BP神经网络在大型斜拉桥施工控制中的应用研究

《BP神经网络在大型斜拉桥施工控制中的应用研究》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、河海大学硕士学位论文BP神经网络在大型斜拉桥施工控制中的应用研究姓名：于涛申请学位级别：硕士专业：大地测量学与测量工程指导教师：黄张裕;赵仲荣20070401AbstractWitllthefastdevelopmentofthetransportationbusinessandtheunremittinglytechniqueofbridgeconstruction，bridgeconstructiontoacrossthedegree,highlydifficultdirectiondevel

2、opmentgreatly,structuresupple,beautyandsafetyofcable-stayedbridgegraduallydriveextensiveadoption．ConstructioncontrolisanimportantpartofcaMe-stayedbridgeconstructiontechnique；itisakeytomakeSUretheconstructionqualityofbridge．Soitisanimportantworktocable

3、-stayedbridge’Sconstructioncontr01．Factorsthataffectconstructioncontrolmanyandcomplex，andtheseinformationindependenteachother．Howwillvalidexploitationistheseinformation,combinemodernmathematicstheories，builduprelatedsupervisionmodelandealTyonavalidcon

4、structioncontrol，isthemaincontentsofthistextresearch．Thefulltextcombinedwimthecharacteristicsoflargecable-stayedbridgeconstruction，researchtheartificialneuralnetworkusedtoconstructioncontrol，worksmainlydidasfollows：(1)wimtheanalyticalanddiscussionofth

5、ebasicprincipletotheartificialneuralnetwork，fromtheories，thearticleexpoundsandprovedthepossibilityoftheneuralnetworkmethodusedtoconstructioncontr01．Inthisfoundationtop，thefactorsthatinfluencecable—stayedbridge’sconstructioncontrolanditsinteractionrela

6、tionareresearched．(2)AtthefoundationofanalyticaloftheblemishmatBPstudycalculateswayexistentinphysicallyapplied，giveahomologousimprovementmethod．CombineananalyticalresulttoestablishamodelthatinkeepingwimtheBPartificialneuralnetwork，topredicttheformedge

7、aringlinecontrolofthesteelboxofcabled-bridge．(3)Aimatagreaterproblemoftheconstructiondataunitandthequantityclassdifference，theadoptionamethodofdatareturnonturning，andovercametheclassbaddatawhichinfluenceoftheastringencydisadvantagetothenetworkfromthes

8、elf-study．(4)Combinetheexampletheell百neeringofthethirdNanjingYangtzeRiverBridge，drawledupthelineformcontrolpredictionprocedurewiththeMATLABprocedurelanguage，obtainedagoodresult．(5)Comparethegraytheoriespredictionandtheneuralnetworkmethodpredic