under extreme weather based on a standardized monitoring data set by constructing a disaster remote sensing model and calculating the fault occurrence probability; The grading early warning and prevention and control module is used for generating grading early warning information according to risk identification and dynamic evaluation results and pushing the grading early warning information to the distribution network regulation command center.
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 a distribution network side high risk area intelligent monitoring method according to any 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 implements the steps of a distribution network side high risk area intelligent monitoring method according to any of claims 1 to 7.

Description

Intelligent monitoring method, system, equipment and medium for high-risk area on distribution network side Technical Field The invention relates to the technical field of distribution networks, in particular to an intelligent monitoring method, system, equipment and medium for a high-risk area on a distribution network side. Background Frequent occurrence of typhoons, heavy rainfall, ice and snow in extreme weather, easy waterlogging points of heavy rainfall on the distribution network side, easy occurrence points of geological disasters, easy occurrence of wire breakage, pole tower dumping, equipment faults and the like on typhoon high-speed line sections and the like, and the safe operation of a power grid and the reliability of power supply of users are directly threatened. The existing distribution network monitoring technology has obvious limitations that a three-dimensional monitoring network covering the environment, the equipment body and the channels is not formed, the capability of monitoring equipment for adapting to different disaster hidden danger points is insufficient, the data transmission is easily interfered in a single mode, the multi-source data isomerism is prominent, the effective radiation normalization processing and the standardized flow are lacked, and the requirement of integrated intelligent monitoring cannot be met. Disclosure of Invention In view of the existing problems, the invention provides a distribution network side high risk area intelligent monitoring method, system, equipment and medium, which are used for solving the problems that monitoring equipment in the prior art is insufficient in capability of adapting to different disaster hidden danger points, multi-source data isomerism is prominent and the requirements of integrated intelligent monitoring cannot be met. In order to solve the technical problems, the intelligent monitoring method for the high risk area of the distribution network side is provided, which comprises the following steps, The method comprises the steps of constructing a multi-dimensional intelligent monitoring system of a high-risk area on a distribution network side, collecting multi-source monitoring data, transmitting the multi-source monitoring data to a data processing center in real time in a dual-mode transmission mode, carrying out fusion processing and data standardization to generate a standardized monitoring data set, carrying out risk identification and dynamic assessment based on the standardized monitoring data set, combining an unsafe state cause process model and a risk identification method, identifying risk types, assessing risk occurrence probability and consequence severity, generating grading early warning information according to risk identification and dynamic assessment results, pushing the grading early warning information to a distribution network regulation command center, and carrying out real-time monitoring and advanced prevention and control of the high-risk area risk on the distribution network side. The method for intelligently monitoring the high-risk area on the distribution network side comprises the following steps of carrying out radiation normalization processing on a multi-source remote sensing image, constructing a multi-source heterogeneous data interaction channel and converting multi-source monitoring data into a standard structural format; And extracting risk association features in the standardized data, and establishing a mapping relation between the features and risk types to obtain a feature library for risk analysis. The method for intelligently monitoring the high-risk area on the distribution network side comprises the steps of identifying the risk types caused by extreme weather and secondary disasters from three levels of an area level, a line level and a user level based on an unsafe state cause process model and combining historical fault data and disaster cases; And quantifying the cause weight of risk occurrence by adopting a risk identification method, and determining the association relation between the risk type and the disaster causing factor by analyzing the cause chain and the deviation state. The dynamic evaluation comprises the steps of constructing a disaster remote sensing model, simulating a distribution network fault scene under extreme weather based on a standardized monitoring data set, and calculating the fault occurrence probability; and introducing a risk assessment index, calculating a risk value based on the occurrence probability and the result severity, and quantifying the overall risk level by a weighted summation mode. The radiation normalization processing comprises the steps of calculating normalized pixel values by means of a multi-scale histogram matching method and utilizing the mean value and standard deviation of source data and the mean value and standard deviation of target data, introducing a stabilization coefficient, fusing a satellite remote sensing imag