CN113837458A - Edge calculation model of power distribution network data, and distribution network equipment defect elimination auxiliary system and defect elimination method based on edge calculation model - Google Patents

Edge calculation model of power distribution network data, and distribution network equipment defect elimination auxiliary system and defect elimination method based on edge calculation model Download PDF

Info

Publication number
CN113837458A
CN113837458A CN202111082952.9A CN202111082952A CN113837458A CN 113837458 A CN113837458 A CN 113837458A CN 202111082952 A CN202111082952 A CN 202111082952A CN 113837458 A CN113837458 A CN 113837458A
Authority
CN
China
Prior art keywords
data
cluster
distribution network
clustering
center
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111082952.9A
Other languages
Chinese (zh)
Other versions
CN113837458B (en
Inventor
肖惠
慈白山
袁帅
胡超
彭立
于鹏
邹欢
张坚
李强
李海涛
刘云祥
欧阳耀军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanchang Honggutan Power Supply Branch State Grid Jiangxi Electric Power Co ltd
Nanchang Power Supply Branch State Grid Jiangxi Province Electric Power Co ltd
State Grid Corp of China SGCC
Original Assignee
Nanchang Honggutan Power Supply Branch State Grid Jiangxi Electric Power Co ltd
Nanchang Power Supply Branch State Grid Jiangxi Province Electric Power Co ltd
State Grid Corp of China SGCC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanchang Honggutan Power Supply Branch State Grid Jiangxi Electric Power Co ltd, Nanchang Power Supply Branch State Grid Jiangxi Province Electric Power Co ltd, State Grid Corp of China SGCC filed Critical Nanchang Honggutan Power Supply Branch State Grid Jiangxi Electric Power Co ltd
Publication of CN113837458A publication Critical patent/CN113837458A/en
Application granted granted Critical
Publication of CN113837458B publication Critical patent/CN113837458B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00—Economic sectors
    • G16Y10/35—Utilities, e.g. electricity, gas or water
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00—Information sensed or collected by the things
    • G16Y20/10—Information sensed or collected by the things relating to the environment, e.g. temperature; relating to location
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00—Information sensed or collected by the things
    • G16Y20/20—Information sensed or collected by the things relating to the thing itself
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00—Information sensed or collected by the things
    • G16Y20/30—Information sensed or collected by the things relating to resources, e.g. consumed power
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/10—Detection; Monitoring
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/20—Analytics; Diagnosis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/40—Maintenance of things
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/50—Safety; Security of things, users, data or systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/60—Positioning; Navigation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Computing Systems (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Health & Medical Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • General Health & Medical Sciences (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Toxicology (AREA)
  • Artificial Intelligence (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Accounting & Taxation (AREA)
  • Probability & Statistics with Applications (AREA)
  • General Engineering & Computer Science (AREA)
  • Water Supply & Treatment (AREA)
  • Environmental & Geological Engineering (AREA)
  • Biomedical Technology (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Security & Cryptography (AREA)

Abstract

The invention discloses a data edge calculation model of a power distribution network, which comprises the steps of (1) collecting state data of distribution network equipment to obtain a data set; (2) determining a clustering number K value; (3) determining a clustering radius-Euclidean distance; (4) updating the clustering center and iterative output. The invention also discloses a distribution network equipment defect elimination auxiliary system and a defect elimination method based on the power distribution network data edge calculation model. The device of the invention is embedded and installed on the power distribution network equipment, so that the sharp increase of information can be relieved, the pressure of background data processing can be relieved, the background data processing time can be reduced, and the data processing rate of a power distribution network system can be improved. The system stores related information such as the primary and secondary wiring diagrams of the power distribution network equipment, so that the loss rate of the information is reduced, the efficiency and the accuracy of eliminating the defects of the operation and maintenance team and group personnel are improved, and the defect eliminating time cannot be delayed due to the fact that the models of the power distribution network equipment are different and the wiring modes are different.

Description

Edge calculation model of power distribution network data, and distribution network equipment defect elimination auxiliary system and defect elimination method based on edge calculation model
Technical Field
The invention belongs to the technical field of electric power defect elimination assistance, particularly relates to a distribution network equipment defect elimination assistance system based on edge calculation, and further relates to a defect elimination method of the defect elimination assistance system.
Background
With the continuous development and progress of the society, the electricity consumption requirements of industrial, commercial and residential people are sharply increased, the construction of a power grid is rapidly developed, the load rate of the power grid is continuously increased, and the sharply increased number of distribution network equipment provides higher and higher requirements for lean operation and maintenance work of the distribution network. On one hand, the information uploaded by the distribution network equipment is increased sharply, so that information transmission delay, time and bandwidth are increased, and the data processing rate of the distribution network system is seriously influenced; on the other hand, due to the large missing task quantity and the high quality index of the high quality service, the low data processing rate of the distribution network system can lead to the extension of the working time of the staff of a team and the high psychological pressure. Under the normal condition, before the distribution network equipment is put into operation, drawings of a primary and secondary electrical wiring diagrams of the distribution network equipment are placed in the cabinet body, but the loss probability is increased along with the increase of the defect eliminating times of operation and maintenance. And because distribution network equipment is many kinds and the model differs, there is a little difference in the first, secondary electric wiring of different models, brings very big challenge for fortune dimension team personnel accurate quick disappearance.
Disclosure of Invention
The invention mainly solves the technical problem of providing a distribution network equipment defect elimination auxiliary system based on edge computing, realizing the transparency of the geographic information of a ring main unit, the precise simplification of data and the paperless management of defect elimination electrical drawings, and further solving the problem of time delay when the existing distribution network equipment is deleted.
The invention solves the technical problems through the following technical proposal,
a data edge calculation model of a power distribution network comprises the following steps,
(1) collecting state data of distribution network equipment to obtain data set
Xn＝{(x1,y1,...,z1),(x2,y2,...,z2),…,(xn,yn,...,zn)}；
(2) Clustering number K value determination
Setting a coefficient a as a convergence condition parameter of the clustering number K value, wherein 0< a < 1;
(2.1) randomly generating data set XnOne m point is selected as the 1 st clustering center
Figure BDA0003264745770000011
Figure BDA0003264745770000012
Represents the 1 st cluster center with m points as the starting point, wherein
Figure BDA0003264745770000013
Representing m points as the 1 st cluster center in data set XnThe value of (b) is the value of the starting point m;
(2.2) in data set XnIn the method, the 1 st cluster center is selected
Figure BDA0003264745770000021
The farthest point is taken as the 2 nd clustering center
Figure BDA0003264745770000022
Selecting the clustering center after the point selected as the clustering center does not enter;
(2.3) in data set XnIn the method, the distance from each point except the cluster center to all the cluster centers is calculated, and the distance formula is as follows:
Figure BDA0003264745770000023
1 ≦ i ≦ n, i being the data set X excluding the cluster centernA point of (1);
j is more than or equal to 1 and less than or equal to k, and k is the number of the selected clustering centers;
(2.4) obtaining the minimum distance of each point except the cluster center from all the cluster centers
Figure BDA0003264745770000024
Figure BDA0003264745770000025
1 ≦ i ≦ n, i being the data set X excluding the cluster centernA point of (1);
(2.5) judging the minimum distance
Figure BDA0003264745770000026
Whether or not:
Figure BDA0003264745770000027
a is the set coefficient in the step (2), and 0< a < 1;
if it is
Figure BDA0003264745770000028
The point i is the cluster of the cluster center closest to the point i;
if it is
Figure BDA0003264745770000029
Selecting the point i as a (k + 1) th clustering center;
(2.6) repeating the operation according to the step (2.3) to the step (2.5), and selecting the Kth clustering center
Figure BDA00032647457700000210
Determining a value of the cluster number K and setting the data set X except the cluster centernThe point in the cluster is divided into a cluster center closest to the point in the cluster to become a cluster of the cluster center;
(3) cluster radius-euclidean distance
The similarity index of the clusters is represented by Euclidean distance, and the selected clustering radius is the maximum value of the distance from all points in the cluster to the clustering center;
through the step (2), K clustering centers are obtained in total, namely K clusters exist,
then the jth cluster center
Figure BDA0003264745770000031
The radius of (a) is:
Figure BDA0003264745770000032
j is more than or equal to 1 and less than or equal to K, i is a cluster in the jth cluster, and m is more than or equal to 1 and less than or equal to n;
wherein, the point (x)i,yi,...,zi) A value belonging to a class cluster in the jth cluster;
obtaining the clustering radius of K clustering centers;
(4) updating cluster centers and iterative outputs
(4.1) calculating a clustering objective function
Set the clustering objective function as
Figure BDA0003264745770000033
Obtaining a clustering objective function divided by K clustering centers;
(4.2) calculating to obtain K first updated cluster centers by taking the mean value of the clusters in the same cluster as the first updated cluster centers, and calculating corresponding first updated cluster objective functions;
(4.3) if any one of the first updating clustering objective function and the first updating clustering center and the clustering objective function and the clustering center is changed, recalculating the data set XnThe distance of each element from the K first updated cluster centers,
the distance formula is:
Figure BDA0003264745770000034
i is more than or equal to 1 and less than or equal to n, i is a data set XnAll points in (1);
j is more than or equal to 1 and less than or equal to K, and K is the number of clusters, namely the value of K;
obtaining a data set XnEach point in (1) is a minimum distance from all first-update cluster centers
Figure BDA0003264745770000035
Figure BDA0003264745770000036
I is more than or equal to 1 and less than or equal to n, i is a data set XnAll points in (1);
minimum distance
Figure BDA0003264745770000037
The corresponding element is drawn as the minimum distance
Figure BDA0003264745770000038
Correspondingly updating the cluster of the cluster center for the first time;
re-obtaining the cluster elements of the K first-time updating clusters;
(4.4) according to the method in the step (3), obtaining the clustering radius of K first-time updated clustering centers;
(4.5) according to the method in the step (4.2), calculating to obtain K second updated clustering centers by taking the mean value of the clusters in the same cluster as the second updated clustering center, and calculating the corresponding second updated clustering objective function;
(4.6) according to the method described in the step (4.4), if any of the second updated clustering objective function and the second updated clustering center and the first updated clustering objective function and the first updated clustering center is changed, recalculating the data set XnThe distance between each element and K second updated clustering centers is obtained to obtain a data set XnEach point in (1) is a minimum distance from all second-update cluster centers
Figure BDA0003264745770000041
Minimum distance
Figure BDA0003264745770000042
The corresponding element is drawn as the minimum distance
Figure BDA0003264745770000043
Correspondingly updating the cluster of the cluster center for the second time;
re-obtaining the cluster elements of the K second updating clusters;
(4.7) repeating the steps (4.4) - (4.6) until the cluster objective function updated s times and the cluster center updated s times and any one of the cluster objective function updated s-1 times and the cluster center updated s-1 times are not changed, indicating data convergence, and outputting the cluster center updated s times.
The output s-th updated clustering center is used as representative data, so that the data volume of background processing can be reduced.
The invention also discloses a distribution network equipment defect elimination auxiliary system based on the power distribution network data edge calculation model, which comprises a distribution network equipment acquisition module and a distribution network communication module and is characterized in that,
the distribution network equipment acquisition module comprises a data acquisition unit, a position acquisition unit and a data alarm unit,
the data acquisition unit acquires state data of the distribution network equipment;
the position acquisition unit acquires geographical position information data of the distribution network equipment;
the data alarm unit compares the state data and/or the geographical position information data with preset data, and when the state data and/or the geographical position information data exceed the preset data range, the data alarm unit sends alarm data, sends the alarm data to the distribution network communication module, and sends the alarm data to the rear-end distribution network system server by the distribution network communication module;
the distribution network communication module comprises a data supervision unit, a data processing unit and a data transmission unit,
the data monitoring unit monitors state data of the distribution network equipment in real time, monitors whether the distribution network equipment normally operates or not, and sends a monitoring signal to a rear-end distribution network system server for recording if the equipment abnormally operates;
the data processing unit preprocesses the state data through the power distribution network data edge calculation model to obtain preprocessed state data, and uploads the preprocessed state data to a rear-end power distribution network system server, so that the information data amount transmitted to the power distribution network system by the data transmission unit is reduced, and the pressure of the power distribution network system in calculating and processing the state data is reduced;
the data transmission unit transmits the preprocessed data processed by the data processing unit to a back-end power distribution network system server through optical fibers or wireless.
In order to obtain a better technical effect, the state data comprises real-time voltage, real-time current, a voltage transformer, a current transformer, a switch on/off switch, relay protection on/off and temperature;
in order to obtain a better technical effect, the distribution network equipment acquisition module further comprises a data storage unit, the data storage unit stores historical data of the data acquisition unit and the position acquisition unit and equipment information data of the distribution network equipment, and the historical data can be automatically cleaned in a period selected;
in order to obtain better technical effect, the equipment information data comprises equipment delivery time, equipment manufacturers, equipment specifications and stored equipment information;
in order to obtain a better technical effect, the distribution network communication module further comprises a safety guarantee unit, wherein the safety guarantee unit is an encryption module containing an encryption algorithm, and is used for carrying out encryption protection on various running state data of the distribution network equipment so as to ensure that the data is not easy to hijack and crack in the transmission process.
The invention also discloses a defect elimination method of the distribution network equipment defect elimination auxiliary system based on the power distribution network data edge calculation model, which comprises the following steps,
(1) displaying fault information and pre-judged specific information on a web page and a mobile phone terminal by using a prediction result of a rear-end power distribution network system server, and automatically issuing fault short messages to a maintainer and a responsible person of an operation and maintenance class of the geographical information by the system according to the geographical information of the fault;
(2) the maintainers and the operation and maintenance team and group responsible person receive the equipment fault short message and log in the mobile phone terminal to check the detailed fault message;
(3) the operation and maintenance team personnel judge the fault message according to the detailed fault message and whether the fault message needs to be eliminated when going out;
(4) if the fault does not occur emergently, the normal operation of the equipment is not hindered, and the equipment is not suitable for live working, the fault message can be loaded into historical fault messages, and the personnel of the operation and maintenance team and team can go out in batch uniformly to eliminate the fault according to the historical fault messages in the later period;
(5) if the fault needs to be solved, the operation and maintenance team personnel navigate to the position of the fault equipment according to the equipment positioning data acquired by the position acquisition unit of the power distribution equipment acquisition module;
(6) in the operation, maintenance and defect elimination process of the equipment, the wiring diagram and the schematic diagram of the equipment are slightly different due to various distribution network equipment, and the challenges are brought to the refined operation, maintenance and defect elimination. And the operation and maintenance team personnel search the related information such as the primary and secondary wiring diagrams of the fault equipment through the mobile phone terminal of the system and then accurately eliminate the fault equipment.
The device of the invention is embedded and installed on the power distribution network equipment, so that the sharp increase of information can be relieved, the pressure of background data processing can be relieved, the background data processing time can be reduced, and the data processing rate of a power distribution network system can be improved. The system stores related information such as the primary and secondary wiring diagrams of the power distribution network equipment, so that the loss rate of the information is reduced, the efficiency and the accuracy of eliminating the defects of the operation and maintenance team and group personnel are improved, and the defect eliminating time cannot be delayed due to the fact that the models of the power distribution network equipment are different and the wiring modes are different.
Drawings
FIG. 1 is a schematic structural diagram of an embodiment of the present invention;
FIG. 2 is a flowchart illustrating deletion processing according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings and examples.
Example 1
A data edge calculation model of a power distribution network comprises the following steps,
(1) collecting state data of distribution network equipment to obtain data set
Xn＝{(x1,y1,...,z1),(x2,y2,...,z2),…,(xn,yn,...,zn)}；
(2) Clustering number K value determination
Setting a coefficient a as a convergence condition parameter of the clustering number K value, wherein 0< a < 1;
(2.1) randomly generating data set XnOne m point is selected as the 1 st clustering center
Figure BDA0003264745770000061
Figure BDA0003264745770000062
Represents the 1 st cluster center with m points as the starting point, wherein
Figure BDA0003264745770000063
Representing m points as the 1 st cluster center in data set XnThe value of (b) is the value of the starting point m;
(2.2) in data set XnIn the method, the 1 st cluster center is selected
Figure BDA0003264745770000064
The farthest point is taken as the 2 nd clustering center
Figure BDA0003264745770000071
Selecting the clustering center after the point selected as the clustering center does not enter;
(2.3) in data set XnCalculating the distance from each point except the cluster center to each cluster center, wherein the distance formula is as follows:
Figure BDA0003264745770000072
1 ≦ i ≦ n, i being the data set X excluding the cluster centernA point of (1);
j is more than or equal to 1 and less than or equal to k, and k is the number of the selected clustering centers;
(2.4) obtaining the minimum distance of each point except the cluster center from all the cluster centers
Figure BDA0003264745770000073
Figure BDA0003264745770000074
1 is more than or equal to i is less than or equal to n, i is a cluster of eliminationOff-center data set XnA point of (1);
(2.5) judging the minimum distance
Figure BDA0003264745770000075
Whether or not:
Figure BDA0003264745770000076
a is the set coefficient in the step (2), and 0< a < 1;
if it is
Figure BDA0003264745770000077
The point i is the cluster of the cluster center closest to the point i;
if it is
Figure BDA0003264745770000078
Selecting the point i as a (k + 1) th clustering center;
(2.6) repeating the operation according to the step (2.3) to the step (2.5), and selecting the Kth clustering center
Figure BDA0003264745770000079
Determining a value of the cluster number K and setting the data set X except the cluster centernThe point in the cluster is divided into a cluster center closest to the point in the cluster to become a cluster of the cluster center;
(3) cluster radius-euclidean distance
The similarity index of the clusters is represented by Euclidean distance, and the selected clustering radius is the maximum value of the distance from all points in the cluster to the clustering center;
through the step (2), K clustering centers are obtained in total, namely K clusters exist,
then the jth cluster center
Figure BDA00032647457700000710
The radius of (a) is:
Figure BDA0003264745770000081
j is more than or equal to 1 and less than or equal to K, i is a cluster in the jth cluster, and m is more than or equal to 1 and less than or equal to n;
therein, a point
Figure BDA0003264745770000082
As the jth cluster center
Figure BDA0003264745770000083
Value of (x)i,yi,...,zi) A value of a cluster-like element belonging to the jth cluster;
obtaining the clustering radius of K clustering centers;
(4) updating cluster centers and iterative outputs
(4.1) calculating a clustering objective function
Set the clustering objective function as
Figure BDA0003264745770000084
Obtaining a clustering objective function divided by K clustering centers, wherein
Figure BDA0003264745770000085
As the jth cluster center
Figure BDA0003264745770000086
A value of (d);
(4.2) calculating to obtain K first-time updated cluster centers by taking the mean value of the cluster elements in the same cluster as the first-time updated cluster centers, and calculating corresponding first-time updated cluster objective functions;
(4.3) if any one of the first updating clustering objective function and the first updating clustering center and the clustering objective function and the clustering center is changed, recalculating the data set XnThe distance of each element from the K first updated cluster centers,
the distance formula is:
Figure BDA0003264745770000087
i is more than or equal to 1 and less than or equal to n, i is a data set XnAll points in (1);
j is more than or equal to 1 and less than or equal to K, and K is the number of clusters, namely the value of K;
obtaining a data set XnEach point in (1) is a minimum distance from all first-update cluster centers
Figure BDA0003264745770000088
Figure BDA0003264745770000089
I is more than or equal to 1 and less than or equal to n, i is a data set XnAll points in (1);
minimum distance
Figure BDA00032647457700000810
The corresponding element is drawn as the minimum distance
Figure BDA00032647457700000811
Correspondingly updating the cluster of the cluster center for the first time;
re-obtaining the cluster elements of the K first-time updating clusters;
(4.4) according to the method in the step (3), obtaining the clustering radius of K first-time updated clustering centers;
(4.5) according to the method in the step (4.2), calculating to obtain K second updated clustering centers by taking the mean value of the clusters in the same cluster as the second updated clustering center, and calculating the corresponding second updated clustering objective function;
(4.6) according to the method described in the step (4.4), if any of the second updated clustering objective function and the second updated clustering center and the first updated clustering objective function and the first updated clustering center is changed, recalculating the data set XnThe distance between each element and K second updated clustering centers is obtained to obtain a data set XnEach point in (1) is a minimum distance from all second-update cluster centers
Figure BDA0003264745770000091
Minimum distance
Figure BDA0003264745770000092
The corresponding element is drawn as the minimum distance
Figure BDA0003264745770000093
Correspondingly updating the cluster of the cluster center for the second time;
re-obtaining the cluster elements of the K second updating clusters;
(4.7) repeating the steps (4.4) - (4.6) until the cluster objective function updated s times and the cluster center updated s times and any one of the cluster objective function updated s-1 times and the cluster center updated s-1 times are not changed, indicating data convergence, outputting the cluster center updated s times, and taking the output cluster center updated s times as representative data, so that the data volume processed in the background can be reduced.
Example 2
A data edge calculation model of a power distribution network comprises the following steps,
(1) in this embodiment, to simply illustrate a specific implementation process, parameters of real-time current (a) and real-time voltage (kV) of a distribution network device are selected and collected to obtain a data set Xn＝{X1(5.1,4.98)，X2(5,5.07)，X3(0.11,0)，X4(0.01,0.01)，X5(6.13,5.77)，X6(6.3,5.77)，X7(6.29,5.79)}；
In the actual working process, according to the complexity of a power grid and the actual operation condition of a power supply company, the selectable state data comprise real-time voltage, real-time current, a voltage transformer, a current transformer, a switch on/off switch, relay protection on/off and temperature; wherein, the real-time voltage selects the current value (unit A) when extracting data; selecting a voltage value (unit kV) when data are extracted from the real-time voltage; primary and secondary current values (unit a) of the voltage transformer; the primary and secondary voltage values (unit kV) of the current transformer, the states of a switch and a relay when data are extracted are respectively selected for switching on and off of the switch and switching off of the relay protection, the states of switching on and switching off of the switch and the relay protection are 1, and the states of switching off and switching off of the switch and the relay protection are 0; selecting a temperature value (unit ℃ C.) when data are extracted;
(2) class number K value determination
Setting a coefficient a to be 0.2, wherein the coefficient a is a convergence condition parameter of a clustering number K value;
(2.1) randomly generating data set XnOne m point is selected as the 1 st clustering center
Figure BDA0003264745770000101
In this example, the 2 nd point X is selected2(5,5.07) as the 1 st clustering center
Figure BDA0003264745770000102
Namely, it is
Figure BDA0003264745770000103
Is (5, 5.07);
(2.2) in data set XnIn the method, the 1 st cluster center is selected
Figure BDA0003264745770000104
The farthest point is taken as the 2 nd clustering center
Figure BDA0003264745770000105
Found by calculation, the 4 th point X4(0.01 ) as the 2 nd clustering center
Figure BDA0003264745770000106
Namely, it is
Figure BDA0003264745770000107
Is (0.01 ); selecting the clustering center after the point selected as the clustering center does not enter;
(2.3) in data set XnIn (2), each point except the cluster center is calculated to each cluster center
Figure BDA0003264745770000108
Is a distance ofThe distance formula is:
Figure BDA0003264745770000109
1 ≦ i ≦ n, i being the data set X excluding the cluster centernA point of (1);
j is more than or equal to 1 and less than or equal to k, and k is the number of the selected clustering centers;
(2.4) obtaining the minimum distance of each point except the cluster center from all the cluster centers
Figure BDA00032647457700001010
Figure BDA00032647457700001011
X1To
Figure BDA00032647457700001012
The distances of (a) are respectively:
Figure BDA00032647457700001013
Figure BDA00032647457700001014
Figure BDA00032647457700001015
(2.5) judging the minimum distance
Figure BDA00032647457700001016
Whether or not:
Figure BDA00032647457700001017
a is the set coefficient in the step (2), and 0< a < 1;
in this example
Figure BDA0003264745770000111
Is a distance of
Figure BDA0003264745770000112
Discovery, for data set Xn1 st point X in (1)1The condition is not satisfied:
Figure BDA0003264745770000113
according to the X obtained by calculation in the step (2.4)1To each cluster center
Figure BDA0003264745770000114
Is a distance of
Figure BDA0003264745770000115
Figure BDA0003264745770000116
Discovery dataset Xn1 st point X in (1)1To
Figure BDA0003264745770000117
Nearest in distance, i.e. X1To be composed of
Figure BDA0003264745770000118
An element of the 1 st cluster of cluster centers;
(2.6) data set Xn2 nd point X in (1)2Is selected as the 1 st cluster center, thus skipping data set Xn2 nd point X in (1)2Directly computing the data set XnPoint 3 in (1)3，
Calculating data set X according to steps (2.3) - (2.5)nPoint 3 in (1)3To each cluster center
Figure BDA0003264745770000119
Figure BDA00032647457700001110
The distances of (a) are respectively:
Figure BDA00032647457700001111
Figure BDA00032647457700001112
Figure BDA00032647457700001113
discovery, for data set XnPoint 3 in (1)3The condition is not satisfied:
Figure BDA00032647457700001114
x obtained by calculation3To each cluster center
Figure BDA00032647457700001115
Is a distance of
Figure BDA00032647457700001116
Discovery dataset XnPoint 3 in (1)3To
Figure BDA00032647457700001117
Nearest in distance, i.e. X3To be composed of
Figure BDA00032647457700001118
An element of the 2 nd cluster that is a cluster center;
(2.7) data set XnAt the 4 th point X in4Is selected as the 2 nd cluster center, thus skippingOver data set XnAt the 4 th point X in4Directly computing the data set XnAt point 5 in (1)5，
Calculating data set X according to steps (2.3) - (2.5)nAt point 5 in (1)5To each cluster center
Figure BDA00032647457700001119
Figure BDA00032647457700001120
The distances of (a) are respectively:
Figure BDA0003264745770000121
Figure BDA0003264745770000122
Figure BDA0003264745770000123
discovery, for data set XnAt point 5 in (1)5The condition is not satisfied:
Figure BDA0003264745770000124
x obtained by calculation5To each cluster center
Figure BDA0003264745770000125
Is a distance of
Figure BDA0003264745770000126
Discovery dataset XnAt point 5 in (1)5To
Figure BDA0003264745770000127
Nearest in distance, i.e. X5To be composed of
Figure BDA0003264745770000128
An element of the 1 st cluster of cluster centers;
(2.8) computing the data set XnAt the 6 th point X in6，
Calculating data set X according to steps (2.3) - (2.5)nAt the 6 th point X in6To each cluster center
Figure BDA0003264745770000129
Figure BDA00032647457700001210
The distances of (a) are respectively:
Figure BDA00032647457700001211
Figure BDA00032647457700001212
ρ6＝min{ρ61,ρ62}＝1.4765；
discovery, for data set XnAt the 6 th point X in6The conditions are satisfied:
Figure BDA00032647457700001213
thus, data set XnAt the 6 th point X in6Is selected as the third cluster center
Figure BDA00032647457700001214
(2.9) computing the data set XnAt 7 th point X in7，
Calculating data set X according to steps (2.3) - (2.5)nAt 7 th point X in7To each cluster center
Figure BDA00032647457700001215
Figure BDA00032647457700001216
The distances of (a) are respectively:
Figure BDA00032647457700001217
Figure BDA00032647457700001218
Figure BDA0003264745770000131
Figure BDA0003264745770000132
discovery, for data set XnAt 7 th point X in7The condition is not satisfied:
Figure BDA0003264745770000133
x obtained by calculation5To each cluster center
Figure BDA0003264745770000134
Is a distance of
Figure BDA0003264745770000135
Figure BDA0003264745770000136
Discovery dataset XnAt 7 th point X in7To
Figure BDA0003264745770000137
More recently, namely X7To be composed of
Figure BDA0003264745770000138
An element of the 3 rd cluster that is the cluster center;
thus, for this embodiment, the number of clusters K is 3, there are 3 cluster centers:
Figure BDA0003264745770000139
Figure BDA00032647457700001310
the list is as follows:
Figure BDA00032647457700001311
(3) cluster radius-euclidean distance
The similarity index of the clusters is represented by Euclidean distance, and the selected clustering radius is the maximum value of the distance from all points in the cluster to the clustering center;
after step 2), there are 3 cluster centers K, i.e. 3 clusters K,
then the jth cluster center
Figure BDA00032647457700001312
The radius of (a) is:
Figure BDA00032647457700001313
j is more than or equal to 1 and less than or equal to K is 3, i is a cluster element in the jth cluster, and m is more than or equal to 1 and less than or equal to n is 7;
wherein, the point (x)i,yi) A value of a cluster-like element belonging to the jth cluster;
according to
Figure BDA00032647457700001314
Calculating to obtain K as 3 clustering radiuses:
Figure BDA00032647457700001315
Figure BDA0003264745770000141
Figure BDA0003264745770000142
Figure BDA0003264745770000143
(4) updating cluster centers and iterative outputs
(4.1) calculating a clustering objective function
Sm＝|X1-M1|2+|X1-M2|2+|X1-M3|2+…+|X7-M1|2+|X7-M2|2+|X7-M3|2To obtain Sm＝572.2357；
(4.2) calculating to obtain K first updated cluster centers by taking the mean value of the cluster elements in the same cluster as the first updated cluster centers, and calculating a corresponding first updated cluster objective function, wherein the cluster radius is recalculated as follows:
first update 1 st cluster center
Figure BDA0003264745770000144
Figure BDA0003264745770000145
Figure BDA0003264745770000146
First update 2 nd cluster center
Figure BDA0003264745770000147
Figure BDA0003264745770000148
Figure BDA0003264745770000149
First update of 3 rd cluster center
Figure BDA00032647457700001410
Figure BDA00032647457700001411
Figure BDA0003264745770000151
And (4) according to the method in the step (3), respectively calculating three corresponding first updated clustering radiuses through the three first updated clustering centers, wherein the results are as follows:
Figure BDA0003264745770000152
calculating a first updated clustering objective function according to the step (4.1): s-579.4782;
(4.3) since the first-time updated clustering objective function and the first-time updated clustering center change from either of the clustering objective function and the clustering center, it is necessary to recalculate the data set X in steps (2.3) - (2.6)nDetermining cluster-like elements of the cluster-like center according to the distance from each element to the first updated cluster center; the mean value of the cluster elements in the same cluster is used as the calculation for updating the cluster center for the second time,
data set XnThe distance result from each element to the first updated cluster center is shown as the following graph:
Figure BDA0003264745770000153
according to step (2), from the data set XnDetermining cluster-like elements of the cluster-like center according to the minimum distance from each element to the cluster-like center updated for the first time, and calculating the cluster-like radius updated for the second time according to the mean value of the cluster-like elements in the same cluster, wherein the specific result is as follows:
Figure BDA0003264745770000154
Figure BDA0003264745770000161
calculating a clustering objective function: 567.9279 ═ S
(4.4) since the second updated clustering objective function and the second updated clustering center are changed from the first updated clustering objective function and the first updated clustering center, it is necessary to recalculate the data set X according to the steps (2.3) - (2.6)nDetermining the cluster of the cluster center according to the distance from each element to the second updated cluster center; updating the cluster center for the third time according to the average value of the cluster elements in the same cluster,
the calculation result according to the step (4.3) is as follows:
Figure BDA0003264745770000162
calculating a clustering objective function: 567.9279 ═ S
And (4) calculating and finding that the clustering objective function and the clustering center of the step (4.3) and the step (4.4) are not changed, indicating data convergence, and outputting the clustering center M { (5.05,5.025), (0.06,0.005), (6.24,5.7767) }.
The output 3 rd time updating cluster center is used as representative data, so that the data volume of background processing can be reduced. In contrast to conventional methods, which require an output dataset Xn7 elements ofIn the invention, only 3 representative data need to be output, so that the data volume of background processing is greatly reduced.
Example 3
A distribution network equipment defect elimination auxiliary system based on edge calculation comprises a distribution network equipment acquisition module and a distribution network communication module,
the distribution network equipment acquisition module comprises a data acquisition unit, a position acquisition unit, a data storage unit and a data alarm unit,
the data acquisition unit acquires state data of the distribution network equipment;
the state data comprises real-time voltage, real-time current, a voltage transformer, a current transformer, a switch on-off switch, relay protection on-off and temperature;
the position acquisition unit acquires geographical position information data of the distribution network equipment;
the data alarm unit compares the state data and/or the geographical position information data with preset data, and when the state data and/or the geographical position information data exceed the preset data range, the data alarm unit sends alarm data, sends the alarm data to the distribution network communication module, and sends the alarm data to the rear-end distribution network system server by the distribution network communication module;
the data storage unit stores historical data of the data acquisition unit and the position acquisition unit and equipment information data of the distribution network equipment, and the historical data can be automatically cleaned in a period selected;
in order to obtain better technical effect, the equipment information data comprises equipment delivery time, equipment manufacturers, equipment specifications and stored equipment information;
the distribution network communication module comprises a data supervision unit, a data processing unit, a data transmission unit and a safety guarantee unit,
the data monitoring unit monitors state data of the distribution network equipment in real time, monitors whether the distribution network equipment normally operates or not, and sends a monitoring signal to a rear-end distribution network system server for recording if the equipment abnormally operates;
the data processing unit preprocesses the state data through the power distribution network data edge calculation model in embodiment 1 or embodiment 2 to obtain preprocessed state data, and uploads the preprocessed state data to a back-end power distribution network system server, so that the information data amount transmitted to a power distribution network system by the data transmission unit is reduced, and the pressure of the power distribution network system in calculating the processed state data is reduced;
the data transmission unit transmits the preprocessed data processed by the data processing unit to a back-end power distribution network system server through optical fibers or wireless.
The safety guarantee unit is an encryption module containing an encryption algorithm, and is used for carrying out encryption protection on various running state data of the distribution network equipment, so that the data is not easy to hijack and crack in the transmission process.
Example 4
A defect elimination method of a distribution network equipment defect elimination auxiliary system based on edge calculation comprises the following steps,
(1) displaying fault information and pre-judged specific information on a web page and a mobile phone terminal by using a prediction result of a rear-end power distribution network system server, and automatically issuing fault short messages to a maintainer and a responsible person of an operation and maintenance class of the geographical information by the system according to the geographical information of the fault;
(2) the maintainers and the operation and maintenance team and group responsible person receive the equipment fault short message and log in the mobile phone terminal to check the detailed fault message;
(3) the operation and maintenance team personnel judge the fault message according to the detailed fault message and whether the fault message needs to be eliminated when going out;
(4) if the fault does not occur emergently, the normal operation of the equipment is not hindered, and the equipment is not suitable for live working, the fault message can be loaded into historical fault messages, and the personnel of the operation and maintenance team and team can go out in batch uniformly to eliminate the fault according to the historical fault messages in the later period;
(5) if the fault needs to be solved, the operation and maintenance team personnel navigate to the position of the fault equipment according to the equipment positioning data acquired by the position acquisition unit of the power distribution equipment acquisition module;
(6) in the operation, maintenance and defect elimination process of the equipment, the wiring diagram and the schematic diagram of the equipment are slightly different due to various distribution network equipment, and the challenges are brought to the refined operation, maintenance and defect elimination. And the operation and maintenance team personnel search relevant information such as the primary and secondary wiring diagrams of the fault equipment through the mobile phone terminal of the system and then accurately eliminate the fault equipment.
The device disclosed by the invention is embedded and installed on the power distribution network equipment, so that the pressure of information sharp increase can be greatly relieved, the pressure of background data processing is relieved, the background data processing time is reduced, and the data processing rate of a power distribution network system is improved. The loss rate of the information is reduced by the aid of the stored related information such as the primary wiring diagram and the secondary wiring diagram of the power distribution network equipment, the efficiency and the accuracy of eliminating the defects of operation and maintenance teams and groups are improved, and the defect eliminating time cannot be delayed due to the fact that different models, different wiring modes and the like of the power distribution network equipment exist.

Claims (7)

1. A data edge calculation model of a power distribution network comprises the following steps,
(1) collecting state data of distribution network equipment to obtain data set
Xn＝{(x1,y1,...,z1),(x2,y2,...,z2),…,(xn,yn,...,zn)}；
(2) Clustering number K value determination
Setting a coefficient a as a convergence condition parameter of the clustering number K value, wherein 0< a < 1;
(2.1) randomly generating data set XnOne m point is selected as the 1 st clustering center
Figure FDA0003264745760000011
Figure FDA0003264745760000012
Represents the 1 st cluster center with m points as the starting point, wherein
Figure FDA0003264745760000013
Denotes m points as1 clustering center in dataset XnThe value of (b) is the value of the starting point m;
(2.2) in data set XnIn the method, the 1 st cluster center is selected
Figure FDA0003264745760000014
The farthest point is taken as the 2 nd clustering center
Figure FDA0003264745760000015
Selecting the clustering center after the point selected as the clustering center does not enter;
(2.3) in data set XnIn the method, the distance from each point except the cluster center to all the cluster centers is calculated, and the distance formula is as follows:
Figure FDA0003264745760000016
1 ≦ i ≦ n, i being the data set X excluding the cluster centernA point of (1);
j is more than or equal to 1 and less than or equal to k, and k is the number of the selected clustering centers;
(2.4) obtaining the minimum distance of each point except the cluster center from all the cluster centers
Figure FDA0003264745760000017
Figure FDA0003264745760000018
1 ≦ i ≦ n, i being the data set X excluding the cluster centernA point of (1);
(2.5) judging the minimum distance
Figure FDA0003264745760000019
Whether or not:
Figure FDA00032647457600000110
a is the set coefficient in the step (2), and 0< a < 1;
if it is
Figure FDA00032647457600000111
The point i is the cluster of the cluster center closest to the point i;
if it is
Figure FDA0003264745760000021
Selecting the point i as a (k + 1) th clustering center;
(2.6) repeating the operation according to the step (2.3) to the step (2.5), and selecting the Kth clustering center
Figure FDA0003264745760000022
Determining a value of the cluster number K and setting the data set X except the cluster centernThe point in the cluster is divided into a cluster center closest to the point in the cluster to become a cluster of the cluster center;
(3) cluster radius-euclidean distance
The similarity index of the clusters is represented by Euclidean distance, and the selected clustering radius is the maximum value of the distance from all points in the cluster to the clustering center;
through the step (2), K clustering centers are obtained in total, namely K clusters exist,
then the jth cluster center
Figure FDA0003264745760000023
The radius of (a) is:
Figure FDA0003264745760000024
i is a cluster in the jth cluster, and m is more than or equal to 1 and less than or equal to n;
wherein, the point (x)i,yi,...,zi) A value belonging to a class cluster in the jth cluster;
obtaining the clustering radius of K clustering centers;
(4) updating cluster centers and iterative outputs
(4.1) calculating a clustering objective function
Set the clustering objective function as
Figure FDA0003264745760000025
Obtaining a clustering objective function divided by K clustering centers;
(4.2) calculating to obtain K first updated cluster centers by taking the mean value of the clusters in the same cluster as the first updated cluster centers, and calculating corresponding first updated cluster objective functions;
(4.3) if any one of the first updating clustering objective function and the first updating clustering center and the clustering objective function and the clustering center is changed, recalculating the data set XnThe distance of each element from the K first updated cluster centers,
the distance formula is:
Figure FDA0003264745760000031
i is more than or equal to 1 and less than or equal to n, i is a data set XnAll points in (1);
j is more than or equal to 1 and less than or equal to K, and K is the number of clusters, namely the value of K;
obtaining a data set XnEach point in (1) is a minimum distance from all first-update cluster centers
Figure FDA0003264745760000032
Figure FDA0003264745760000033
I is more than or equal to 1 and less than or equal to n, i is a data set XnAll points in (1);
minimum distance
Figure FDA0003264745760000034
The corresponding element is drawn as the minimum distance
Figure FDA0003264745760000035
Correspondingly updating the cluster of the cluster center for the first time;
re-obtaining the cluster elements of the K first-time updating clusters;
(4.4) according to the method in the step (3), obtaining the clustering radius of K first-time updated clustering centers;
(4.5) according to the method in the step (4.2), calculating to obtain K second updated clustering centers by taking the mean value of the clusters in the same cluster as the second updated clustering center, and calculating the corresponding second updated clustering objective function;
(4.6) according to the method described in the step (4.4), if any of the second updated clustering objective function and the second updated clustering center and the first updated clustering objective function and the first updated clustering center is changed, recalculating the data set XnThe distance between each element and K second updated clustering centers is obtained to obtain a data set XnEach point in (1) is a minimum distance from all second-update cluster centers
Figure FDA0003264745760000036
Minimum distance
Figure FDA0003264745760000037
The corresponding element is drawn as the minimum distance
Figure FDA0003264745760000038
Correspondingly updating the cluster of the cluster center for the second time;
re-obtaining the cluster elements of the K second updating clusters;
(4.7) repeating the steps (4.4) - (4.6) until the cluster objective function updated s times and the cluster center updated s times and any one of the cluster objective function updated s-1 times and the cluster center updated s-1 times are not changed, indicating data convergence, and outputting the cluster center updated s times.
2. A distribution network equipment defect elimination auxiliary system based on the power distribution network data edge calculation model of claim 1, which comprises a distribution network equipment acquisition module and a distribution network communication module,
the distribution network equipment acquisition module comprises a data acquisition unit, a position acquisition unit and a data alarm unit,
the data acquisition unit acquires state data of the distribution network equipment;
the position acquisition unit acquires geographical position information data of the distribution network equipment;
the data alarm unit compares the state data and/or the geographical position information data with preset data, and when the state data and/or the geographical position information data exceed the preset data range, the data alarm unit sends alarm data, sends the alarm data to the distribution network communication module, and sends the alarm data to the rear-end distribution network system server by the distribution network communication module;
the distribution network communication module comprises a data supervision unit, a data processing unit and a data transmission unit,
the data monitoring unit monitors state data of the distribution network equipment in real time, monitors whether the distribution network equipment normally operates or not, and sends a monitoring signal to a rear-end distribution network system server for recording if the equipment abnormally operates;
the data processing unit preprocesses the state data through the power distribution network data edge calculation model to obtain preprocessed state data, and uploads the preprocessed state data to a rear-end power distribution network system server, so that the information data amount transmitted to the power distribution network system by the data transmission unit is reduced, and the pressure of the power distribution network system in calculating and processing the state data is reduced;
the data transmission unit transmits the preprocessed data processed by the data processing unit to a back-end power distribution network system server through optical fibers or wireless.
3. The distribution network equipment defect elimination auxiliary system of claim 2, wherein the status data comprises real-time voltage, real-time current, a voltage transformer, a current transformer, a switch on/off switch, relay protection on/off and temperature.
4. The distribution network equipment defect elimination auxiliary system as claimed in claim 2, wherein the distribution network equipment acquisition module further comprises a data storage unit, the data storage unit stores historical data of the data acquisition unit and the position acquisition unit, and equipment information data of the distribution network equipment, and the historical data can be automatically cleaned in a period selected.
5. The distribution network equipment defect-eliminating auxiliary system of claim 2, wherein the equipment information data comprises equipment information stored in equipment factory time, equipment manufacturer, equipment specifications.
6. The distribution network equipment defect elimination auxiliary system as claimed in claim 2, wherein the distribution network communication module further comprises a security guarantee unit, and the security guarantee unit is an encryption module containing an encryption algorithm, and is used for performing encryption protection on various operation state data of the distribution network equipment to ensure that the data is not easy to be hijacked and cracked in the transmission process.
7. A method for eliminating the defect of the auxiliary system for eliminating the defect of the distribution network equipment in any claim of 2 to 6,
(1) displaying fault information and pre-judged specific information on a web page and a mobile phone terminal by using a prediction result of a rear-end power distribution network system server, and automatically issuing fault short messages to a maintainer and a responsible person of an operation and maintenance class of the geographical information by the system according to the geographical information of the fault;
(2) the maintainers and the operation and maintenance team and group responsible person receive the equipment fault short message and log in the mobile phone terminal to check the detailed fault message;
(3) the operation and maintenance team personnel judge the fault message according to the detailed fault message and whether the fault message needs to be eliminated when going out;
(4) if the fault does not occur emergently, the normal operation of the equipment is not hindered, and the equipment is not suitable for live working, the fault message can be loaded into historical fault messages, and the personnel of the operation and maintenance team and team can go out in batch uniformly to eliminate the fault according to the historical fault messages in the later period;
(5) if the fault needs to be solved, the operation and maintenance team personnel navigate to the position of the fault equipment according to the equipment positioning data acquired by the position acquisition unit of the power distribution equipment acquisition module;
(6) in the operation, maintenance and defect elimination process of the equipment, the wiring diagram and the schematic diagram of the equipment are slightly different due to various distribution network equipment, and the challenges are brought to the refined operation, maintenance and defect elimination. And the operation and maintenance team personnel search the related information such as the primary and secondary wiring diagrams of the fault equipment through the mobile phone terminal of the system and then accurately eliminate the fault equipment.
CN202111082952.9A 2020-12-29 2021-09-15 Distribution network data edge calculation model of distribution network equipment and defect elimination auxiliary method and system Active CN113837458B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2020115898639 2020-12-29
CN202011589863 2020-12-29

Publications (2)

Publication Number Publication Date
CN113837458A true CN113837458A (en) 2021-12-24
CN113837458B CN113837458B (en) 2023-05-16

Family

ID=78959595

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111082952.9A Active CN113837458B (en) 2020-12-29 2021-09-15 Distribution network data edge calculation model of distribution network equipment and defect elimination auxiliary method and system

Country Status (1)

Country Link
CN (1) CN113837458B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104899229A (en) * 2014-03-07 2015-09-09 上海市玻森数据科技有限公司 Swarm intelligence based behavior clustering system
CN106022614A (en) * 2016-05-22 2016-10-12 广州供电局有限公司 Data mining method of neural network based on nearest neighbor clustering
JP2016206985A (en) * 2015-04-23 2016-12-08 日本電信電話株式会社 Clustering device, method, and program
CN106710164A (en) * 2017-03-06 2017-05-24 国电南瑞科技股份有限公司 Power distribution network fault early warning method aiming at multiple factors
CN109145957A (en) * 2018-07-26 2019-01-04 国网浙江省电力有限公司温州供电公司 The identification and processing method and processing device of power distribution network abnormal index based on big data
CN111564906A (en) * 2020-06-16 2020-08-21 国网湖北省电力有限公司黄石供电公司 Method for efficiently processing automation defects of urban network distribution network

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104899229A (en) * 2014-03-07 2015-09-09 上海市玻森数据科技有限公司 Swarm intelligence based behavior clustering system
JP2016206985A (en) * 2015-04-23 2016-12-08 日本電信電話株式会社 Clustering device, method, and program
CN106022614A (en) * 2016-05-22 2016-10-12 广州供电局有限公司 Data mining method of neural network based on nearest neighbor clustering
CN106710164A (en) * 2017-03-06 2017-05-24 国电南瑞科技股份有限公司 Power distribution network fault early warning method aiming at multiple factors
CN109145957A (en) * 2018-07-26 2019-01-04 国网浙江省电力有限公司温州供电公司 The identification and processing method and processing device of power distribution network abnormal index based on big data
CN111564906A (en) * 2020-06-16 2020-08-21 国网湖北省电力有限公司黄石供电公司 Method for efficiently processing automation defects of urban network distribution network

Also Published As

Publication number Publication date
CN113837458B (en) 2023-05-16

Similar Documents

Publication Publication Date Title
CN111650921A (en) Intelligent power grid regulation and control system equipment fault diagnosis method and system
CN108471168A (en) A kind of substation&#39;s wireless data transmission and inspection base and method
CN106204330A (en) A kind of power distribution network intelligent diagnosis system
CN103647276A (en) Electric energy quality early warning system and method thereof
CN112713658A (en) Intelligent control method and system for monitoring defects of power grid equipment
CN117078017A (en) Intelligent decision analysis system for monitoring power grid equipment
CN111579917A (en) Low-voltage distribution network fault positioning system and method
CN113409162A (en) Centralized control station equipment monitoring system and control method based on intelligent event processing
CN108921443A (en) A kind of power grid early warning scheduling system, method, apparatus and server
CN107295553A (en) A kind of base station maintenance method and device
CN113708350A (en) Power distribution station heavy overload abnormity judgment method and system based on cloud edge cooperation
CN113837458A (en) Edge calculation model of power distribution network data, and distribution network equipment defect elimination auxiliary system and defect elimination method based on edge calculation model
CN112783939A (en) Low-voltage distribution network running state evaluation method based on data mining
CN104134168A (en) Data management system
CN116094174A (en) Knowledge graph-based power grid operation and maintenance monitoring method, system, equipment and medium
CN116884159A (en) Intelligent fire-fighting early warning method and system based on AI (advanced technology attachment) identification
CN111142003A (en) System and method for power capacitor state monitoring and reliability analysis
CN114741447B (en) Distributed energy station data processing method and device
CN114501755B (en) Lighting facility monitoring method, system, device and computer readable storage medium
CN112766636B (en) Distribution transformer power failure judgment method based on electric quantity information acquired by power consumption information acquisition terminal
CN112101422B (en) Typical case self-learning method for power system fault case
CN113129160A (en) Electric power communication network inspection method based on equipment state perception and intellectualization
CN116663785B (en) Topology identification-based low-voltage fault positioning method
CN116298675B (en) Intelligent algorithm-based power distribution network line loss anomaly detection method and system
CN113159993B (en) Knowledge base construction method and system based on power system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant