欢迎来到天天文库
浏览记录
ID:36800288
大小:5.76 MB
页数:53页
时间:2019-05-15
《基于神经网络的逆变器的研究》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、摘要逆变器在当今社会具有很广泛的应用,随着电力电子技术的快速发展,大量高功率开关器件相继出现,为逆变器在各行各业的应用提供了保障。本文研究的逆变器主电路采用三相桥式结构。在控制策略方面,传统的正弦脉宽调制(SPwM)具有直流电压利用率较低的缺点,空间矢量脉宽调制(SVPWM)是一种优化的脉宽调制技术,能明显减小逆变器输出电流的谐波成分并能提高电压的利用率,已有取代传统SPWM的趋势,本文即采用电压空间矢量脉宽调制作为控制策略。本文首先对电力电子技术及逆变器的相关知识作了简要的介绍。对空间电压矢量脉宽调制的工作原理做了深入的研究,并且在此基础上构建
2、了基于SVPWM算法的逆变器仿真模型,得到了仿真结果。之后本文提出了基于神经网络的逆变器的构想,在计算逆变器三相导通时间时,运用了3层前馈网络,并采用Levenberg.Marquardt算法对神经网络进行训练。最后,在MATLAB/Simulink环境下,对系统进行了仿真研究,仿真结果表明基于神经网络的SVPWM控制方法是可行的,能达到预期的效果,同时通过合理的运用神经网络,可以提高PWM的频率,减小输出电压中的谐波含量。关键词:逆变器空间矢量神经网络MATLAB/SimulinkAbstractTheinverterisusedwidelyi
3、nsociety.Withtherapiddevelopmentofpowerelectronicstechnology,alargeamountofhigh-frequencyswitchingequipmentareappearedwhichprovidetheguaranteefortheinverter’Sapplicationinallkindsoffields.Theinverter’Smaincircuitdiscussedinthispaperadoptsthethree-phasebridgestructure.Intermso
4、fcontrollerstrategy,thetraditionalmethodofSPWMhasthesetbackwhichislowerelectricsourceutilizationratio.SVPWMisakindofoptimizedmethod,itcanreducetheoutputcurrent’Sharmoniccomponentandimprovetheelectricsourceutilizationratio,havingthetrendofreplacingthetraditionalSPWMmethod,this
5、paperjustadoptstheSVPWMasthecontrollerstrategy.Thispaperfirstlyintroducesthepowerelectronicstechnologyandinverter’Scorrespondingknowledge,andthepaperdeeplyresearchesthework’SprincipleoftheSVPWMandestablishestheinverter’SsimulationmodelbasedonthearithmeticofSVPWM,achievedthesi
6、mulationresult.Andthenthispaperputsforwardtheinverter’Simaginationbasedonneuralnetwork,whencalculatingthethree-phasetimes,thepaperusesathree—layerforward—feednetwork,adoptesthearithmeticofLevenberg—Marquardetotrainthenetwork.Atlast,intheenvironmentofMATLAB/‘Simulink,thePaperm
7、akesthesimulationresearchtothesystem,thesimulationresultsshowsthatthemethodofSVPWMbasedonneuralnetworkisfeasibleanditcanachievetheanticipatedeffect,atthesametimebyusingtheneuralnetworkrationally,themethodcanenhancethePWMfrequencyandreducetheharmonicofoutputvoltage.Keywords:In
8、verterSpaceVectorNeuralNetworkMATLAB/Simulink独创性声明本人声明所呈交的学位论文是本人在导师
此文档下载收益归作者所有