A comprehensive real‐time state evaluation strategy for distribution equipment based on MI‐PSO‐BP algorithm

Summary This paper studies a comprehensive state assessment strategy for distribution equipment with the aim to enhance the reliability and real‐time performance of the state evaluation. In an implementation framework of two‐stage evaluation, a qualitative assessment method is employed to carry out...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:International transactions on electrical energy systems Ročník 29; číslo 11
Hlavní autori: Zhichun, Yang, Yu, Shen, Fan, Yang, Zilin, Wan, Jian, Le
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Hoboken John Wiley & Sons, Inc 01.11.2019
Predmet:
Algorithms
BP neural network
Computer simulation
distribution equipment
Evaluation
Mathematical models
mutual information
Neural networks
Particle swarm optimization
Performance evaluation
Prediction models
real‐time assessment
Reliability analysis
Strategy
Training
ISSN:2050-7038, 2050-7038
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Summary This paper studies a comprehensive state assessment strategy for distribution equipment with the aim to enhance the reliability and real‐time performance of the state evaluation. In an implementation framework of two‐stage evaluation, a qualitative assessment method is employed to carry out the initial state evaluation, in order to provide sufficient training data samples for the real‐time evaluation in the second stage. To improve the accuracy and timelessness of the distribution equipment state evaluation, we use a mutual information‐based technology to refine the training data set and apply particle swarm optimization approach to optimize the parameters of the neural network‐based prediction model that is crucial to the real‐time evaluation. Numerical simulation results are provided to demonstrate the performance of the developed strategy.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2050-7038
2050-7038
DOI:10.1002/2050-7038.12108