機(jī)械加工刀具中英文對照外文翻譯文獻(xiàn)

中英文對照外文翻譯文獻(xiàn)PAGE17中英文對照外文翻譯英文原文SelectionofoptimumtoolgeometryandcuttingconditionsusingasurfaceroughnesspredictionmodelforendmillingAbstractInfluenceoftoolgeometryonthequalityofsurfaceproducediswellknownandhenceanyattempttoassesstheperformanceofendmillingshouldincludethetoolgeometry.Inthepresentwork,experimentalstudieshavebeenconductedtoseetheeffectoftoolgeometry(radialrakeangleandnoseradius)andcuttingconditions(cuttingspeedandfeedrate)onthemachiningperformanceduringendmillingofmediumcarbonsteel.Thefirstandsecondordermathematicalmodels,intermsofmachiningparameters,weredevelopedforsurfaceroughnesspredictionusingresponsesurfacemethodology(RSM)onthebasisofexperimentalresults.ThemodelselectedforoptimizationhasbeenvalidatedwiththeChisquaretest.Thesignificanceoftheseparametersonsurfaceroughnesshasbeenestablishedwithanalysisofvariance.Anattempthasalsobeenmadetooptimizethesurfaceroughnesspredictionmodelusinggeneticalgorithms(GA).TheGAprogramgivesminimumvaluesofsurfaceroughnessandtheirrespectiveoptimalconditions.1IntroductionEndmillingisoneofthemostcommonlyusedmetalremovaloperationsinindustrybecauseofitsabilitytoremovematerialfastergivingreasonablygoodsurfacequality.Itisusedinavarietyofmanufacturingindustriesincludingaerospaceandautomotivesectors,wherequalityisanimportantfactorintheproductionofslots,pockets,precisionmouldsanddies.Greaterattentionisgiventodimensionalaccuracyandsurfaceroughnessofproductsbytheindustrythesedays.Moreover,surfacefinishinfluencesmechanicalpropertiessuchasfatiguebehaviour,wear,corrosion,lubricationandelectricalconductivity.Thus,measuringandcharacterizingsurfacefinishcanbeconsideredforpredictingmachiningperformance.Surfacefinishresultingfromturningoperationshastraditionallyreceivedconsiderableresearchattention,whereasthatofmachiningprocessesusingmultipointcutters,requiresattentionbyresearchers.Astheseprocessesinvolvelargenumberofparameters,itwouldbedifficulttocorrelatesurfacefinishwithotherparametersjustbyconductingexperiments.Modellinghelpstounderstandthiskindofprocessbetter.Thoughsomeamountofworkhasbeencarriedouttodevelopsurfacefinishpredictionmodelsinthepast,theeffectoftoolgeometryhasreceivedlittleattention.However,theradialrakeanglehasamajoraffectonthepowerconsumptionapartfromtangentialandradialforces.Italsoinfluenceschipcurlingandmodifieschipflowdirection.Inadditiontothis,researchers[1]havealsoobservedthatthenoseradiusplaysasignificantroleinaffectingthesurfacefinish.Thereforethedevelopmentofagoodmodelshouldinvolvetheradialrakeangleandnoseradiusalongwithotherrelevantfactors.Establishmentofefficientmachiningparametershasbeenaproblemthathasconfrontedmanufacturingindustriesfornearlyacentury,andisstillthesubjectofmanystudies.Obtainingoptimummachiningparametersisofgreatconcerninmanufacturingindustries,wheretheeconomyofmachiningoperationplaysakeyroleinthecompetitivemarket.Inmaterialremovalprocesses,animproperselectionofcuttingconditionscausesurfaceswithhighroughnessanddimensionalerrors,anditisevenpossiblethatdynamicphenomenaduetoautoexcitedvibrationsmaysetin[2].Inviewofthesignificantrolethatthemillingoperationplaysintoday’smanufacturingworld,thereisaneedtooptimizethemachiningparametersforthisoperation.So,anefforthasbeenmadeinthispapertoseetheinfluenceoftoolgeometry(radialrakeangleandnoseradius)andcuttingconditions(cuttingspeedandfeedrate)onthesurfacefinishproducedduringendmillingofmediumcarbonsteel.Theexperimentalresultsofthisworkwillbeusedtorelatecuttingspeed,feedrate,radialrakeangleandnoseradiuswiththemachiningresponsei.e.surfaceroughnessbymodelling.Themathematicalmodelsthusdevelopedarefurtherutilizedtofindtheoptimumprocessparametersusinggeneticalgorithms.2ReviewProcessmodellingandoptimizationaretwoimportantissuesinmanufacturing.Themanufacturingprocessesarecharacterizedbyamultiplicityofdynamicallyinteractingprocessvariables.Surfacefinishhasbeenanimportantfactorofmachininginpredictingperformanceofanymachiningoperation.Inordertodevelopandoptimizeasurfaceroughnessmodel,itisessentialtounderstandthecurrentstatusofworkinthisarea.Davisetal.[3]haveinvestigatedthecuttingperformanceoffiveendmillshavingvarioushelixangles.CuttingtestswereperformedonaluminiumalloyL65forthreemillingprocesses(face,slotandside),inwhichcuttingforce,surfaceroughnessandconcavityofamachinedplanesurfaceweremeasured.Thecentralcompositedesignwasusedtodecideonthenumberofexperimentstobeconducted.Thecuttingperformanceoftheendmillswasassessedusingvarianceanalysis.Theaffectsofspindlespeed,depthofcutandfeedrateonthecuttingforceandsurfaceroughnesswerestudied.Theinvestigationshowedthatendmillswithlefthandhelixanglesaregenerallylesscosteffectivethanthosewithrighthandhelixangles.Thereisnosignificantdifferencebetweenupmillinganddownmillingwithregardtothecuttingforce,althoughthedifferencebetweenthemregardingthesurfaceroughnesswaslarge.Bayoumietal.[4]havestudiedtheaffectofthetoolrotationangle,feedrateandcuttingspeedonthemechanisticprocessparameters(pressure,frictionparameter)forendmillingoperationwiththreecommerciallyavailableworkpiecematerials,11L17freemachiningsteel,62-35-3freemachiningbrassand2024aluminiumusingasingleflutedHSSmillingcutter.Ithasbeenfoundthatpressureandfrictionactonthechip–toolinterfacedecreasewiththeincreaseoffeedrateandwiththedecreaseoftheflowangle,whilethecuttingspeedhasanegligibleeffectonsomeofthematerialdependentparameters.Processparametersaresummarizedintoempiricalequationsasfunctionsoffeedrateandtoolrotationangleforeachworkmaterial.However,researchershavenottakenintoaccounttheeffectsofcuttingconditionsandtoolgeometrysimultaneously;besidesthesestudieshavenotconsideredtheoptimizationofthecuttingprocess.Asendmillingisaprocesswhichinvolvesalargenumberfparameters,combinedinfluenceofthesignificantparametersanonlybeobtainedbymodelling.MansourandAbdallaetal.[5]havedevelopedasurfaceroughnessmodelfortheendmillingofEN32M(asemi-freecuttingcarboncasehardeningsteelwithimprovedmerchantability).Themathematicalmodelhasbeendevelopedintermsofcuttingspeed,feedrateandaxialdepthofcut.Theaffectoftheseparametersonthesurfaceroughnesshasbeencarriedoutusingresponsesurfacemethodology(RSM).Afirstorderequationcoveringthespeedrangeof30–35m/minandasecondorderequationcoveringthespeedrangeof24–38m/minweredevelopedunderdrymachiningconditions.Alauddinetal.[6]developedasurfaceroughnessmodelusingRSMfortheendmillingof190BHNsteel.Firstandsecondordermodelswereconstructedalongwithcontourgraphsfortheselectionofthepropercombinationofcuttingspeedandfeedtoincreasethemetalremovalratewithoutsacrificingsurfacequality.Hasmietal.[7]alsousedtheRSMmodelforassessingtheinfluenceoftheworkpiecematerialonthesurfaceroughnessofthemachinedsurfaces.Themodelwasdevelopedformillingoperationbyconductingexperimentsonsteelspecimens.Theexpressionshows,therelationshipbetweenthesurfaceroughnessandthevariousparameters;namely,thecuttingspeed,feedanddepthofcut.Theabovemodelshavenotconsideredtheaffectoftoolgeometryonsurfaceroughness.Sincetheturnofthecenturyquitealargenumberofattemptshavebeenmadetofindoptimumvaluesofmachiningparameters.Usesofmanymethodshavebeenreportedintheliteraturetosolveoptimizationproblemsformachiningparameters.JainandJain[8]haveusedneuralnetworksformodelingandoptimizingthemachiningconditions.Theresultshavebeenvalidatedbycomparingtheoptimizedmachiningconditionsobtainedusinggeneticalgorithms.Sureshetal.[9]havedevelopedasurfaceroughnesspredictionmodelforturningmildsteelusingaresponsesurfacemethodologytoproducethefactoraffectsoftheindividualprocessparameters.Theyhavealsooptimizedtheturningprocessusingthesurfaceroughnesspredictionmodelastheobjectivefunction.Consideringtheabove,anattempthasbeenmadeinthisworktodevelopasurfaceroughnessmodelwithtoolgeometryandcuttingconditionsonthebasisofexperimentalresultsandthenoptimizeitfortheselectionoftheseparameterswithinthegivenconstraintsintheendmillingoperation.3MethodologyInthiswork,mathematicalmodelshavebeendevelopedusingexperimentalresultswiththehelpofresponsesurfacemethodology.Thepurposeofdevelopingmathematicalmodelsrelatingthemachiningresponsesandtheirfactorsistofacilitatetheoptimizationofthemachiningprocess.Thismathematicalmodelhasbeenusedasanobjectivefunctionandtheoptimizationwascarriedoutwiththehelpofgeneticalgorithms.3.1MathematicalformulationResponsesurfacemethodology(RSM)isacombinationofmathematicalandstatisticaltechniquesusefulformodellingandanalyzingtheproblemsinwhichseveralindependentvariablesinfluenceadependentvariableorresponse.Themathematicalmodelscommonlyusedarerepresentedby:whereYisthemachiningresponse,?istheresponsefunctionandS,f,α,raremillingvariablesand∈istheerrorwhichisnormallydistributedabouttheobservedresponseYwithzeromean.Therelationshipbetweensurfaceroughnessandotherindependentvariablescanberepresentedasfollows，whereCisaconstantanda,b,canddareexponents.Tofacilitatethedeterminationofconstantsandexponents,thismathematicalmodelwillhavetobelinearizedbyperformingalogarithmictransformationasfollows:TheconstantsandexponentsC,a,b,canddcanbedeterminedbythemethodofleastsquares.Thefirstorderlinearmodel,developedfromtheabovefunctionalrelationshipusingleastsquaresmethod,canberepresentedasfollows:whereY1istheestimatedresponsebasedonthefirst-orderequation,Yisthemeasuredsurfaceroughnessonalogarithmicscale,x0=1(dummyvariable),x1,x2,x3andx4arelogarithmictransformationsofcuttingspeed,feedrate,radialrakeangleandnoseradiusrespectively,∈istheexperimentalerrorandbvaluesaretheestimatesofcorrespondingparameters.Thegeneralsecondorderpolynomialresponseisasgivenbelow:whereY2istheestimatedresponsebasedonthesecondorderequation.Theparameters,i.e.b0,b1,b2,b3,b4,b12,b23,b14,etc.aretobeestimatedbythemethodofleastsquares.Validityoftheselectedmodelusedforoptimizingtheprocessparametershasbeentestedwiththehelpofstatisticaltests,suchasF-test,chisquaretest,etc.[10].3.2OptimizationusinggeneticalgorithmsMostoftheresearchershaveusedtraditionaloptimizationtechniquesforsolvingmachiningproblems.Thetraditionalmethodsofoptimizationandsearchdonotfarewelloverabroadspectrumofproblemdomains.Traditionaltechniquesarenotefficientwhenthepracticalsearchspaceistoolarge.Thesealgorithmsarenotrobust.Theyareinclinedtoobtainalocaloptimalsolution.Numerousconstraintsandnumberofpassesmakethemachiningoptimizationproblemmorecomplicated.So,itwasdecidedtoemploygeneticalgorithmsasanoptimizationtechnique.GAcomeundertheclassofnon-traditionalsearchandoptimizationtechniques.GAaredifferentfromtraditionaloptimizationtechniquesinthefollowingways:1.GAworkwithacodingoftheparameterset,nottheparameterthemselves.2.GAsearchfromapopulationofpointsandnotasinglepoint.3.GAuseinformationoffitnessfunction,notderivativesorotherauxiliaryknowledge.4.GAuseprobabilistictransitionrulesnotdeterministicrules.5.ItisverylikelythattheexpectedGAsolutionwillbetheglobalsolution.Geneticalgorithms(GA)formaclassofadaptiveheuristicsbasedonprinciplesderivedfromthedynamicsofnaturalpopulationgenetics.Thesearchingprocesssimulatesthenaturalevaluationofbiologicalcreaturesandturnsouttobeanintelligentexploitationofarandomsearch.ThemechanicsofaGAissimple,involvingcopyingofbinarystrings.Simplicityofoperationandcomputationalefficiencyarethetwomainattractionsofthegeneticalgorithmicapproach.Thecomputationsarecarriedoutinthreestagestogetaresultinonegenerationoriteration.Thethreestagesarereproduction,crossoverandmutation.InordertouseGAtosolveanyproblem,thevariableistypicallyencodedintoastring(binarycoding)orchromosomestructurewhichrepresentsapossiblesolutiontothegivenproblem.GAbeginwithapopulationofstrings(individuals)createdatrandom.Thefitnessofeachindividualstringisevaluatedwithrespecttothegivenobjectivefunction.Thenthisinitialpopulationisoperatedonbythreemainoperators–reproductioncrossoverandmutation–tocreate,hopefully,abetterpopulation.Highlyfitindividualsorsolutionsaregiventheopportunitytoreproducebyexchangingpiecesoftheirgeneticinformation,inthecrossoverprocedure,withotherhighlyfitindividuals.Thisproducesnew“offspring”solutions,whichsharesomecharacteristicstakenfromboththeparents.Mutationisoftenappliedaftercrossoverbyalteringsomegenes(i.e.bits)intheoffspring.Theoffspringcaneitherreplacethewholepopulation(generationalapproach)orreplacelessfitindividuals(steadystateapproach).Thisnewpopulationisfurtherevaluatedandtestedforsometerminationcriteria.Thereproduction-crossovermutation-evaluationcycleisrepeateduntiltheterminationcriteriaaremet.中文翻譯選擇最佳工具，幾何形狀和切削條件

利用表面粗糙度預(yù)測模型端銑摘要：刀具幾何形狀對工件表面質(zhì)量產(chǎn)生的影響是人所共知的，因此，任何成型面端銑設(shè)計(jì)應(yīng)包括刀具的幾何形狀。在當(dāng)前的工作中，實(shí)驗(yàn)性研究的進(jìn)行已看到刀具幾何（徑向前角和刀尖半徑）和切削條件（切削速度和進(jìn)給速度），對加工性能，和端銑中碳鋼影響效果。第一次和第二次為建立數(shù)學(xué)模型，從加工參數(shù)方面，制訂了表面粗糙度預(yù)測響應(yīng)面方法（丹參），在此基礎(chǔ)上的實(shí)驗(yàn)結(jié)果。該模型取得的優(yōu)化效果已得到證實(shí)，并通過了卡方檢驗(yàn)。這些參數(shù)對表面粗糙度的建立，方差分析極具意義。通過嘗試也取得了優(yōu)化表面粗糙度預(yù)測模型，采用遺傳算法（GA）。在加文的程式中實(shí)現(xiàn)了最低值，表面粗糙度及各自的值都達(dá)到了最佳條件。

1導(dǎo)言端銑是最常用的金屬去除作業(yè)方式，因?yàn)樗軌蚋焖偃コ镔|(zhì)并達(dá)到合理良好的表面質(zhì)量。它可用于各種各樣的制造工業(yè)，包括航空航天和汽車這些以質(zhì)量為首要因素的行業(yè)，以及在生產(chǎn)階段，槽孔，精密模具和模具這些更加注重尺寸精度和表面粗糙度產(chǎn)品的行業(yè)內(nèi)。此外，表面光潔度還影響到機(jī)械性能，如疲勞性能，磨損，腐蝕，潤滑和導(dǎo)電性。因此，測量表面光潔度，可預(yù)測加工性能。車削過程對表面光潔度造成的影響歷來倍受研究關(guān)注，對于加工過程采用多刀，用機(jī)器制造處理，都是研究員需要注意的。由于這些過程涉及大量的參數(shù)，使得難以將關(guān)聯(lián)表面光潔度與其他參數(shù)進(jìn)行實(shí)驗(yàn)。在這個(gè)過程中建模有助于更好的理解。在過去，雖然通過許多人的大量工作，已開發(fā)并建立了表面光潔度預(yù)測模型，但影響刀具幾何方面受到很少注意。然而，除了切向和徑向力量，徑向前角對電力的消費(fèi)有著重大的影響。它也影響著芯片冰壺和修改芯片方向人流。此外，研究人員[1]也指出，在不影響表面光潔度情況下，刀尖半徑發(fā)揮著重要作用。因此，發(fā)展一個(gè)很好的模式應(yīng)當(dāng)包含徑向前角和刀尖半徑連同其他相關(guān)因素。對于制造業(yè)，建立高效率的加工參數(shù)幾乎是將近一個(gè)世紀(jì)的問題，并且仍然是許多研究的主題。獲得最佳切削參數(shù)，是在制造業(yè)是非常關(guān)心的，而經(jīng)濟(jì)的加工操作中及競爭激烈的市場中發(fā)揮了關(guān)鍵作用。在材料去除過程中，不當(dāng)?shù)倪x擇切削條件造成的表面粗糙度高和尺寸誤差，它甚至可能發(fā)生動(dòng)力現(xiàn)象：由于自動(dòng)興奮的震動(dòng)，可以設(shè)定在[2]。鑒于銑削運(yùn)行在今天的全球制造業(yè)中起著重要的作用，就必要優(yōu)化加工參數(shù)。因此通過努力，在這篇文章中看到刀具幾何（徑向前角和刀尖半徑）和切削條件（切削速度和進(jìn)給速度），表面精整生產(chǎn)過程中端銑中碳鋼的影響。實(shí)驗(yàn)顯示，這項(xiàng)工作將被用來測試切削速度，進(jìn)給速度，徑向前角和刀尖半徑與加工反應(yīng)。數(shù)學(xué)模型的進(jìn)一步利用，尋找最佳的工藝參數(shù)，并采用遺傳算法可促進(jìn)更大發(fā)展。2回顧建模過程與優(yōu)化，是兩部很重要的問題，在制造業(yè)。生產(chǎn)過程的特點(diǎn)是多重性的動(dòng)態(tài)互動(dòng)過程中的變數(shù)。表面光潔度一直是一個(gè)重要的因素，在機(jī)械加工性能預(yù)測任何加工操作。為了開發(fā)和優(yōu)化表面粗糙度模型，有必要了解目前在這方面的工作的狀況。迪維斯等人[3]調(diào)查有關(guān)切削加工性能的五個(gè)銑刀具有不同螺旋角。分別對鋁合金L65的3向銑削過程（面，槽和側(cè)面）進(jìn)行了切削試驗(yàn)，并對其中的切削力，表面粗糙度，凹狀加工平面進(jìn)行了測量。所進(jìn)行的若干實(shí)驗(yàn)是用來決定該中心復(fù)合設(shè)計(jì)的。切削性能的立銑刀則被評定采用方差分析。對主軸速度，切削深度和進(jìn)給速度對切削力和表面粗糙度的影響進(jìn)行了研究。調(diào)查顯示銑刀與左手螺旋角一般不太具有成本效益比。上下銑方面切削力與右手螺旋角，雖然主要區(qū)別在于表面粗糙度大，但不存在顯著差異。拜佑密等人[4]研究過工具對旋轉(zhuǎn)角度，進(jìn)給速度和切削速度在機(jī)械工藝參數(shù)（壓力，摩擦參數(shù)）的影響，為端銑操作常用三種商用工件材料，11L17易切削鋼，62-35-3易切削黃銅和鋁2024年使用單一槽高速鋼立銑刀。目前已發(fā)現(xiàn)的壓力和摩擦法對芯片-工具接口減少，增加進(jìn)給速度，并與下降的氣流角，而切削速度已微不足道，對一些材料依賴參數(shù)，工藝參數(shù)，歸納為經(jīng)驗(yàn)公式，作為職能的進(jìn)給速度和刀具旋轉(zhuǎn)角度為每個(gè)工作材料。不過，研究人員也還有沒有考慮到的影響，如切削條件和刀具幾何同步，而且這些研究都沒有考慮到切削過程的優(yōu)化。因?yàn)槎算娺^程介入多數(shù)f參量，重大參量的聯(lián)合只能通過塑造得到。曼蘇爾和艾布達(dá)萊特基地[5]已開發(fā)出一種表面粗糙度模式，為年底銑EN32M（半自由切削碳硬化鋼并改進(jìn)適銷性）。數(shù)學(xué)模型已經(jīng)研制成功，可用在計(jì)算切削速度，進(jìn)給速度和軸向切深。這些參數(shù)對表面粗糙度的影響已進(jìn)行了響應(yīng)面分析法（丹參）。分別制定了一階方程涵蓋的速度范圍為30-35米/分，一類二階方程涵蓋速度范圍的24-38米/分的干切削條件。艾爾艾丁等人[6]開發(fā)出一種表面粗糙度模型，用丹參，為端銑190BHN鋼。為選擇適當(dāng)?shù)慕M合，切割速度和伺服，增加金屬