3560 挖掘機(jī)工作裝置設(shè)計(jì),挖掘機(jī),工作,裝置,設(shè)計(jì) 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 1 -外文翻譯原文出處 Submitted to ASME/JDSMC Special Issue on Sensors作 者 Jagannath Yammada, Terrence L. Chambers, Suren N. Dwivedi原文標(biāo)題 Intelligent Mold Design Tool For Plastic Injection Molding譯文標(biāo)題 注塑成型的智能模具設(shè)計(jì)工具摘要注塑成型是一個(gè)生產(chǎn)熱塑性塑料制品最流行的制造工藝，而模具設(shè)計(jì)是這個(gè)過(guò)程的一個(gè)重要方面。模具設(shè)計(jì)需要專業(yè)的知識(shí)、技能，最重要的是擁有該領(lǐng)域的經(jīng)驗(yàn)。三者缺一不可。生產(chǎn)塑料組件需要選擇恰當(dāng)?shù)哪＞?，如果缺乏其中之一，這種選擇就得在反復(fù)試驗(yàn)的基礎(chǔ)上進(jìn)行。這會(huì)增加生產(chǎn)成本，并造成設(shè)計(jì)上的不一致。本文介紹了智能模具設(shè)計(jì)工具的發(fā)展。該工具捕獲模具設(shè)計(jì)過(guò)程的知識(shí)，并且以符合邏輯的方式將這些知識(shí)反映出來(lái)。所獲得的知識(shí)將是確定性的，但模具設(shè)計(jì)過(guò)程中的信息是非確定的。一旦開發(fā)了模具設(shè)計(jì)工具，它將指導(dǎo)使用者根據(jù)不同客戶的要求，為其塑料零件選擇合適的模具。導(dǎo)言注塑成型工藝過(guò)程需要專業(yè)的知識(shí)、技能，最重要的是需要它成河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 2 -功的實(shí)踐經(jīng)驗(yàn)。通常是工藝參數(shù)控制過(guò)程的效率。在制造過(guò)程中，有效地控制和優(yōu)化這些參數(shù)能實(shí)現(xiàn)一致性，這種一致性會(huì)在零件質(zhì)量和零件成本上表現(xiàn)出來(lái)的問(wèn)題。1 智能化工程模塊注塑成型工藝（IKEM）基于知識(shí)的智能化工程模塊的注塑成型工藝（IKEM）是一種軟件技術(shù)，它領(lǐng)先于并行工程和 CAD / CAM 系統(tǒng)。它集成工程的設(shè)計(jì)和制造工藝的最新知識(shí)，給用戶各種設(shè)計(jì)方面的指示，通過(guò)減少在產(chǎn)品開發(fā)設(shè)計(jì)階段的工程變更，有助于減少一些工時(shí)。該系統(tǒng)將用于注塑設(shè)計(jì)，設(shè)計(jì)迭代和流程整合。目前的過(guò)程由許多手工計(jì)算、CAD 圖形結(jié)構(gòu)和從以前項(xiàng)目取得的經(jīng)驗(yàn)三部分組成。一旦工程師完成設(shè)計(jì)，這將是性能評(píng)估。該 IKEM 項(xiàng)目已分為三大模塊。 (1) 費(fèi)用估算模塊(2) 模具設(shè)計(jì)模塊(3) 生產(chǎn)模塊IKEM 系統(tǒng)有兩種形式輸入。在一個(gè) CAD 模型的形式（Pro/E 文件）下輸入，和在給出的用戶界面形式下輸入。圖 1-1 說(shuō)明了那種進(jìn)入每個(gè)模塊的輸入形式和用戶輸出形式。制造商的經(jīng)驗(yàn)水平將決定如何有效地控制工藝參數(shù)。有時(shí)這就導(dǎo)致人為錯(cuò)誤引起的不一致性。還有經(jīng)驗(yàn)不足，時(shí)間、資源短缺和創(chuàng)新的空間不大的情況。通過(guò)創(chuàng)造所謂的“智能模型”的問(wèn)題，工程學(xué)知識(shí)提供了一個(gè)可行的方案去解決所有這樣2 智能模具設(shè)計(jì)工具在它的基本形式中模具設(shè)計(jì)工具是一個(gè)從文本文件中提取輸入的Visual Basic 應(yīng)用程序，這種文本文件包含關(guān)于零件和用戶輸入程序。該文本文件包含來(lái)自 Pro/E 的一個(gè)信息文件的零件的幾何解析。輸入河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 3 -是用來(lái)估測(cè)模具得尺寸和其它各種特性。2.1 文獻(xiàn)回顧模具設(shè)計(jì)的是另一種注塑成型過(guò)程的階段，有經(jīng)驗(yàn)的工程師在很大程度上有助于自動(dòng)化進(jìn)程，提高其效率。這個(gè)問(wèn)題需要注意的是深入研究設(shè)計(jì)模具的時(shí)間。通常情況下，當(dāng)設(shè)計(jì)工程師設(shè)計(jì)模具時(shí)，他們會(huì)參閱表格和標(biāo)準(zhǔn)手冊(cè)，這會(huì)消耗大量的時(shí)間。另外，在標(biāo)準(zhǔn)的CAD 軟件中需要大量的時(shí)間去考慮模具的建模組件。不同的研究人員已經(jīng)解決了縮短用不同的方式來(lái)設(shè)計(jì)模具所花費(fèi)的時(shí)間的問(wèn)題。凱爾奇和詹姆斯采用成組技術(shù)來(lái)減少模具設(shè)計(jì)時(shí)間。聚合一類注塑成型件的獨(dú)特的編碼系統(tǒng)和在注射模具中所需的工具已開發(fā)，它可以適用于其它產(chǎn)品生產(chǎn)線。實(shí)施編碼系統(tǒng)的軟件系統(tǒng)也已經(jīng)被開發(fā)。通過(guò)獲取在這方面領(lǐng)域的工程師的經(jīng)驗(yàn)和知識(shí)，嘗試直接使模具設(shè)計(jì)過(guò)程的自動(dòng)化。并行模具設(shè)計(jì)系統(tǒng)的研究開發(fā)就是這樣的一個(gè)過(guò)程，在并行工程環(huán)境中試圖制定一個(gè)系統(tǒng)的注塑模具設(shè)計(jì)流程。他們的研究目標(biāo)是研制一個(gè)有利于并行工程實(shí)踐的模具開發(fā)的進(jìn)程，和研制開發(fā)一個(gè)以知識(shí)為基礎(chǔ)的為注塑模具設(shè)計(jì)提供工藝問(wèn)題和產(chǎn)品要求的輔助設(shè)計(jì)。通過(guò)各種方式獲取關(guān)于模具設(shè)計(jì)過(guò)程的確定信息和不確定信息，研究人員一直試圖使模具設(shè)計(jì)流程自動(dòng)化。這個(gè)研究試圖研制開發(fā)一個(gè)獨(dú)特的模具設(shè)計(jì)應(yīng)用程序，它一確定性和不確定性兩種形式獲取信息。 2.2 采用的方法為了發(fā)展智能模具設(shè)計(jì)工具，傳統(tǒng)的模具設(shè)計(jì)方法在被研究。應(yīng)用程序開發(fā)人員和設(shè)計(jì)工程師合作設(shè)計(jì)一種特定塑料零件的模具。在此期間，被工程師采納用來(lái)選擇模底座的方法正在被地密切關(guān)注和篩選過(guò)程的各個(gè)方面，需要他的知識(shí)經(jīng)驗(yàn)來(lái)確定。此外，有時(shí)候工程師將參考圖表和手冊(cè)以規(guī)范其甄選過(guò)程。這耗費(fèi)時(shí)間的過(guò)程，稍后也被記錄在應(yīng)用程序中。河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 4 -系統(tǒng)的闡述依據(jù)輸入和輸出的應(yīng)用程序是下一階段。這涉及到如何定義什么養(yǎng)的模具布局信息是用戶最需要的，也是他輸入最少卻得到相同的輸出。根據(jù)在模具設(shè)計(jì)工作中收集到的信息，由工程師遵循的公約被轉(zhuǎn)化為 if - then 規(guī)則。決策表是用來(lái)解釋各種可能出現(xiàn)的情況，它們是當(dāng)處理模具設(shè)計(jì)工程中某一特定的方面所提出的。這樣被制定規(guī)則，然后被組織在相互交融的模塊中，使用應(yīng)用程序開發(fā)環(huán)境。最后，應(yīng)用程序是檢驗(yàn)其正確性，當(dāng)涉及到為塑料零件設(shè)計(jì)模具在工業(yè)生產(chǎn)中。2.3 選擇合適的模架通常情況下，為制造塑料零件選擇適當(dāng)?shù)哪＜芩婕暗挠校海?）估計(jì)模腔數(shù) 模腔數(shù)量的決定取決于在一定時(shí)間內(nèi)所需部件的數(shù)量，像機(jī)器的塑化能力，廢品率等問(wèn)題也會(huì)影響到模架的模腔數(shù)量。（2）確定鑲塊及其尺寸 鑲塊有助于模架重用，因此有助于降低生產(chǎn)成本。當(dāng)涉及到尺寸和數(shù)量的選擇，作出決定取決于現(xiàn)有的鑲塊的重用性和新的鑲塊的成本。（3）確定澆道的尺寸和定位 澆道的尺寸取決于所成型的材料。盡管還有其它要考慮材料特性來(lái)決定它的澆道的尺寸供符合它的流量要求。轉(zhuǎn)輪的定位，取決于所用流道的拓?fù)洳季帧ｋm然循環(huán)的澆道系統(tǒng)始終是最好的，支道系統(tǒng)的平衡，避免流道均衡補(bǔ)償?shù)臉渲顫驳老到y(tǒng)是一個(gè)最被廣泛應(yīng)用的系統(tǒng)。（4）確定澆道直徑 澆道直徑?jīng)Q定于模具的尺寸，模腔的數(shù)量或在一定的時(shí)間內(nèi)用來(lái)填補(bǔ)的塑料的總數(shù)。 （5）澆口的定位 塑料在某一點(diǎn)進(jìn)入模腔，在這點(diǎn)可以均勻填充滿模腔。澆口可以設(shè)在循環(huán)模腔的任何周圍點(diǎn)，但當(dāng)填補(bǔ)矩形腔時(shí)，必須從中部流進(jìn)。（6）確定供水道的的尺寸和定位 供水道之間和從模具中的任何壁上以標(biāo)準(zhǔn)的距離定位。該公約不是用一個(gè)直徑范圍定位水道在模具河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 5 -壁上。 （7）根據(jù)以上結(jié)論確定模具的尺寸 根據(jù)以上的所有結(jié)論，模具的大概尺寸可以被估計(jì)，并四舍五入至最接近的產(chǎn)品目錄號(hào)。在模架以前，如果重新設(shè)計(jì)，考慮到以上所有方面會(huì)降低成本和減少設(shè)計(jì)時(shí)間，進(jìn)入重新設(shè)計(jì)。2.4 問(wèn)題的提出建立問(wèn)題，需要人的知識(shí)和經(jīng)驗(yàn)，模具設(shè)計(jì)方面消耗的時(shí)間涉及到圖表，數(shù)據(jù)表等，為開發(fā)應(yīng)用程序的問(wèn)題解釋所示。雖然大部分的輸入如模腔數(shù)、腔的圖像尺寸、周期時(shí)間，都是根據(jù)客戶要求，其他輸入如塑化能力、每分鐘注射量等，可從機(jī)器的說(shuō)明書中獲得。應(yīng)用程序的輸出包含模具尺寸和其他資料，這顯然有助于在目錄中選擇標(biāo)準(zhǔn)模架。除了輸入和輸出，圖 2 也顯示了產(chǎn)生的最終輸出的各種模塊。2.5 制定規(guī)則模具設(shè)計(jì)模塊的組織如果材料=“縮醛”和分流道長(zhǎng)度“0，那么分流道直徑 = 0.062 結(jié)束當(dāng)制定了規(guī)則，重要的是我們用一種嚴(yán)謹(jǐn)?shù)姆绞降膩?lái)表現(xiàn)這些信息，同時(shí)要避免重復(fù)、不完整和不一致的現(xiàn)象。決策表可以幫助處理上述問(wèn)題，它是通過(guò)對(duì)過(guò)于冗余和廣泛的問(wèn)題陳述的檢查實(shí)現(xiàn)的。比如說(shuō)，在選擇適當(dāng)?shù)哪＜艿倪^(guò)程中，模架尺寸取決于型腔和鑲件的數(shù)目。為確情況 A:模具寬度 =(鑲塊長(zhǎng)度 + 2)；模具長(zhǎng)度=(鑲塊長(zhǎng)度 + 2)；模具厚度=鑲塊厚度。情況 B:模具寬度=(2*鑲塊寬度 + 3.5)；模具長(zhǎng)度=(鑲塊長(zhǎng)度 + 2)；模具厚度=鑲塊厚度。情況 C: 模具寬度=(2*鑲塊寬度 + 3.5)；模具長(zhǎng)度=(鑲塊長(zhǎng)度 + 3)；模具河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 6 -厚度=鑲塊厚度。型腔的數(shù)目是一個(gè)，鑲件的數(shù)目也是一個(gè)的情況和型腔數(shù)目是兩個(gè)和四個(gè)的情況具有相同的模具尺寸，這三種情況可以歸結(jié)為一個(gè)單一的規(guī)則： 如果 鑲塊的數(shù)目= 1，則模具寬度 =（鑲塊寬度 + 2）模具長(zhǎng)度 =（鑲塊長(zhǎng)度+ 2）模具厚度 = 鑲塊厚度 結(jié)束為了方便和清楚起見，用一種標(biāo)準(zhǔn)的編程語(yǔ)言將這些規(guī)則模塊化。每個(gè)模塊生成一組輸出，這個(gè)輸出又將是對(duì)其他模塊的輸入。2.6 測(cè)試應(yīng)用通過(guò)使用各種測(cè)試案例對(duì)智能模具設(shè)計(jì)中的應(yīng)用程序進(jìn)行了驗(yàn)證。對(duì)于每一個(gè)案件的零件信息，模具和機(jī)器的信息資料種類繁多，人類專家證實(shí)了把這些信息輸入到應(yīng)用程序的結(jié)果。表 2 顯示了一個(gè)這樣的試驗(yàn)，需要兩個(gè)模腔，也沒(méi)有鑲件的存在。應(yīng)用程序提供近似的模具尺寸，執(zhí)行尺寸，澆口尺寸和亞軍的模腔長(zhǎng)度基于模腔圖尺寸和其他信息。通過(guò)使用各種測(cè)試案例對(duì)智能模具設(shè)計(jì)中的應(yīng)用程序進(jìn)行了驗(yàn)證。對(duì)于每一個(gè)案件的零件信息，模具和機(jī)器的信息資料種類繁多，人類專家證實(shí)了把這些信息輸入到應(yīng)用程序的結(jié)果。表 2 顯示了一個(gè)這樣的試驗(yàn)，需要兩個(gè)模腔，也沒(méi)有鑲件的存在。應(yīng)用程序提供近似的模具尺寸，執(zhí)行尺寸，澆口尺寸和亞軍的模腔長(zhǎng)度基于模腔圖尺寸和其他信息。獲得的模具尺寸非常接近人類專家的一個(gè)典型設(shè)計(jì)，但并沒(méi)有明確地說(shuō)明了一個(gè)模具標(biāo)準(zhǔn)件的用途，就像 D-M-E 模具目錄中的一種特定的模具。模具尺寸是基于所用材料而定的，因此它被限制在一定的范圍。河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 7 -3 總結(jié)本文介紹了在發(fā)展智能模具設(shè)計(jì)應(yīng)用中所采用的方法，這種應(yīng)用是根據(jù)用戶輸入進(jìn)行模架選擇的。獲取知識(shí)的過(guò)程首先是通過(guò)與業(yè)內(nèi)專家密切協(xié)商設(shè)計(jì)一種模架，也通過(guò)從舊書和數(shù)據(jù)表中收集確定性信息。收集到的資料，表示了在不同的模塊中規(guī)則的排列形式。這些資料可定性和定量地對(duì)模具進(jìn)行選擇。決策表是用來(lái)減少規(guī)則庫(kù)的規(guī)模，使規(guī)則庫(kù)中的問(wèn)題域全面。在不同的模塊中使用這些規(guī)則來(lái)開發(fā)應(yīng)用程序，當(dāng)談到在給業(yè)內(nèi)生產(chǎn)的塑件選擇適當(dāng)?shù)哪＜軙r(shí)就為應(yīng)用程序的有效性作測(cè)試。參考文獻(xiàn)[1] 錢伯斯 T. L.帕金森 A. R. “知識(shí)代表及專家系統(tǒng)的混合轉(zhuǎn)換。 ” 美國(guó)機(jī)械工程師學(xué)會(huì)，1998，120:468-474.[2] 凱爾其·詹姆斯 R.“軟件升壓模具設(shè)計(jì)效率”的成型系統(tǒng)，1999， 3:16-23. [3] 李榮顯，陳育民，鄒昶,“開發(fā)一個(gè)并行模具設(shè)計(jì)系統(tǒng)：以知識(shí)為基礎(chǔ)的辦法” ，計(jì)算機(jī)集成制造系統(tǒng)，1997，4:287-307. [4] 斯特德曼薩利 ·佩爾 M,“在工程設(shè)計(jì)專家系統(tǒng)：一種注塑成型的塑料件的應(yīng)用”智能制造，發(fā)動(dòng)機(jī) 1995，2:347-353.[5] 費(fèi)爾南德斯 A，卡斯塔尼 J,賽爾 F, “CAD / CAE 信息的模具和熱塑性塑料注射原型設(shè)計(jì)的”信息技術(shù) 1997:117-124. [6] 道格拉斯 M 布萊斯,“塑料注射成型，材料選擇和產(chǎn)品設(shè)計(jì)”1997:1-48.[7] 道格拉斯 M 布萊斯， “塑料注射成型模具設(shè)計(jì)基礎(chǔ)” ，1997，2:1-120. 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 8 -Session VA4Intelligent Mold Design Tool For Plastic Injection MoldingJagannath Yammada, Terrence L. Chambers, Suren N. DwivediDepartment of Mechanical EngineeringUniversity of Louisiana at LafayetteAbstractPlastic Injection molding is one of the most popular manufacturing processes for making thermoplastic products, and mold design is a key aspect of the process. Design of molds requires knowledge, expertise and most importantly experience in the field. When one of these is lacking, selection of an appropriate mold for manufacturing a plastic component is done on a trial-and-error basis. This increases the cost of production and introduces inconsistencies in the design.This paper describes the development of an intelligent mold design tool. The tool captures knowledge about the mold design process and represents the knowledge in logical fashion. The knowledge acquired will be 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 9 -deterministic and non-deterministic information about the mold design process. Once developed the mold design tool will guide the user in selecting an appropriate mold for his plastic part based on various client specifications.IntroductionThe plastic injection molding process demands knowledge, expertise and, most important, experience for its successful implementation. Often it is the molding parameters that control the efficiency of the process. Effectively controlling and optimizing these parameters during the manufacturing process can achieve consistency, which takes the form of part quality and part cost.The level of experience of the manufacturer(s) determines how effectively the process parameters are controlled. This sometimes leads to inconsistency introduced by human error. There is also the case where there is inexperience, shortage of time, resources and little scope for innovation. Knowledge-based engineering provides a feasible solution to all these problems by creating what is called an “intelligent model” of the problem.1 IKEMIntelligent Knowledge based Engineering modules for the plastic injection molding process (IKEM) is a software technology that is a step 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 10 -ahead of the concurrent engineering and CAD/CAM systems. It integrates current knowledge about the design and manufacturing processes and helps to reduce several man-hours by reducing engineering changes in the design phase of product development by giving users instruction about various design aspects. The system will be used for injection molding design, design iterations, and process integration. The current process consists of many manual computations, CAD graphical constructions, and experience attained from previous projects. Once the engineer completes the design, it will be evaluated for performance. The IKEM project has been divided into three major modules.1. The cost estimation module2. The mold design module3. The Manufacturing moduleInput to the IKEM system is of two forms. Input in the form of a CAD model (Pro-E file) and input given at the User Interface form. Figure 1 illustrates the kind of input that goes into each module and the output given to the user. Figure 1. Organization of the IKEM Project2 Intelligent Mold Design ToolThe mold design tool in its basic form is a Visual Basic application taking input from a text file that contains information about the part and a User Input form. The text file contains information about the part geometry parsed from a Pro/E information file. The input is used to estimate the dimensions of mold and various other features.2.1 Literature Review Design of molds is another stage of the injection molding process where the experience of an engineer largely helps automate the process and increase its efficiency. The issue that needs attention is the time that goes 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 11 -into designing the molds. Often, design engineers refer to tables and standard handbooks while designing a mold, which consumes lot of time. Also, a great deal of time goes into modeling components of the mold in standard CAD software. Different researchers have dealt with the issue of reducing the time it takes to design the mold in different ways. Koelsch and James have employed group technology techniques to reduce the mold design time. A unique coding system that groups a class of injection molded parts, and the tooling required ininjection molding is developed which is general and can be applied to other product lines. A software system to implement the coding system has also been developed. Attempts were also directed towards the automation of the mold design process by capturing experience and knowledge of engineers in the field. The development of a concurrent mold design system is one such approach that attempts to develop a systematic methodology for injection mold design processes in a concurrent engineering environment. The objective of their research was to develop a mold development process that facilitates concurrent engineering-based practice, and to develop a knowledge-based design aid for injection molding mold design that accommodates manufacturability concerns, as well as product requirements.Researchers have been trying to automate the mold design process either by capturing only the deterministic information on the mold design process or the non-deterministic information, in various ways. This research uniquely attempts to develop a mold design application that captures information in both forms; deterministic and non-deterministic.2.2 Approach AdoptedIn order to develop an intelligent mold design tool, the conventional method of designing molds is studied. The application developer and the design engineer work together in designing a mold for a particular plastic part. During this time, the approach adopted by the engineer to select the 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 12 -mold base is closely observed and aspects of the selection process that require his knowledge/experience are identified. Also, there will be times when the engineer will refer to tables and handbooks in order to standardize his selection process. This time consuming process is also recorded to incorporate it later in the application.Formulating the problem for the application in terms of inputs and outputs is the next stage. This involves defining what information about the mold layout is most required for the user and also the minimum number of inputs that can be taken from him to give those outputs.Based on the information gathered in the mold design exercise, the conventions followed by the engineer are transformed into if-then rules. Decision tables are used to account for all possible cases that arise when dealing with a particular aspect of the mold design process. The rules so framed are then organized into modules interacting with each other, using an application development environment. Finally the application is tested for its validity when it comes to designing molds for plastic parts manufactured in the industry.2.3 Selection of Appropriate Mold Base Typically, selection of appropriate mold base for manufacturing a plastic part involvesEstimating the number of cavities The number of cavities is decided depending on the number of parts required within a given time. There are also other issues like the plasticizing capacity of the machine, reject rate etc that affect the number of cavities to be present in the mold base.Deciding on the presence of inserts and their dimensionsInserts facilitate the reusability of the mold base and therefore help in reducing cost of manufacturing. When it comes to selecting the dimensions and the number, a decision is made depending on the reusability of existing 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 13 -old inserts and cost of ordering new ones.Determining the size and location of runnersThe runner size depends on the material being molded. Although there are other considerations material properties determines the channel size required for its flow. Location of runners mainly depends on the topology of runners being used. Though a circular runner system is always preferable, the branched runner system that avoids runner balancing is the one most widely used.Determining the diameter of sprueThe diameter of the sprue is decided based on the size of the mold, number of cavities, or the amount of plastic that is to be filled within a given time.Locating gatesPlastic enters the cavity at a point where it can uniformly fill the cavity. A gate can be located at any point on the perimeter of a circular cavity but has to enter at the midsection when it comes to filling rectangular cavities.Determining the size and location of water lines Water lines are located at standard distances form each other and from any wall in the mold. The convention is not to locate a waterline within one diameter range on the mold wall.Deciding mold dimensions based on above conclusions河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 14 -Based on all the above decisions the approximate mold dimensions can be estimated and rounded off to the nearest catalog number. Considering all the above aspects before even modeling the mold base reduces the cost and time that go into redesigning.2.4 Formulation of the ProblemBased on issues that require human knowledge/experience, and aspects of mold designthat consume time referring to tables, data sheets etc., the problem for developing theWhile most of the input, like the number of cavities, cavity image dimensions, cycle time are based on the client specifications, other input like the plasticizing capacity, shots per minute etc., can be obtained from the machine specifications. The output of the application contains mold dimensions and other information, which clearly helps in selecting the standard mold base from catalogs. Apart from the input and output, the Figure 2 also shows the various modules that produce the final output.2.5 Framing rules At this stage, the expert’s knowledge is represented in the form of multiple If-Then statements. The rules may be representations of both qualitative and quantitative knowledge. By qualitative knowledge, we mean 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 15 -deterministic information about a problem that can be solved computationally. By qualitative we mean information that is not deterministic, but merely followed as a rule based on previous cases where the rule has worked. A typical rule is illustrated below: If Material = “Acetal” AndRunner Length 0 ThenRunner Diameter =0.062End IfWhen framing the rules it is important that we represent the information in a compact way while avoiding redundancy, incompleteness and inconsistency. Decision tables help take care of all the above concerns by checking for redundancy and comprehensive expression of the problem statement. As an example, in the process of selecting an appropriate mold base, the size of mold base depends on the number of cavities and inserts. To ensure that all possible combinations of cavities and inserts have been considered we use a decision table and subsequently use the decision table to frame rules. Figure 3. Mold Dimensions for various combinations of Inserts and CavitiesThe case where the number of cavities is one and the number of inserts is one has the same mold dimensions as the case where the number of cavities is two and four. The three cases can be reduced to one single rule:If Number Of Inserts=1 ThenMold Width = (Insert Width + 2)Mold Length = (Insert Length + 2)Mold Thickness = Insert ThicknessEnd IfThe rules are arranged in modular fashion using a standard programming language for the sake of convenience and clarity. Each module generates a set of outputs, which would be inputs for other modules.2.6 Testing the applicationThe intelligent mold design application is validated using various test 河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 16 -cases. For each case the part information, mold information and the machine information are varied and a human expert validates the results of feeding this info into the application. Table 2 shows one such test case where the part requires two cavities and there are no inserts present.The application gives the approximate mold dimensions, runner dimension, sprue dimension and runner length based on the cavity image dimensions and other information.Table 2. Typical test case showing program input and output.The mold dimensions obtained are very close to a typical human expert design for the test case but do not suggest explicitly the use of a standard mold base, like a specific mold from the D-M-E mold base catalog. The mold dimensions are however useful in selecting appropriate mold base from the mold catalogs. The runner dimensions are based on the material being used and therefore are limited to a specific range of shot size.3 SummaryThis paper presents the approach adopted towards developing an intelligent mold design application that performs mold base selection based on user input. The knowledge acquisition process is done by first designing a mold base in close consultation with an industry expert and also by collecting deterministic information from hand books and data sheets. The collected information, which can be both qualitative and quantitative knowledge about the mold selection process, is represented in the form of rules arranged in different modules. Decision tables are used to reduce the size of rule base and make the rule base comprehensive in the problem domain. The application developed using the rules in different modules is then tested for its validity when it comes to selecting appropriate mold bases for plastic parts manufactured in the industryReferences河南理工大學(xué)萬(wàn)方科技學(xué)院本科畢業(yè)設(shè)計(jì)- 17 -1. Chambers T. L., Parkinson A. R., 1998, “Knowledge Representation and Conversion of Hybrid Expert Systems.” Transactions of the ASME, v 120,pp 468-474 2. Koelsch, James R., 1999, “Software boosts mold design efficiency“ Molding Systems, v 57, n 3,p 16-23. 3. Lee, Rong-Shean, Chen, Yuh-Min, Lee, Chang-Zou,1997 “Development of a concurrent mold design system: A knowledge-based approach”, Computer Integrated Manufacturing Systems, v 10,n 4, p 287-307 4. Steadman Sally, Pell Kynric M, 1995, “ Expert systems in engineering design: An application forinjection molding of plastic parts“ Journal of Intelligent Manufacturing, v6, p 347-353.5. Fernandez A., Castany J., Serraller F., Javierre C., 1997, “CAD/CAE assistant for the design of molds and prototypes for injection of thermoplastics “Information Technological, v 8, p 117-124.6. Douglas M Bryce, 1997, “Plastic injection molding -Material selection and product design”, v 2, pp 1-48.7. Douglas M Bryce, 1997, “Plastic injection molding-Mold design fundamentals”, v2, pp 1-120.