mctd

of 14
19 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
Journal of Materials Processing Technology 185 (2007) 46–59 Development of hybrid predictive models and optimization techniques for machining operations I.S. Jawahir ∗ , X. Wang Center for Manufacturing, Department of Mechanical Engineering, University of Kentucky, Lexington, KY 40506-0108, USA Abstract This paper presents a summary of recent developments in modeling and optimization of machining processes, focusing on turning and milling operations. With a brief analysis of past research on p
Document Share
Document Tags
Document Transcript
  Journal of Materials Processing Technology 185 (2007) 46–59 Development of hybrid predictive models and optimizationtechniques for machining operations I.S. Jawahir ∗ , X. Wang Center for Manufacturing, Department of Mechanical Engineering, University of Kentucky, Lexington, KY 40506-0108, USA Abstract This paper presents a summary of recent developments in modeling and optimization of machining processes, focusing on turning and millingoperations. With a brief analysis of past research on predictive modeling, the paper presents the analytical, numerical and empirical modelingeffortsfor2Dand3Dchipformationcoveringthedevelopmentofauniversalslip-linemodel,acomprehensivefiniteelementmodel,andintegratedhybrid models. This includes a newly developed equivalent toolface (ET) model and new tool-life relationships developed for machining withcomplex grooved tools. At the end, a performance-based machining optimization method developed for predicting optimum cutting conditionsand cutting tool selection is presented. The paper also highlights the need for considering a machining systems approach to include the integratedeffects of workpiece, cutting tool and machine tool.© 2006 Elsevier B.V. All rights reserved. Keywords: Machining performance; Modeling; Optimization; Cutting tools; Slip-line model; Finite element model 1. Introduction Machining operations constitute a large segment of the man-ufacturing sector in the U.S. However, a recent CIRP (TheInternational Institution for Production Engineering Research)working paper[1]reports the survey results of a major cutting tool manufacturer as “ ... In the USA, the correct cutting toolis selected less than 50% of the time, the tool is used at therated cutting speed only 58% of the time, and only 38% of thetools are used up to their full tool-life capability. ... ” This situ-ation urges the need for developing more scientific approachesto select cutting tools and cutting conditions for optimum eco-nomic and technological machining performance.Selectionofcuttingtoolsandcuttingconditionsrepresentsanessentialelementinprocessplanningformachining.Thistaskistraditionallycarriedoutonthebasisoftheexperienceofprocessplanners with the help of data from machining handbooks andtool catalogs.Turning and milling operations are among the most commonmachining operations performed in automotive, aerospace andother application industries. Process planners continue to expe-rience great difficulties due to lack of performance data on thenumerous new commercial cutting tools with different materi- ∗ Corresponding author. Tel.: +1 859257 6262; fax: +1 859257 1071.  E-mail address: jawahir@engr.uky.edu(I.S. Jawahir). als, coatings, geometry and chip-groove configurations for highwear resistance and effective chip breaking (in turning), etc.Also, specific data on relevant machining performance mea-sures such as tool-life, surface roughness, chip-form, etc., arehard to find due to lack of predictive models for these measures.Consequently, the process planners are forced to choose andrecommend sub-optimal cutting conditions for machining oper-ations.Whilesignificantemphasisisbeingplacedondevelopingcomprehensive predictive models for machining performance,the progress so far has been very limited due to the inherentcomplexities involved in the actual machining operations oftenassociated with the inconsistencies in work material and cut-ting tool properties and the machine tool conditions. Also, theanalytical and numerical tools and techniques used for predict-ingmachiningperformancemeasuresarelessaccurateandquitetimeconsuming,eventhoughthecomputationalpowerhasbeencontinuing to increase to date. 2. A summary of the state of the art in modeling andoptimization of machining operations 2.1. Fundamental work and renewed interest for developing predictive performance models The well known early model for 2D machining by Ernst andMerchant presented an analytical approach as an alternative to 0924-0136/$ – see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.jmatprotec.2006.03.133   I.S. Jawahir, X. Wang / Journal of Materials Processing Technology 185 (2007) 46–59 47 thehighlyempirical-basedapproachesprevalentuntilthen[2,3].LeeandShaffer[4]f ollowedMerchant’sworkbyapplyingplas- ticity theory to the machining problem. Palmer and Oxley[5]studied the chip formation process using cine’ filming meth-ods and provided a variable flow stress model for shear zonein machining. Roth and Oxley[6]f ollowed up this work with flow streamlines using printed grids and constructed a slip-linefield from measured flow velocities. Dewhurst[7]showed that the machining process, with quasi-static curled chip formation,is not uniquely defined by any given set of steady state cut-ting conditions. It was also shown that a decrease in frictionreduced the chip curvature but increased the forces, chip thick-ness and contact length. Oxley and his co-workers developed acomprehensive predictive theory for machining involving vari-able flow stress theory[8].In recent years, a greater emphasis has been placed on developing predictive models for machiningperformance measures, most notably by Usui in 1988[9];Mer- chant in 1993[10];Armarego in 1996[11];Oxley in 1998[12]; and Jawahir and Balaji in 2000[13].International collaborative effort in this area has resulted in three major keynote papers:one at ASME in 1997[14]and two other papers at CIRP in1998 and 2003[15,16],all highlighting the need for developing predictive performance models for machining operations. TheCIRP has been largely instrumental in bringing out a renewedinterest among international researchers with its founding of aninternationalworkinggroupin1995undertheleadershipofPro-fessor C.A. van Luttervelt. This major milestone activity pickedup its momentum with the beginning of an annual internationalworkshop series on modeling of machining operations in 1998,and the first workshop was held in Atlanta, GA, USA whichwas attended by 72 researchers from 14 countries. This work-shopwasalsoco-sponsoredbytheNationalScienceFoundation(USA), and the North American Manufacturing Research Insti-tuteoftheSocietyofManufacturingEngineers(NAMRI/SME).In continuing with this major event, so far six workshops havebeen organized in different countries, and an excellent collec-tionofcontributionshasbeenproducedintheformofworkshopproceedingscontainingallpresentedpapers,summaryofdiscus-sions, etc.[17–22].The 7th workshop will be held in May 2004 at ENSAM, Cluny, France, where a total of 40 technical papersare expected to be presented, while the 8th workshop is plannedfor 2005 to be held in Chemnitz, Germany. The recent CIRP-sponsored survey conducted by Armarego et al. concludes thata majority of research groups worldwide are active on devel-oping empirical models for predicting machining performancewhile significant effort is being made on developing analyti-cal and FEM-based numerical models[23].Tool-life, surface roughness, part accuracy and chip control are identified to beamong the most needed machining performance measures forpredictive models. 2.2. Traditional methods for machining optimization The role of objective function in optimization of machin-ing operations is extremely important but is often difficult todefine due to the complex interactions that take place duringthe machining process. Traditional approaches in machiningoptimization have been limited to objective functions relatedto cost or productivity[24–27].Although such an objective is desirable, the more critical role of optimization lies in the needforoptimizingthevariousmachiningperformancemeasuresforhigher productivity and enhanced product quality. The conflict-ingmachiningperformancerequirements,dependingonspecificapplications,resultintheneedfortheoptimizationofmachiningprocessesbytheselectionofthemostsuitablecuttingconditionsandcuttingtools,aswellasbythesuggestionforachievingopti-mizedmachiningperformance.Almostallrelationshipsbetweenthe performance variables and process parameters employed byprevious research are approximated by power functions withfixed empirical coefficients. This may be attributed to the non-availability of quantitatively reliable machining performancemodelsrelatingthemachiningperformancemeasurestothepro-cess variables. The lack of technological performance data andequations, as well as detailed machine tool specifications andcapabilities,haslimitedthewidespreaduseoftheavailableopti-mization strategies[11].Traditional approaches have not fully considered the need for performance predictions. For example,the functional requirement of chip breaking, and the associateduse of commercial complex grooved tools in turning opera-tions, are generally ignored in these methods. Early work onoptimizationofmulti-passmachiningoperationsshowstheeco-nomic benefits of machining[28].Most recent research demon- Fig. 1. (a) Extended universal slip-line model and (b) its hodograph for machining with grooved tools[34].  48 I.S. Jawahir, X. Wang / Journal of Materials Processing Technology 185 (2007) 46–59 strates the use of advanced computer-aided methodologies andintelligent techniques for multi-pass machining optimization[29–32].However, the most needed performance-based opti-mization methods were lacking in these works, despite the useof advanced tools and techniques for optimization. 3. Predictive models for machining operations 3.1. Fundamental work on 2D and 3D chip formationmodels Recent work on modeling of machining operations involvesthe development of an analytical predictive model to takeaccountofthecyclicchipformationprocessincludingchipcurl,andchipbreaking[33].Thefundamentalworkinvolvingplastic- ityapplicationsinmachininghasrecentlyledtothedevelopmentof a universal slip-line model by Fang et al[34].Based on the rigid-plastic and plane-strain assumptions used in the classicslip-line theory, the model takes into account of both, the chipup-curling, and chip back-flow effects and incorporates six pre-viously developed slip-line models for machining during thelast six decades as special cases: Dewhurst’s model[7],Shi and Ramalingam’s model[35],Kudo’s model[36],Johnson’s[37] and Usui and Hoshi’s[38]models, Lee and Shaffer’s model[4], and Merchant’s model[3].This universal slip-line model has been validated through extensive cutting tests covering a widerangeofcuttingconditions.Byincorporatinganadditionalchip-groove backwall force, the universal slip-line model is extendedfor machining with restricted grooved tools as shown inFig. 1,where AC is a convex upward shear plane; θ , ψ , η 1 , and η 2 are four slip-line angles; and F  b and N  b are frictional and nor-mal forces acting on the chip-groove backwall. This model hassubsequently been extended to include the effects of strains,strain-rates and temperature by combining it with Oxley’s clas-sic predictive model[8]through an iterative process matching thecuttingforcesobtainedinmachiningwithgroovedtools[39].Further work includes the establishment of tool–chip interfacefrictional conditions and the effects of chip-groove parameters(groove width, backwall height, etc.) on chip formation[40].Recent efforts by various active researchers on developingnumerical/computational techniques to simulate the machiningprocesses are significant[41–45].A numerical model for 2D cutting with a grooved tool insert was developed with a ther-mal rigid-viscoplastic material model and was later extended toinclude the cyclic chip formation[46].Fig. 2shows the free- body-diagram developed for this cyclic chip formation and atypical set of results obtained from this finite element analysisfor stresses, strain-rates and temperatures. Recently, this modelhasbeenusedinconjunctionwithadamagemechanicsmodeltostudy the development of void growth and coalescence in orderto relate it to the chip breaking process[47].With a thermal elastic–viscoplastic material model, residual stress formation inorthogonal machining has also been studied[48].An operation-based 3D machining model has been devel-oped for predicting cutting forces and chip flow angle in turningoperations involving the use of a nose radius tool having a finitecutting edge radius[49].This work was subsequently extended Fig. 2. (a) The free body diagram showing the forces acting on the chip, and(b) and (c) a set of typical results obtained from the finite element analysis formachining with grooved tools.   I.S. Jawahir, X. Wang / Journal of Materials Processing Technology 185 (2007) 46–59 49 to include the effects of chip side–curl and the related tool–chiptribological interactions[50].Fig. 3shows the variations in tool–chip contact in machining with flat-faced and a groovedtools. This work shows that variable tool–chip interface fric-tionalconditions,prevalentinmachiningtypicallywithcomplexgrooved tools, can be interrelated to the tool-wear mechanismsand the associated chip-forms produced in machining. 3.2. Predictive models for machining performance3.2.1. Surface roughness prediction Surface roughness, an important functional requirement of amachinedcomponent,alsoservestomaintaindimensionalaccu-racy and tolerances. The arithmetic average for surface rough-ness is generally expressed in terms of feed, f  and the tool noseradius, r  e . Although different, but similar, geometric relation-ships for average surface roughness exist, none, however, fullycorrelates with the measured values in finish machining. Theserelationshipsarealsonotuniquelyacceptableinpresent-dayfin-ish machining operations involving complex grooved tools andvarying edge geometry conditions. Hence, there is an urgentneed for developing predictive models for surface roughnessand surface integrity. Fang and Jawahir[51]showed the sur-face roughness variations in relation to changes in operationalparameters, work materials and chip-breaker types. Wide vari-ations are seen in the measured surface roughness for differentwork materials employing the same cutting conditions. Turningoperations also give much higher measured surface roughness(  R a )valuesthanthepredictedtheoreticalvaluesinthefinishturn-ing range[52].Therefore, a database is established for various work materials and cutting tools by measuring R a experimen-tally for a set of different cutting conditions. R a was representedas a function of feed and depth of cut, and the effect of cut-ting speed was found to be very minimal in the finish turningregion[53,54].Developing accurate empirical relationships for predictingsurfaceroughnesshasbeenanongoingresearchwork and recent attempts includes the use of work material proper-ties along with cutting conditions and tool geometry includingthe edge geometry conditions such as the cutting edge (hone)radius. 3.2.2. Chip-form/chip breakability prediction Chip-form/chip breakability is considered as a basic require-mentinautomatedmachining,particularlyinturningoperations.Therecentlydevelopeduniversalslip-linemodelcanbeusedforpredicting chip curl and other machining variables for machin-ing with grooved tools[34,39].This model has subsequently been extended to include the effects of chip-groove parameters[40].Also, by using the custom-built, finite element modelingtechniques,theeffectsofcuttingconditions,toolgeometry,chip-groove parameters on stresses, strains, strain-rates and tempera-tures are predicted[46–48].In the ongoing work, the previously established definition of chip-form/chip breakability has beenused, which assumes that the size, shape and difficulty/easeof chip producibility determine the levels of chip breakability.Accordingtothedefinition,thevaluesofchipbreakabilityrangebetween 0 and 1, with ‘0’ for absolutely unbroken chips and ‘1’ Fig. 3. Variable tool–chip contact in (a) flat-faced tools; (b) grooved tools; (c)complex chip back- and side-flow mechanisms in turning with a grooved tool[50].
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x