Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Particle swarm principle-based power distribution network data preprocessing method adopting big data clustering

A technology for data preprocessing and distribution network, applied in data processing applications, electrical digital data processing, special data processing applications, etc., can solve problems such as inconsistent conclusions, achieve fast convergence speed, good noise removal effect, and overcome the tendency to fall into The effect of local extrema

Active Publication Date: 2020-12-11
SHANGHAI MUNICIPAL ELECTRIC POWER CO
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Aiming at the shortcomings of the bending moment method and the concave-convex coefficient method in the selection of the optimal N value of clustering at present, the particle swarm algorithm is combined with the bending moment method and the concave-convex coefficient algorithm, and the bending moment method is the main method, and the concave-convex coefficient method is supplemented The selection mechanism of the number of clusters, and simultaneously output the optimized distribution network data that eliminates outliers

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Particle swarm principle-based power distribution network data preprocessing method adopting big data clustering
  • Particle swarm principle-based power distribution network data preprocessing method adopting big data clustering
  • Particle swarm principle-based power distribution network data preprocessing method adopting big data clustering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0083] According to the method described in this patent, random data collection is carried out according to the data sources of the distribution line online monitoring system and the intelligent public distribution transformer monitoring system. There are 962 samples in total. In the same interval [0,1], participate in classification with the same magnitude, that is, perform normalization processing, and merge data from multiple sources into a consistent database storage.

[0084] Take the average value of the k-th dimension data of n samples first and standard deviation C,

[0085]

[0086] This gives the normalized value x of the original data ik ':

[0087]

[0088] Using Matlab programming, the curve calculated by the bending moment method is as follows figure 1 shown in .

[0089] Using Matlab programming, the curve calculated by the concave-convex coefficient method is as follows figure 2 shown in .

[0090] It can be seen that the maximum N value of the co...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a particle swarm principle-based power distribution network data preprocessing method adopting big data clustering, and belongs to the field of power distribution network reliability prediction. For the normalized power distribution network data, a clustering number selection mechanism which takes a bending moment method as a main method and takes a concave-convex coefficient method is adopted as an auxiliary method to obtain an optimal clustering number of a sample; after clustering analysis is carried out on the samples, abnormal value identification standards of upper and lower critical graphs are adopted to delimit diagnosis thresholds; if the distance between the sample and the clustering center is greater than a diagnosis threshold,a judgment is made that thesample is an outlier sample and removed. further, de-noised sample data is obtained; Potential rule of the power distribution network fault is predicted by adopting the sample data subjected to de-noising. Not only is the defect that a bending moment method and a concave-convex coefficient algorithm are easy to fall into a local extreme value overcome, but also global optimization of a particle swarm algorithm is maintained, and convergence speed of a bending moment method and a concave-convex coefficient algorithm is relatively high; the invention has the advantages of being good in denoisingeffect and high in sorting accuracy and effectiveness.

Description

technical field [0001] The invention belongs to the field of reliability prediction of distribution networks, and in particular relates to a distribution network data preprocessing method based on particle swarm principle using big data clustering. Background technique [0002] In recent years, the State Grid Corporation of China has intensified the transformation of distribution network automation and deepened the promotion and application of distribution automation systems. And the protection signal of the automatic switch, automatically judge the fault area, send a prompt signal to the control personnel or automatically complete the fault isolation and restore power supply, which improves the reliability of power supply and the quality of transmission. [0003] The power system is a unified whole of production, transmission, distribution, and consumption of electric energy, and any power system failure will affect users. According to statistics, more than 80% of user fai...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F30/18G06F30/25G06F16/906G06Q10/04G06Q50/06H02J3/00G06F113/04G06F119/02
CPCG06F30/18G06F30/25G06F16/906G06Q10/04G06Q50/06H02J3/00G06F2113/04G06F2119/02H02J2203/20Y04S10/50Y02E40/70
Inventor 吴峥嵘石江华周蓝波宋祎波李俊颖忻葆宏张萌亮宗卫国顾珏曹轶毅
Owner SHANGHAI MUNICIPAL ELECTRIC POWER CO
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products