欢迎来到天天文库
浏览记录
ID:36374889
大小:5.48 MB
页数:112页
时间:2019-05-10
《自由曲面数字化建模与自组织拟合的研究》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、华中科技大学博士学位论文自由曲面数字化建模与自组织拟合的研究姓名:邓春梅申请学位级别:博士专业:机械制造及其自动化指导教师:师汉民;陈吉红2001.3.1华中科技大学博士学位论文ABSTRACTDigitalmodelingforfreefornlsurfacesisallimportantproblemofreverseengineering.Itiswidelyusedinthefieldsofaviatiomshipbuilding,automobileandmouldindustry.Themodelofasurfacewilldeterminetheparts’
2、shapedesigningandcuttingqualitydirectly,thusaffecttheproducts’functionandfaimcss.Focusingontheproblemsofthetraditionalmodelsbyconstructingalgebraequationsonsub·patches,anoveldigitalmodelwhichrepresentsurfacesbyasetofmeshpointsprocessedbyself-organizationwaysanditsrelativethcodeswilIbeintr
3、oducedinthisPa廿er.ThechiefcontributionofthisPaDeristhetheoryofsurfacerepresentationmannerbymeshpointscalledself-organizationmodeling.Aself-organizationfittingmethodsimulatingthebehaviorofeellularalltomata,showingasself-adjusting.shoftdistancecommunicating,paralleloperatinganditerationopti
4、mizinghasalsobeenputforwardinthispaper.Asadiscretetopologyhasbeenadoptedinthisnewmodel,nocompositealgorithmsbetweensub-patcheswhichhadtobecarriedoutintraditionalmodelshouldbeconsidered.Asthemeshpointssequenceshasbeenemployedinthisnewroodelinsteadofalgebraequations,noparametricdomainsarene
5、ededwhicharealsonecessaryinlxaditionalmodel.Onlysomesimplecalculationsaretakenwithnoconsiderationoftheshapeof也emeasuredsurface.Inaddition.themodelhasstrongimmunitytonoisefromthemeastLrepoints.whichleadstothemodelisespeciallysuitableforthesituat.ions谢t11badenvironments.Toeliminate血einforma
6、tionof*-hesurfacemodel-thusimprovingthemodelingspeedandenhancingprecision,allalgorithmbasedondigitalimageprocessingforgeneratingself-adaptivedensitymeshfrommeasurcddensescattedpointsisalsointrodueed.111eautomaticselectionofthemeasuringpointsbecomesrealitybythisway.Astheunifon-nmeshgridsha
7、vethecharactersofdigitalimages.imageprocessingalgorithmscanbeutilizeddirectlybymappingagraphtoagrayscaleimage,whichmadethecalculationforadaptivemeshveryeasy.Contrastwithtraditionalmeshgenerators.it’swithmoreconvcniel'iceforrealization.Duetotheabilityofparalleloperat
此文档下载收益归作者所有