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

Transient assessment evaluation feature selection method and device

A feature selection method and a feature selection technology, applied in the direction of instruments, data processing applications, resources, etc., can solve the problems that the stable samples are not considered, the unstable samples are very few, and the criteria for determining the selection threshold are not given, so as to reduce the impact , to ensure the effect of rationality

Pending Publication Date: 2017-06-27
CHINA ELECTRIC POWER RES INST +3
View PDF4 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the existing technology can use random forest to rank the importance of features, it does not consider the characteristics of many stable samples and very few unstable samples in the online historical data of the actual power system, and does not give the criteria for determining the selection threshold

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
  • Transient assessment evaluation feature selection method and device
  • Transient assessment evaluation feature selection method and device
  • Transient assessment evaluation feature selection method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0104] Embodiment 1. Construction of sample set

[0105] This patent uses the New England 39-node system as the test system, and the system structure is as follows figure 2 As shown, it contains 10 generators, 39 nodes, 46 branches, and 19 load points. The generator model adopts a 4th-order model, and the load model is a constant impedance model.

[0106] During the operation of the power system, the power flow state of the random power grid should satisfy the power flow equation, and the equation is f(A,p,D,u,x)=0. Among them, A is an association matrix, which represents the structural variables of the network, which is determined by the topology of the power grid; p is the parameters of network components, such as transmission line parameters, transformer parameters, etc.; D is an uncontrollable variable, such as the system load in a general sense; u is a control variable , such as generator active power and machine terminal voltage, capacitive reactor switching, etc.; x ...

Embodiment 2

[0111] Embodiment 2. Construction of original input features

[0112] A key step in realizing transient stability assessment based on machine learning technology is to select reasonable state quantities as the input features of the classifier. For a specific power system, when factors such as generator distribution, generator output level, load level, load distribution, and fault conditions of the system are determined, the stability level of the system is determined. Therefore, the present invention uses the steady-state operation information before the failure as the candidate input feature set of the original feature set. When the steady-state operating variables are used as the original feature set, the evaluation speed is fast, and online evaluation does not require numerical simulation; generally, the key features identified through feature selection are the operating variables that operators should focus on monitoring. Moreover, using steady-state operating variables a...

Embodiment 3

[0115] Embodiment 3, feature selection results

[0116] The random forest algorithm needs to set two main parameters: the number of trees in the random forest and the number of preselected features in the tree nodes. The number of preselected features in the tree node takes the default value (p is the number of features in the training set), and the number of decision trees is set to 300.

[0117] The present invention compares the effect of feature selection using the WRF-RFE method formed by weighted random forest and the RF-RFE method (Random Forest-Recursive Feature Elimination) formed by ordinary random forest in the process of combining the recursive feature elimination strategy. For the class with a small number of samples, a larger weight is set, and a reasonable setting of the weight can improve the effect of feature selection. In the invention, the weight value of the stable sample and the unstable sample is set to 1:3.

[0118] According to the relationship betwe...

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 present invention provides a transient assessment evaluation feature selection method and device. The method comprises the following steps that: a weighted random forest model is constructed; features with the lowest importance are deleted from calculated feature importance; final feature sequencing is performed on the retained technical features; and features which are high in ranking are retained according to OOB (Out-Of-Bag) error rate curve changes. According to the transient assessment evaluation feature selection method and device provided by the technical schemes of the invention, few unstable sample data weight factors are introduced into the weighted random forest, so that the influence of unbalanced data on feature selection can be decreased, feature subsets of which the performance can be better than that of feature subsets selected by an ordinary random forest algorithm can be selected; and threshold values are not required to be set manually, and the rationality of the selected feature subsets can be ensured.

Description

technical field [0001] The invention belongs to the technical field of geographic system security and stability analysis, and in particular relates to a transient evaluation feature selection method and device. Background technique [0002] Power system transient stability assessment (TSA) is one of the important means to ensure the safe and stable operation of the power system. With the gradual formation of UHV AC / DC hybrid power grid pattern, the safety and stability characteristics and mechanism of power system are becoming more and more complex, and the operation control of power grid is becoming more and more difficult, which puts forward new requirements for the accurate evaluation of power system transient stability. With the rapid development of computer technology in recent years, the transient stability assessment method based on machine learning technology has the advantages of fast online assessment and potential information mining, and has a good development pro...

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
IPC IPC(8): G06Q10/06G06Q50/06
CPCG06Q10/06393G06Q50/06Y04S10/50
Inventor 张春张军于之虹杨超平鲁广明张爽戴红阳高峰田蓓田芳马军李岩松马天东苏明昕
Owner CHINA ELECTRIC POWER RES INST
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com