資源描述：
《基于模糊神經(jīng)網(wǎng)絡(luò)的智能優(yōu)化pid控制器研究》由會(huì)員上傳分享，免費(fèi)在線(xiàn)閱讀，更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫(kù)。
1、摘要基于模糊神經(jīng)網(wǎng)絡(luò)的智能優(yōu)化PID控制器研究PID控制算法簡(jiǎn)單、魯棒性好且規(guī)則容易理解，被廣泛應(yīng)用于工業(yè)過(guò)程控制中。但隨著工業(yè)生產(chǎn)的發(fā)展，對(duì)象越來(lái)越復(fù)雜，傳統(tǒng)PID控制的弊端日益暴露。模糊控制和神經(jīng)網(wǎng)絡(luò)均不依賴(lài)被控對(duì)象的數(shù)學(xué)模型，且具有較強(qiáng)的自適應(yīng)和自學(xué)習(xí)能力。針對(duì)復(fù)雜非線(xiàn)性對(duì)象，利用神經(jīng)網(wǎng)絡(luò)與模糊控制的優(yōu)點(diǎn)，設(shè)計(jì)PID控制器的在線(xiàn)調(diào)整控制系統(tǒng)，改善系統(tǒng)性能，無(wú)論在理論還是實(shí)踐上都將具有重要意義。本文在吸取傳統(tǒng)的經(jīng)典控制理論強(qiáng)大的分析能力基礎(chǔ)上，結(jié)合神經(jīng)網(wǎng)絡(luò)和模糊控制的特點(diǎn)，將模糊神經(jīng)網(wǎng)絡(luò)控制與傳統(tǒng)PID控制相結(jié)合，明確了模糊神經(jīng)網(wǎng)絡(luò)對(duì)于解決傳統(tǒng)過(guò)程控制問(wèn)題的重要地位。設(shè)計(jì)了
2、一種基于模糊神經(jīng)網(wǎng)絡(luò)的智能優(yōu)化PID控制器，采用遺傳算法優(yōu)化模糊神經(jīng)網(wǎng)絡(luò)的參數(shù)，包括隸屬度函數(shù)均值co、標(biāo)準(zhǔn)差bo及網(wǎng)絡(luò)輸出層所對(duì)應(yīng)的連接權(quán)系數(shù)Wii，在線(xiàn)更新PID參數(shù)。文中給出了基于模糊神經(jīng)網(wǎng)絡(luò)智能優(yōu)化PID控制系統(tǒng)的結(jié)構(gòu)及控制器的一般算法，并討論了利用VC語(yǔ)言實(shí)現(xiàn)控制器軟件化的方法。同時(shí)對(duì)TE過(guò)程，利用該模糊神經(jīng)網(wǎng)絡(luò)智能優(yōu)化PID控制器對(duì)其進(jìn)行了仿真控制研究，仿真結(jié)果表明基于模糊神經(jīng)網(wǎng)絡(luò)的智能優(yōu)化PID控制器具有良好的控制效果，有著廣闊的發(fā)展前景。北京化T大學(xué)碩：卜學(xué)位論文關(guān)鍵詞：PID控制，模糊神經(jīng)網(wǎng)絡(luò)，遺傳算法，智能控制ABSTRACTINTELLIGENTOPTIM
3、IZATIoNPIDCONTROLLERBASEDoNFUZZYNEURALNETWORKSThePIDcontrolalgorithmissimpleinstructure，stronginrobustness，andcanbeunderstoodeasily．Ithasbeenwidelyusedinindustrialprocesscontr01．However,theobjectsbecomemoreandmorecomplexwiththedevelopmentofindustrialproems，andthedrawbacksofthetraditionalPIDc
4、ontrolareincreasinglyexposed．TheFuzzyControlandNeuralNetworksdonotrelayonthemathematicmodelofcontrolledobjects，andhavestronggerabilityofself-adaptationandself-study．Aimingatthecomplexnon-linerobjects，itissignificanttoemploytheexcellenceofNeuralNetworksandFuzzyControltodesigntheon-lineadjusti
5、ngcontrolsystemofPIDcontroltoimprovethesystemperformancebomintheoryandpractice．ThisthesiscombinedthecharacteristicofNeuralNetworksandFuzzyControlonthebasisofpowerfulanalysisabilityoftraditionalclassiccontroltheory．FirstlyitcombinedtheFuzzyNeuralNetworksControlandtraditionalPIDcontrol，determi
6、nedtheimportantstateofFuzzyNeuralNetworksresolvingthetraditionalprocesscontrol，北京化工人學(xué)碩J：學(xué)位論文designedanintelligentoptimizationPIDcontrollerbasedonFuzzyNeuralNetworkswhichusedGeneticAlgorithmtooptimizetheFuzzyNeuralNetworksparametersincludingthemembershipfunctionmean，standarddevimion，connectio
7、nweightswijcorrespondingtothenetworkoutputlayer,andon-lineupdatedPIDparameters．ThenifgavethestructureofintelligentoptimizationPIDcontrolsystemandthegeneralalgorithmofcontrollerbasedontheFuzzyNeuralNetworks，anddiscussedthemethodofusingVCtorealizethe
