渦輪減速器箱體加工工藝及夾具設(shè)計【蝸輪減速箱體】
渦輪減速器箱體加工工藝及夾具設(shè)計【蝸輪減速箱體】,蝸輪減速箱體,渦輪減速器箱體加工工藝及夾具設(shè)計【蝸輪減速箱體】,渦輪,減速器,箱體,加工,工藝,夾具,設(shè)計,蝸輪,減速
Proceedings ofthe2006 IEEE/RSJ International Conference on Intelligent Robots and Systems October9- 15, 2006, Beijing, China ANovelModularFixtureDesignandAssemblySystem BasedonVR PengGaoliang, LiuWenjian SchoolofMechatronicsEngineering HarbinInstituteofTechnology Harbin, 150001, China pgl7782a Abstract - Modular fixtures are one oftheimportant aspects ofmanufacturing. This paper presents a desktop VR system for modular fixture design. The virtual environmentis designed and the design procedure is proposed. It assists the designer to make the feasible design decisions effectively and efficiently. A hierarchical data model is proposed to represent the modular fixture assembly. Based on this structure, the user can manipulate the virtual models precisely in VE during the design and assembly processes. Moreover, the machining simulation for manufacturing interaction checking is discussed and implemented. Finally, the case study has demonstrated the functionality of the proposed system. Compared with the immersive VR system, the proposed system has offered an affordable andportable solutionformodularfixtures design. Index Terms - Modularfixture, desktop VR, assembly design, machiningsimlulation. I. INTRODUCTION Modular fixtures are one of the important aspects of manufacturing. Proper fixture design is crucial to product quality in terms of precision, accuracy, and finish of the machined part. Modular fixture is a system of interchange- eable and highly standardized components designed to securely and accurately position, hold, and support the workpiece throughout the machining process 1. Tradition- ally, fixture designers rely on experience or use trial-and- error methods to determine an appropriate fixturing scheme. With the advent of computer technology, computer aided design has been prevalent in the area of modular fixture design. In general, the associated fixture design activities, namely setup planning, fixture element design, and fixture layout design are often dealt with at the downstream end of the machine tool development life-cycle. These practices do not lend themselves well to the bridging of design and manufacturing activities. Forexample, very few systems have incorporated the functionality of detecting machining interference. This leads to a gap between the fixture design andmanufacturing operationswheretheaspectofcutterpaths is not considered during the design stage 2. As a result, re- designcannotbeavoidedwhenthecutterisfoundtointerfere with the fixture components in the manufactu- ring set-up. Therefore, in orderto bring machining fixture design into the arenaofflexiblemanufacturing, amoresystematicandnatural designenvironmentisrequired. As a synthetic, 3D, interactive environment typically generated by a computer, VR has been recognized as a very powerful human-computer interface for decades 4. VR holds great potential in manufacturing applications to solve problems before being employed in practical manufacturing thereby preventing costly mistakes. The advances in VR technology in the last decade have provided the impetus for applying VR to different engineering applications such as product design 5, assembly 6, machining simulation 7, andtraining 8. The goal ofthis paper is to develop a VR- basedmodular fixtures design system (VMJFDS). This is the firststepto develop anintegratedandimmersiveenvironment for modular fixture design. This application has the advantages of making the fixture design in a natural and instructive manner, providing better match to the working conditions, reducing lead-time, and generally providing a significantenhancementoffixtureproductivityandeconomy. II. OVERVIEWOFTHEPROPOSEDSYSTEM The system architecture of the proposed desktop VR systemismodularisedbasedonthefunctionalrequirements of thesystem,whichisshowninFig.1. Atthesystemlevel,three modules of proposed system, namely, Graphic interface (GUI), Virtual environment (VE) and Database modules are designed. For each ofthe modules, a set ofobjects has been identified to realize its functional requirements. The detailed objectdesignandimplementation are omittedfromthispaper. Instead, the briefdescription ofthese three modules is given below. 1) Graphic Interface (GUI): The GUI is basically a friendly graphic interface that is used to integrate the virtual environmentandmodularfixturedesignactions. 2) Virtual environment (VE): TheVEprovidestheusers with a 3D display for navigating and manipulating the models of modular fixture system and its components in the virtual environment. As shown in Fig. 1, the virtual environment module comprises two parts, namely assembly design environment andmachiningsimulationenvironment. Theuser selects appropriate elements andputs downthese elements on the desk in the assembly design area. Then he assembles the selected elements one by one to build up the final fixture systemwiththeguidanceofthesystem. 1-4244-0259-X/06/$20.00 C)2006IEEE 2650 Authorized licensed use limited to: Nanchang University. Downloaded on December 20, 2009 at 22:44 from IEEE Xplore. Restrictions apply. Fig.1.OverviewofthedesktopVRbasedmodularfixturedesignsystem. 3) Database: The database deposit all of the models of environment and modular fixture elements, as well as the domain knowledge and useful cases. There are 5 databases shown in Fig.1. Among them, knowledge & rule base governing all fixture planning principles forms the brains of thesystem. III. PROCEDUREOFMODULARFIXTUREDESIGN In this section, an instructive modular fixture design procedure within VE is presented. Besides the 3D depth that the users feel and the real-world like operation process, this procedure features intelligence and introduction. During the design process, some useful cases and suggestion will be presented to the user for reference based on intelligent inference method such as Case based reasoning (CBR) and Rule based reasoning (RBR). Further more, relative knowledge andrules arepresented ashelppages thattheuser caneasilybrowsedduringthedesignprocess. Overview of modular fixture design process is summarized in Fig. 2. After the VE environment is initialed andthe workpiece is loaded, the first step is fixtureplanning. Inthis step, theuserfirstdecides thefixturing scheme, thatis specifies the fixturing faces of the workpiece interactively. Forhelptheusersdecision-making, someusefulcasesaswell as their fixturing scheme will be presented via the automatic CBR retrieval method. Once the fixturing faces are selected, theusermaybepromptto specifythefixturingpoints. Inthis task, somesuggestions andrulesaregiven. After the fixturing planning, the next step is fixture FUs design stage. In this stage, the user may be to select suitable fixture elements andassembletheseindividualparts into FUs. According to the spatial information ofthe fixturingpoints in relation to the fixture base and the workpiece, some typical FUs and suggestions may be presented automatically. These willbehelpfulfortheuser. AftertheplanningandFUs design stage, the next stage is interactively assembling the designed fixtureFUstoconnecttheworkpiecetothebaseplate. When the fixture configuration is completed, the result will be checked and evaluated within the machining environment. The tasks executed in this environment including assembly planning, machining simulation, and fixture evaluation. Assemblyplanning isusedto gain optimal assembly sequence and assembly path of each component. Machining simulation is responsible for manufacturing interaction detection. Fixture evaluation will check and evaluate the design result. In conclusion, the whole design process isinanaturemannerforthebenefitofVE. Moreover, the presented information of suggestion and knowledge can advise the user on how to make decisions ofthe best design selection. IV. ASSEMBLYMODELINGOFMODULARFIXTURE A. Modularfixturestructureanalysis A functionalunit(FU) is acombination offixture elements to provide connectionbetweenthebaseplate and aworkpiece 11. Generally, modularfixture structuremaybe dividedinto three functional units according to its basic structure characteristics, namely locating unit, clamping unit, and supporting unit. The number offixture elements in aFU may consist ofone or more elements, in which only one element serves as a locator, support or clamp. The major task ofthe modularfixture assembly is to selectthe supporting, locating, clamping and accessory elements to generate the fixture FUs toconnecttheworkpiecetothebaseplate. By analyzing the practical application ofmodular fixtures, it is found that the assembly ofmodular fixtures begins by selecting the suitable fixture elements to construct FUs, then subsequentlymountingtheseFUs onthebaseplate. Therefore, the FUs can be regarded as subassemblies ofmodular fixture system.Further,thestructureofmodularfixturesystemcanbe representedasahierarchalstructureasshowninFig.3. 2651 Authorized licensed use limited to: Nanchang University. Downloaded on December 20, 2009 at 22:44 from IEEE Xplore. Restrictions apply. UsefTa6 *T- siikg&Sugge lr,l Fixtui e Elemenets rUetrieval i0 Tools rKetrieval 4 Fig.2Modularfixturedesignprocedureinproposedsystem B. Hierarchically structured data modelfor modularfixture representation in VE It is common that the corresponding virtual environment may contain millions ofgeometric polygon primitives. Over thepastyears, anumberofmodel sub-division schemes, such asBSP-tree 10 andOctrees,havebeenproposedto organize largepolygonalmodels.However, formodular Ba 1I_ 1 Hsreplalte Bansepla1nte Elements *Locatng ElementsL,cating Units AccessoryEllements ClamnpingElemnents !ClampingUnits SupportingElemntsSupporting Ufnits Accessory Elements Fig. 3Hierarchical structureofmodularfixture system design applications, the scene is also dynamically changing, due to interactions. For example, in design process, the part object may change its spatial position, orientation and assembly relations. This indicates that a static representation, such as BSP-tree, is not sufficient. Further more, the above models can only represent the topology structure of fixture system in the component level. However, to the assembly relationship among fixture components, which refers to the mating relationship between assembly features that is not concerned. In this section, we present a hierarchically structuredandconstraint-baseddatamodelformodularfixture system representation, real-time visualization and precise 3D manipulationinVE. As shown in Fig.4, the high-level component based model is used for interactive operations involving assemblies or disassembles. It provides both topological structure and link relationsbetweencomponents. Theinformationrepresent- ed in the high-level model can be divided into two types, i.e. component objects and assembly relationships. Component objects can be a subassembly or a part. A subassembly consists of individual parts and assembly relationships betweentheparts. Component Level (Pt Part S Subassembly Assembly relationship Feature Level Ft3 Feature Feature mating relationship t- -t Polygon Level FZ-ll. Polygon Fig.4ThehierarchicalstructuredatamodelinVE Themiddle-levelfeaturebasedmodelisbuiltuponfeatures and feature constraints. In general, the assembly relationship often treated as the mating relationships between assembly features. Thus the featurebasedmodel isusedto describethe assembly relationship andprovides necessary information for spatial relationship calculating during assembly operation. In this model, only the feature relationships between two different components are considered. The relationship between features ofone element will be discussed in feature basedmodularfixtureelementmodelingbelow. The low-level polygon based model corresponds to the above two level models for real-time visualization and interaction. It describes the entire surface as an inter- connected triangular surface mesh. More about how the polygons organized of a single element will be discussed is thenextsection. C. Modularfixtureelementsmodeling As we know, in VE, the part is only represented as a number ofpolygon primitives. This result in the topological 2652 Authorized licensed use limited to: Nanchang University. Downloaded on December 20, 2009 at 22:44 from IEEE Xplore. Restrictions apply. relations- hips and parametric information are lost during the translation process of models from CAD systems to VR systems. However, this important information is necessary in design and assembly process. In order to fulfill the requirements, we present a modeling scheme for fixture elementsrepresentationinthissection. The modular fixture elements are pre-manufactured parts withstandarddimensions. Afterthefixturingschemedesigned, the left job is to select suitable standard elements and assemblethese elements to formafixture systeminafeasible andeffectivemanner. Therefore, intheproposed system, only the assembly features of the fixture elements need to be considered. Inthispaperanassemblyfeature isdefinedas apropertyof afixture element, whichprovidesrelatedinformationrelevant to modular fixture design and assembly/disassembly. The following eight function faces are defined as assembly featuresoffixtureelements: supportingfaces, supportedfaces, locating holes, counterbore holes, screw holes, fixing slots, andscrewbolts. Besidestheinformation aboutthefeaturelike typeanddimension, otherparameters, i.e. therelativeposition andorientationofthe featureintheelements localcoordinate system are recorded with the geometric model in the fixture element database. When one element assembles with another, the information aboutthematedfeatures isretrieved andused to decide the spatial relationship ofthe two elements. More information about the assembly features and their mating relationship arediscusseddetailedinRef 1. D. Constraintbasedfixtureassemblyin VE 1)Assemblyrelationshipbetweenfixtureelements Mating relationships have been used to define assembly relationships between part components in the field of assembly. According to the assembly features summarized in the above section, there are fivetypes ofmating relationships between fixture elements. Namely against, fit, screw fit, across, andT-slotfit,which are illustrated inFig. 5. Based on these mating relationships, we can reason the possible assemblyrelationshipofanytwoassembledfixtureelements. 2)Assemblyrelationshipreasoning Ingeneral, the assemblyrelationship oftwo assembledpart isrepresented as thematedassembly featurepairs ofthem. In the above section, we defined five basic mating relationships between fixture elements. Therefore, it is enabled to decide the possible assembly relationships through finding the possible mating assembly feature pairs. These possible assembly relationships are saved in assembly relationships database(ARDB)forfixtureassemblyinnextstage. However, when the fixture is complicated and the numbers ofcomposite fixture elements is large, the possible assembly relationships are too much to take much time for reasoning andtreating. To avoidthis situation, wefirstdecide the possible assembled elements pairs. That is to avoid reasoning the assembly relationship between a clamp andthe baseplate, for they never were assembled together. In this stage, some rules are utilized to find the possible assembled elementspairs. The algorithm of assembly relationships reasoning is similar to what discussed in Ref 12. Thus the detailed descriptionofthealgorithmisomittedfromthispaper. (a) AIlai.ns .2 l.I.F LIi I7 F d) Asicmie 1f-isxkt Elmn Fig. 5Fivebasicmatingrelationshipsbetweenfixtureelements 3)Constraint-basedfixtureassembly Aftercarrying outthe assemblyrelationships reasoning, all possible assembly relationships ofthe selected elements are establishedandsavedinARDB. Basedontheserelationships, the trainee can assemble these individual parts to a fixture system. This section is about the discussion of interactive assembly operation in VE. The process ofa single assembly operation is presented in Fig.5 and illustrated by two simple partsassemblyasshowninFig.6. In general, the assembly operation process is divided into three steps, namely assembly relationship recognizing, constraint analysis and applying, constraint-based motion. Firstly, the trainee selects an element and moves it to the assembled component. Once an inference between the assembling and assembled component is detected during the moving,the inferredfeatures is checked. Ifthetwo features is one of the assembly relationships in ARDB, they will be highlighted and will await the users confirmation. Once it is confirmed, the recognized assembly relationship will be appliedby constraint analyzing and solving, that is adjustthe translationandorientationoftheassemblingelementtosatisfy the position relationship ofthese two components, as well as applythenew constrainttotheassemblingelement.Whenthe new constraint is applied, the motion of the assembling element will be mapped into a constraint space. This is done bytransferring 3Dmotiondatafromtheinputdevicesintothe allowable motions ofthe object. The constraint-based motion notonlyensuresthattheprecisepositionsofacomponentcan be obtained, but also guarantee that the existing constraints will not be violated during the future operations. The assembling element will reach to the final position through succession assembly relationship recognizing and constraint applying. 2653 Ii 1-11 4- (b) F.t Authorized licensed use limited to: Nanchang University. Downloaded on December 20, 2009 at 22:44 from IEEE Xplore. Restrictions apply. NO Assembly relationship Iis possible checking elatioohship? Fig. 6Processofassemblyconstraintestablishment No V. MACHINING SIMULATION A. Manufacturinginteractions During the machining process, there are many types of manufacturing interactions associated with the fixture may occur. These interactions can be divided into two broad categories illustrated below, namely static interactions and dynamicinteractions. 1) Static interactions refer to the interference between fixture components, the interference between fixture components and machine tool, and the interference between fixture components andmaching feature ofworkpiece during theworkpiecesetup. 2)Dynamicinteractionsrefertothetool-fixtureinteractions, which occur within a single operation when the tool and the fixtureusedinthatoperationmaycollideduringcutting. Generally, the aspects of machining process and cutter paths are not considered duringthe fixture design stage. As a result, these interactions may often occur during the practical manufacturing. Thus the human machinists have to spend muchoftheirtimeidentifyingtheseinteractions andresolving them. Itis oftenresults inmodification orre-designoffixture system. Thatistediousandtimecostly. B.Interferencedetection Although the currently commercial software, like VERICUT, can simulates NC machining to detect tool path errors and inefficient motion prior to machining an actual workpiece. It is available to eliminate errors that could ruin the part, damage the fixture, break the cutting tool, or crash the machine during the part programming stage. However, these software are expensive and oriented to NC program- mertherebynotsuitableforfixturedesigners. During the fixture design stage, it should be ensured that the associated fixture interactions can be avoided. In this system, after the fixture configuration is complete, the machining simulation module is presented to the user to identifytheinteractionsandresolvethem. Within the machining simulation environment, the 3D digitalmodelofmachinetoolispresented. The canassemble the fixture components on the work bench and setup the workpiece, just as what the machining engineers do in the actual site. During the setup, the fixture components and the workpiece are move to their assembly position under manipulation. Theinterferencecheckingmoduleiscarriedout. Ifinterference occurs, the inferred objectwill be highlight. It is p
收藏