首页> 外文期刊>AEU: Archiv fur Elektronik und Ubertragungstechnik: Electronic and Communication >Design ofageneralproposeneuro-fuzzycontrollerbyusingmodified adaptive-network-basedfuzzyinferencesystem
【24h】

Design ofageneralproposeneuro-fuzzycontrollerbyusingmodified adaptive-network-basedfuzzyinferencesystem

机译:利用改进的基于自适应网络的模糊推理系统设计通用的神经模糊控制器

获取原文
获取原文并翻译 | 示例
           

摘要

This paperpresentsanewgeneralpurposeneuro-fuzzycontrollertorealizeadaptive-network-basedfuzzyinferencesys- tem (ANFIS)architecture.Atwoinput,singleoutput,16rulesANFISarchitectureisdesignedin0.35m standardCMOS technology.Thiscontrollercanalsobeusedasastandard(Mamdani)typefuzzylogiccontroller(FLC)havingbell-shaped input andsingletonoutputmembershipfunctions.Mixedmoderealizationofthecircuitmakesthedesignprogrammable and extendable,whilehavingrelativelylowpowerconsumption.Currentmoderealizationoftherulebaseanddefuzzifier circuits leadstosimpleandintuitiveconfigurations.Foraparticularsetofprogrammingparameters,simulationresultsofthe controller usingHSPICEsimulatorandlevel49parameters(BSIM3V3),showsanaveragepowerconsumptionof8mWand an RMSerrorof1.3%comparedtoidealresultsobtainedfromMATLAB
机译:此paperpresentsanewgeneralpurposeneuro-fuzzycontrollertorealizeadaptive-网络basedfuzzyinferencesys- TEM(ANFIS)architecture.Atwoinput,singleoutput,16rulesANFISarchitectureisdesignedin0.35m standardCMOS technology.Thiscontrollercanalsobeusedasastandard(马丹尼)typefuzzylogiccontroller(FLC)havingbell形输入andsingletonoutputmembershipfunctions.Mixedmoderealizationofthecircuitmakesthedesignprogrammable和可扩展的,whilehavingrelativelylowpowerconsumption.Currentmoderealizationoftherulebaseanddefuzzifier电路leadstosimpleandintuitiveconfigurations.Foraparticularsetofprogrammingparameters ,使用HSPICE模拟器和级别49参数(BSIM3V3)进行仿真的软件结果显示,平均功耗为8 mW,而RMS误差为MATLAB理想结果的1.3%

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号