資源描述:
《鍋爐運(yùn)行燃燒優(yōu)化模型-研究與其驗證》由會員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在應(yīng)用文檔-天天文庫。
1、上的運(yùn)行最優(yōu)區(qū)間,構(gòu)建出基于規(guī)則庫的燃燒優(yōu)化模型,并且在運(yùn)行中不斷完善這個模型,以保證其實時性。通過建立的燃燒優(yōu)化模型,控制相應(yīng)調(diào)節(jié)量來實現(xiàn)電廠的燃燒優(yōu)化,實現(xiàn)方法是找到相應(yīng)的控制方案使運(yùn)行狀況迅速向最近的優(yōu)化區(qū)間靠近,通過DCS完成閉環(huán)控制,確保鍋爐盡量處在優(yōu)化區(qū)間內(nèi)運(yùn)行,以求達(dá)到歷史因素下的最優(yōu),完成電廠的優(yōu)化目標(biāo)。利用支持向量機(jī)在小樣本學(xué)習(xí)上的優(yōu)越性,建立相應(yīng)的支持向量機(jī)燃燒優(yōu)化模型,用遺傳算法實現(xiàn)模型參數(shù)的尋優(yōu),通過建立的模型預(yù)測鍋爐的運(yùn)行狀況,本文實驗證明支持向量機(jī)建立的模型具有很好的預(yù)測能力,并能找到客觀條件下的最優(yōu)?;诰垲惖臄?shù)據(jù)挖掘是對電站海量歷史數(shù)
2、據(jù)的直接處理,可能會出現(xiàn)某些規(guī)則庫的操作量在電廠實際運(yùn)行中難以實際操作,而以支持向量機(jī)和遺傳算法的燃燒優(yōu)化建模的預(yù)測功能對其則是一個補(bǔ)充,兩者相輔相成。通過VisualC++編程實現(xiàn)燃燒優(yōu)化軟件的開發(fā),并在大同二電7≠}鍋爐投入使用,實現(xiàn)了多目標(biāo)優(yōu)化,提高了鍋爐效率和降低了NOx排放量。關(guān)鍵詞:燃燒優(yōu)化;數(shù)據(jù)挖掘;支持向量機(jī);遺傳算法;華北電力大學(xué)碩士學(xué)位論文AbstractHowthecoal-firedboilercantoachievehighboilerefficiencyandreducedNOxemissionsasthemostconcernedabo
3、utforthepowerPlants.ThroughcombustionoptnnlzatloncanbeverygoodtoimproveboilerefficiencyandreduceNOxemissions,SOthecontentsofthispapm‘hasimportantpracticalsignificance.FortheanalysisofDaTongNo.2PowerPlant撐7beforeandaftertheburnunitofhedgewallaroundtheboilercombustionunitoperatingconditi
4、onstOestablishCombnstionoptimizationmodel.Throughtheanalysisofplantoperatingdata,hathedividedconditionfixedloadrange,(similartothecoalandtheoutsideanabienttemperature)inthecase,pulverizedcoalconcenn'ationandflowrate,thesecondaryairflowandwindburn,lesswarmwaterfloWandgasblockopeningthec
5、ontroIpanelcombustioniSthemaininfluencefactors.Itcanbeadjustedbycontrollingtheappropriateamounttoachieveoptimizedcombustionpowerplant.Plant—levelmonitoringsystemallowsplantstosavevastamountsofhistoricaldata,cleaningmethodbyclusteringtoanalyzethehistoricaldata,tofindtheoptimalrangeofope
6、ratinghistory,buildarule—basedcombustionoptimizationmodeIlibraryandruncontinuouslyimprovethismodeltoei-istirereal—timenature.Bythecorrespondingcontr01programtorunquicklytothenearestopthnalconditionscloserange,completecbsed—loopcontrolbyDCStoensurethatasmuchaspossibleintheoptimizationof
7、itsoperatingrange,thuscompletingthepowerplantoptimizationobjectives.Usingsupportvectormachineinthesmallsamplethesuperiorityofthestudyandsetupacorrespondingsupportvectormachineburningoptimizatbnmodel.withgeneticalgorithmtorealizeoptimizationoftheparametersforthemodel,throughthemodelpr