资源描述：
《应用信度分配的模糊CMAC实现非线性系统的容错控制》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。
1、Vol.32,No.3ACTAAUTOMATICASINICAMay,2006Fault-tolerantControlofNonlinearSystemUsingCreditAssignFuzzyCMAC')ZHUDa-QiKONGMin(ResearchCen￡二ofControlScienceandEngineering,SouthernYangtzeUniversity,Wuxi214122)(E-mail:zdg367Qyahoo.com.cn,zhizi0708Qsohu.com)AbstractTheadaptivefault-tolerantco
2、ntrolschemeofdynamicnonlinearsystembasedonthecreditassignedfuzzyCMACneuralnetworkispresented.Theproposedlearningapproachusesthelearnedtimesofaddressedhypercubesasthecredibility,theamountsofcorrectingerrorsareproportionaltotheinversionofthelearnedtimesofaddressedhypercubes.Withthiside
3、a,thelearningspeedcanindeedbeimproved.BasedontheimprovedCMAClearningapproachandusingtheslidingcontroltechnique,theeffectivecontrollawreconfigurationstrategyispresented.Thesystemstabilityandperformanceareanalyzedunderfailurescenarios.Thenumericalsimulationdemonstratestheeffectivenesso
4、ftheimprovedCMACalgorithmandtheproposedfault-tolerantcontroller.KeywordsCreditassignedfuzzyCMAC,faultdiagnosis,fault-tolerantcontrol,nonlinearsystem1IntroductionDuringthelasttwodecades,extensiveresearchactivitieshavefocusedonfault-tolerantcontrol(FTC)tomaintainthesystemstabilityandto
5、avoidlossesundervariousfailurescenarios.Reference[1-3]providedexcellentoverviewsofrecentresearchworkonFTC.However,linksbetweenfaultdiagnosisandFTCtechniquesarestilllackinglll，andsomerecentresultsontheintegrationoffaultdiagnosiswithFTCcanbefoundin[4-刘.Fromtheserecentresearchresults,tw
6、okeycomponentsmustberesearchedfurther:1)theonlinefaultdiagnosismoduleconsistingofthefaultdetectionmechanismandonlinefaultestimator;and2)thecontrollermoduleconsistingofnominalcontrollerandafault-tolerantcontroller.ThefaultinformationgeneratedbydiagnosisprocedurecanbeveryusefultoFTC,an
7、daccuracyandspeedoffaultdiagnosisareveryimportantforactivefault-tolerantcontrol.Withdevelopmentofartificialneuralnetwork,ithasstrongsuperiorityinfaultdiagnosis,especiallyforthecasesthatcannotbeexpressed饰formula,andincomplexnonlinearsituationwithstrongaccommodation.Invariousneuralnetw
8、orks,theBP(b
