資源描述:
《基于現(xiàn)場(chǎng)數(shù)據(jù)的火電機(jī)組模型參數(shù)辨識(shí)》由會(huì)員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫。
1、華北電力大學(xué)碩士學(xué)位論文摘要摘要本文研究基于現(xiàn)場(chǎng)數(shù)據(jù)的500Mw火電機(jī)組模型參數(shù)辨識(shí),分析討論控制系統(tǒng)閉環(huán)可辨識(shí)性,給出幾種可辨識(shí)條件;選取機(jī)組動(dòng)態(tài)變化過程中數(shù)據(jù),滿足輸入信號(hào)的持續(xù)激勵(lì)條件;針對(duì)電廠運(yùn)行數(shù)據(jù)特點(diǎn),采用野值識(shí)別、剔除與補(bǔ)正,噪聲去除,數(shù)據(jù)平滑等方法做數(shù)據(jù)預(yù)處理;分析制粉過程、鍋爐蓄熱、過熱器差壓、汽輪機(jī)動(dòng)態(tài)的特點(diǎn),建立500MW火電機(jī)組簡化模型,分析模型中參數(shù)可辨識(shí)性,給出參數(shù)辨識(shí)方案;以汽輪機(jī)動(dòng)態(tài)過程模型為辨識(shí)對(duì)象,利用Hopfield神經(jīng)網(wǎng)絡(luò)優(yōu)化算法進(jìn)行參數(shù)辨識(shí);檢驗(yàn)所得模型,結(jié)果表明模型的適應(yīng)度較高,辨識(shí)結(jié)果較好。關(guān)鍵詞:參數(shù)辨識(shí),Hopfi
2、eld神經(jīng)網(wǎng)絡(luò),數(shù)據(jù)預(yù)處理,火電機(jī)組模型ABSTRACTThethesisstudiedtheparameterldentiflcationof。500MWthermalpowerunitmodelbasedontheon—linedata.Afteranalyzinganddiscussingtheidentifyingclosed-loopofcontrolsystem,seVeralidentinableconditionsweregiVen;Thethesisselecteddatafromdynamicprocessofunittomeetthecont
3、inuedinspiritconditionsofinputsignals;InViewofthecharacteristicoftherunningdataofpowerplant,a10tofworkwasdoneondataprocessing,suchaseliminatingandcorrectingtheunreasonabledata,remoVingthenoisefromthecorrupteddatawithwaVelettransform,datasmoothing;AfteranaIyzingthemillprocess,boilerhea
4、tstorageprocess,super—heaterpressurediffbrentialprocess,lurbinedynamic,itestablishedsimplmedmodelof500Mwthermalpowerunitandanalyzedidentifyingparameterofmodel,thenthemethodofparameteridenti矗cationisgiVen;Thesteamturbinedynamicmodelwastakenasidentificationobject.ItisidentifiedbyusingHN
5、Noptimizearithmetic.Throughtestingtheobtainedmodel,theresultsshowthatthemodelhashighfitnessandtheidentificationresultisgood.Y.ouYbnghua(Contr01TheoryandControlEngineering)DirectedbyProf.SunJianpingKEYWoRDS:parameteridentification,Hopfieldneuralnetwork,datapreprocessing,thermalpoweruni
6、tmodel華北電力大學(xué)碩士學(xué)位論文摘要摘要本文研究基于現(xiàn)場(chǎng)數(shù)據(jù)的500Mw火電機(jī)組模型參數(shù)辨識(shí),分析討論控制系統(tǒng)閉環(huán)可辨識(shí)性,給出幾種可辨識(shí)條件;選取機(jī)組動(dòng)態(tài)變化過程中數(shù)據(jù),滿足輸入信號(hào)的持續(xù)激勵(lì)條件;針對(duì)電廠運(yùn)行數(shù)據(jù)特點(diǎn),采用野值識(shí)別、剔除與補(bǔ)正,噪聲去除,數(shù)據(jù)平滑等方法做數(shù)據(jù)預(yù)處理;分析制粉過程、鍋爐蓄熱、過熱器差壓、汽輪機(jī)動(dòng)態(tài)的特點(diǎn),建立500MW火電機(jī)組簡化模型,分析模型中參數(shù)可辨識(shí)性,給出參數(shù)辨識(shí)方案;以汽輪機(jī)動(dòng)態(tài)過程模型為辨識(shí)對(duì)象,利用Hopfield神經(jīng)網(wǎng)絡(luò)優(yōu)化算法進(jìn)行參數(shù)辨識(shí);檢驗(yàn)所得模型,結(jié)果表明模型的適應(yīng)度較高,辨識(shí)結(jié)果較好。關(guān)鍵詞:參數(shù)辨識(shí)
7、,Hopfield神經(jīng)網(wǎng)絡(luò),數(shù)據(jù)預(yù)處理,火電機(jī)組模型ABSTRACTThetheslsstudiedtheparameterldentiflcationof。500MWthermalpowerunitmodelbasedontheon—linedata.A11eranalyzinganddiscussingtheidentifyingclosed-loopofcontrolsystem,seVeralidentinableconditionsweregiVen;Thethesisselecteddatafromdynamicprocessofunittomeett
8、hecon