公差分析和尺寸链方法.ppt

《公差分析和尺寸链方法.ppt》由会员上传分享，免费在线阅读，更多相关内容在教育资源-天天文库。

1、公差分析和车身尺寸链方法ToleranceAnalysis&Auto-bodyDimensionChainMethod车身质量控制系列讲座1-DToleranceanalysisAuto-bodyDimensionChainMethodContentsVariationvs.ToleranceVariationSimulationMethodsAdvantagesandLimitationsofEachMethodIntegrationofKeyCharacteristicofProduct&ProcessAssemblyConstraintsToleranc

2、eandSolvingofDimensionChainPartI1-DToleranceanalysis3CharacterizingtheperformanceofaprocessCentraltendencyormean:Spreadorvariation-4-3-2-101234-4-3-2-101234Deviation4Variationvs.ToleranceVariation:•iswhattheprocessgivesus,•maybequitedifferentfromthetolerance.ToleranceorSpecificati

3、onis•theallowablelevelofvariation,•basedonfunctionalconsideration,•usedtoestablishapart'sconformabilitytodesign.Butthetechniquesforpredictingvariationortoleranceforanassemblyisthesame.5VariationSimulation123x±ax±bx±c123y±t123Givenindividualpartdimensionsandtheirdistribution,whataret

4、heassemblydimensions?Butthemethodcanappliedmorewidelythanmechanicalassembly.Thegeneralformis:givenafunctionY=f(x1,x2,…),andthedistributionsofxi,Whatisthedistributionofy?6DistributionofToleranceWorstCaseStatistical:RootsumsquaresMonteCarloAssemblyModelExplicit:LinearizedSensitivityMe

5、chanisticModelNon-linearModelImplicit:VariationSimulationMethodsTwothingsareessentialinordertoperformvariationanalysis.Oneisanassemblymodelorinput-outputmodel.Theotheristhedistributionofvariables.7CommonlyUsedVariationSimulationMethodsWorstCase:(Conway,1948;ChaseandParkinson,1991)R

6、ootSumSquares(RSS):(Spotts,1978,LeeandWoo,1990)MonteCarloSimulation:(Craig,1989)123x±ax±bx±c123y±t1238SimpleVariationSimulationExampleGivencomponenttolerances,determinethevariationinthemeasureddimension(thegapbetweentheblocksandthebase).Inworstcaseanalysis,itisassumedthatthecontribu

7、tingdimensionsarealwayswithintolerance.Bymakingthisassumptionworstcaselimitscanbefoundwithinwhichthemeasureddimensionmustalwaysfall.WorstCaseAnalysisRootSumSquaresTheideabehindRSSistotreatatoleranceasanormaldistributionwithcertainprocesscapability,anduserandomassembly.Meanandvarianc

8、eofsomelinearfuncti