毕业论文

打赏
当前位置: 毕业论文 > 外文文献翻译 >

互联网的注射模具智能设计系统外文文献和中文翻译(4)

时间:2019-06-01 15:01来源:毕业论文
The coded representation of allnecessary information about a plastic part together with itsrespective plastic mould is illustrated in Fig. 4.5.2. The knowledge baseThe knowledge base in this system wa


The coded representation of allnecessary information about a plastic part together with itsrespective plastic mould is illustrated in Fig. 4.5.2. The knowledge baseThe knowledge base in this system was acquired from aliterature search and from a number of mould makingcompanies. There are a number of useful injection moulddesign and mould-making rules as well as technical data inthe technical papers, handbooks and trade standards. Theexpert knowledge and empirical rules were also collectedthrough interviews and case discussions with the experts ofsome local companies. The scope of knowledge covers themost widely used methods and elements of mould featuresfor the main functional design modules for standard two-plate moulds, stripper moulds, split moulds and three-platemoulds.The internal structure of the knowledge base consists ofthree parts namely the methods file, the alternative mouldfeatures by method database and mould feature geometrydatabase. The methods file contains a long list of methodsof achieving the various functional designs of an injectionmould. The method screening routine (block 8 in Fig. 3)uses the part code to select the recommended list ofmethods. Once method has been selected a code associatedwith each chosen method would be generated and is thenused to access the alternative mould features of the chosenmethod through the mould feature screening routine (block9in Fig. 3).
When particular mould feature of chosenmethod has been selected, a code associated with eachmould feature would be produced and is then used toaccess and prompt the designer to input informationconcerning inpidual dimensions, position and orientationfrom the mould feature geometry database through theautographic routine (Fig. 3). In the mould featuregeometry database, the detailed dimensions of the geome-try of all the mould features are stored in terms ofparameters. Once the graphic information has beeninputted, the geometry of the chosen mould feature wouldbe drawn. The schematic representation of the internalstructure of the KB is illustrated in Fig. 5.A rule-based knowledge-representation method was usedto develop the knowledge base. This method is the mostpopular and versatile of all the representation schemes.Knowledge is represented in the form of production rules.Rules are written as IF-THEN statements to provide aformal way of representing strategies, directives andrecommendations. This scheme is appropriate for a varietyof KB systems problem domains. More than 500 rules havebeen established in this work. The various functionaldesigns of injection moulds are related to part informationand mould specifications by rules developed in the knowl-edge base. For example, the rules for the cavity layoutand part of the rules for the cooling system are shown inTables 1 and 2, respectively.The JESS 4.4 expert system shell [22–24] was selected todevelop the knowledge base of the Internet-based moulddesign system. JESS is a forward chaining productionsystem implemented in the Java programming language.Rules are written in CLIPS/JESS language, which caneasily call and be called by Java programs. The programmefor the knowledge base was written in modular structureconsisting of the five basic functional systems of injectionmould design, namely cavity layout, feed system, mouldconstruction, cooling system, and ejection system. Thecapability of the knowledge base can be further expandedthrough continual refinements in the rules and databases ofthe system. Using the system’s editing facilities, it is fairlyeasy to add and delete rules in the knowledge base. Inaddition, the assignments and combinations of certaintyfactors can be added, and user queries can also be added ordeleted where it seems appropriate.6. System implementationA prototype system has been implemented based onJ2EE. This system is combined with a commercial 3D CAD system (SolidWorks 2003) and a commercialdatabase (Microsoft Access 2000). As mentioned in Section5.2, the JESS 4.4 expert system shell was used to developthe knowledge base of the system. 互联网的注射模具智能设计系统外文文献和中文翻译(4):http://www.youerw.com/fanyi/lunwen_34039.html
------分隔线----------------------------
推荐内容