Active distribution network reliability fast evaluation method based on improved AdaBoost. M1-SVM

An adaboost.m1-svm and reliability technology, applied in the field of electrical information, can solve the problems of one-sidedness of evaluation results, lack of authenticity, and non-objectivity, scientificity, accuracy and speed of reliability evaluation, and achieve good results. Search ability, effect of improving weight growth factor

Active Publication Date: 2019-01-08
国网山东省电力公司聊城供电公司 +2
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a method for rapidly evaluating the reliability of active distribution networks based on improved AdaBoost.M1-SVM. The lack of authenticity makes the whole reliability evaluation not objective, scientific, accurate and fast

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
  • Active distribution network reliability fast evaluation method based on improved AdaBoost. M1-SVM
  • Active distribution network reliability fast evaluation method based on improved AdaBoost. M1-SVM
  • Active distribution network reliability fast evaluation method based on improved AdaBoost. M1-SVM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0060] A method for rapidly evaluating the reliability of active distribution networks based on the improved AdaBoost.M1-SVM includes the following steps:

[0061] 1) Mathematical model based on the improved AdaBoost.M1-SVM rapid evaluation method for active distribution network reliability:

[0062]

[0063] where H T (x) is the output ensemble classifier, T is the number of iterations, α t for SVM t (x) weight coefficient, SVM t (x) is the tth iteration of the input state variable x weak classifier SVM, sign is the sign function, ||α|| 1 for alpha t The 1-norm of .

[0064] 2) Improve the basic principle of AdaBoost.M1-SVM method

[0065] Before building a state recognizer, extracting the input variables most relevant to the operating state of the distribution network is an important aspect of building an identification model and reducing the feature space. The input state variables that can be used to analyze the operating state of the system include: system load l...

Embodiment 2

[0117] Based on the rapid evaluation method for the reliability of the active distribution network in Embodiment 1, the present embodiment adopts the IEEE RBTs-Bus6 feeder F4 test system to simulate the network connection of the active distribution network for reliability analysis, such as figure 2 As shown, the distributed fan power supply is connected between Bus10-Bus19, Bus15-Bus16, and Bus15-Bus25, and the test system line failure rate is 0.039 times / (a km), and the circuit breaker and fuse rejection probability is 0.02 times / In 2019, the failure rate of transformers was 0.021 times / year, and the average failure repair time was 6 hours; the failure state probability of wind turbines was 0.052 times / year, and the peak load of the system was 7.93MW. System load model According to the annual time series load data of IEEE-RTS79 system, the probability density distribution is obtained by non-parametric kernel density estimation, and the correlation coefficient of node load is...

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 an active distribution network reliability fast evaluation method based on an improved SVM. The method introduces the improved AdaBoost. M1-SVM algorithm into the distributionnetwork reliability evaluation. The improved AdaBoost. M1-SVM algorithm uses AdaBoost technology to integrate multiple SVM weak classifiers. In the improved AdaBoost. M1-SVM algorithm, the bat algorithm is used to optimize the c-parameter and g-parameter of SVM in the training process, and the local search is introduced, which has better searching ability. The invention weakens the error weight ofmissed judgment samples, minimizes the total number of misjudgment samples, and overcomes the defect that a single classifier can not achieve effective balance in classification accuracy and generalization ability.

Description

technical field [0001] The invention relates to the field of electrical information technology, in particular to a method for rapidly evaluating the reliability of an active distribution network based on the improved AdaBoost.M1-SVM. Background technique [0002] The reliability assessment of the active distribution network refers to the ability of the power system to provide power and electricity to power users uninterruptedly according to acceptable quality standards and required quantities. The reliability of the power grid includes the meaning of adequacy, that is, when the power grid is running stably, it is within the allowable range of the grid component capacity, bus voltage and system frequency; considering the planned outage and reasonable unplanned outage conditions of the grid components, the The ability of the user to provide all required power and charge. And use this to determine the technical measures to improve the reliability of power supply and seek manag...

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): G06Q10/06G06Q50/06
CPCG06Q10/0639G06Q50/06
Inventor 尹晓敏杨延勇王华莹许强朱辉赵飞桃董丽丽雷霞丁吉吴卓聪
Owner 国网山东省电力公司聊城供电公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products