schliessen

Filtern

 

Bibliotheken

THE RESEARCH ON FLUX LINKAGE CHARACTERISTIC BASED ON BP AND RBF NEURAL NETWORK FOR SWITCHED RELUCTANCE MOTOR

The flux and torque of switched reluctance motor (SRM) have a highly nonlinear functional relationship with rotor position and phase current, as a consequence of the double-salient structure of the stator and rotor pole and highly magnetic saturation, which is difficult to build an accurate analytic... Full description

Journal Title: Progress In Electromagnetics Research M 2014, Vol.35, pp.151-161
Main Author: Cai, Yan
Other Authors: Sun, Siyuan , Wang, Chenhui , Gao, Chao
Format: Electronic Article Electronic Article
Language: English
Subjects:
Quelle: CrossRef
ID: E-ISSN: 1937-8726 ; DOI: http://dx.doi.org/10.2528/PIERM14011604
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: crossref10.2528/PIERM14011604
title: THE RESEARCH ON FLUX LINKAGE CHARACTERISTIC BASED ON BP AND RBF NEURAL NETWORK FOR SWITCHED RELUCTANCE MOTOR
format: Article
creator:
  • Cai, Yan
  • Sun, Siyuan
  • Wang, Chenhui
  • Gao, Chao
subjects:
  • Physics
ispartof: Progress In Electromagnetics Research M, 2014, Vol.35, pp.151-161
description: The flux and torque of switched reluctance motor (SRM) have a highly nonlinear functional relationship with rotor position and phase current, as a consequence of the double-salient structure of the stator and rotor pole and highly magnetic saturation, which is difficult to build an accurate analytic model. In order to achieve the SRM high-performance control, it is necessary to build an accurate nonlinear model for SRM. On the basis of the adequate and precise sample data, by taking advantage of neural network that has outstanding nonlinear mapping capability, this paper adopts the Back Propagation (BP) based on Levenberg-Marquardt (LM) algorithm and Radial Basis Function (RBF) neural networkto build offline models for SRM respectively. Under different requirements of model accuracy, two kinds of network are studied and compared with each other on accuracy, scale and other aspects. The research results indicate that the network scale built as SRM nonlinear model by BP neural network based on LM algorithm is smaller than the one built by RBF. Additionally, the model accuracy is higher. In terms of the Switched Reluctance Drive (SRD) which requires real-time controller, reducing the network scale will be beneficial to the online real-time control of the system.
language: eng
source: CrossRef
identifier: E-ISSN: 1937-8726 ; DOI: http://dx.doi.org/10.2528/PIERM14011604
fulltext: fulltext
issn:
  • 19378726
  • 1937-8726
url: Link


@attributes
ID548057899
RANK0.07
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
LOCALfalse
PrimoNMBib
record
control
sourcerecordid10.2528/PIERM14011604
sourceidcrossref
recordidTN_crossref10.2528/PIERM14011604
sourceformatXML
sourcesystemOther
display
typearticle
titleTHE RESEARCH ON FLUX LINKAGE CHARACTERISTIC BASED ON BP AND RBF NEURAL NETWORK FOR SWITCHED RELUCTANCE MOTOR
creatorCai, Yan ; Sun, Siyuan ; Wang, Chenhui ; Gao, Chao
ispartofProgress In Electromagnetics Research M, 2014, Vol.35, pp.151-161
identifierE-ISSN: 1937-8726 ; DOI: http://dx.doi.org/10.2528/PIERM14011604
languageeng
sourceCrossRef
lds412014
descriptionThe flux and torque of switched reluctance motor (SRM) have a highly nonlinear functional relationship with rotor position and phase current, as a consequence of the double-salient structure of the stator and rotor pole and highly magnetic saturation, which is difficult to build an accurate analytic model. In order to achieve the SRM high-performance control, it is necessary to build an accurate nonlinear model for SRM. On the basis of the adequate and precise sample data, by taking advantage of neural network that has outstanding nonlinear mapping capability, this paper adopts the Back Propagation (BP) based on Levenberg-Marquardt (LM) algorithm and Radial Basis Function (RBF) neural networkto build offline models for SRM respectively. Under different requirements of model accuracy, two kinds of network are studied and compared with each other on accuracy, scale and other aspects. The research results indicate that the network scale built as SRM nonlinear model by BP neural network based on LM algorithm is smaller than the one built by RBF. Additionally, the model accuracy is higher. In terms of the Switched Reluctance Drive (SRD) which requires real-time controller, reducing the network scale will be beneficial to the online real-time control of the system.
subjectPhysics;
lds50peer_reviewed
links
openurl$$Topenurl_article
thumbnail
0$$TPCamazon_thumb
1$$TPCgoogle_thumb
openurlfulltext$$Topenurlfull_article
addlink$$Uhttp://exlibris-pub.s3.amazonaws.com/aboutCrossref.html$$EView_CrossRef_copyright_notice
search
creatorcontrib
0Cai, Yan
1Sun, Siyuan
2Wang, Chenhui
3Gao, Chao
titleTHE RESEARCH ON FLUX LINKAGE CHARACTERISTIC BASED ON BP AND RBF NEURAL NETWORK FOR SWITCHED RELUCTANCE MOTOR
general
0English
110.2528/PIERM14011604
2CrossRef
sourceidcrossref
recordidcrossref10.2528/PIERM14011604
issn
019378726
11937-8726
rsrctypearticle
addtitle
0Progress In Electromagnetics Research M
1PIER M
searchscope
0crossref_rest
1CrossRef
2Crossref
3crossref
scope
0crossref_rest
1CrossRef
2Crossref
3crossref
lsr402014
lsr412014
lsr432014
lsr452014
tmp012
tmp022
creationdate2014
startdate20140101
enddate20141231
citationpf 151 pt 161 vol 35
lsr30VSR-Enriched:[description, subject]
sort
titleTHE RESEARCH ON FLUX LINKAGE CHARACTERISTIC BASED ON BP AND RBF NEURAL NETWORK FOR SWITCHED RELUCTANCE MOTOR
authorCai, Yan ; Sun, Siyuan ; Wang, Chenhui ; Gao, Chao
creationdate20140000
lso0120140000
facets
frbrgroupid7212336808969899281
frbrtype5
languageeng
creationdate2014
prefilterarticles
rsrctypearticles
creatorcontrib
0Cai, Yan
1Sun, Siyuan
2Wang, Chenhui
3Gao, Chao
jtitleProgress In Electromagnetics Research M
toplevelpeer_reviewed
delivery
delcategoryRemote Search Resource
fulltextfulltext
addata
aulast
0Cai
1Sun
2Wang
3Gao
aufirst
0Yan
1Siyuan
2Chenhui
3Chao
au
0Cai, Yan
1Sun, Siyuan
2Wang, Chenhui
3Gao, Chao
atitleTHE RESEARCH ON FLUX LINKAGE CHARACTERISTIC BASED ON BP AND RBF NEURAL NETWORK FOR SWITCHED RELUCTANCE MOTOR
jtitleProgress In Electromagnetics Research M
stitlePIER M
date2014
risdate2014
adddate2014
volume35
spage151
epage161
pages151-161
issnProgress In Electromagnetics Research M
eissn1937-8726
formatjournal
genrearticle
ristypeJOUR
doi10.2528/PIERM14011604
lad012