機械類外文翻譯---自動車床系統(tǒng)設計的研究

1、<p>　　自動車床系統(tǒng)設計的研究</p><p>　　摘要：車床數(shù)字化設計系統(tǒng)的開發(fā)，是通過工程知識相關的技術和產(chǎn)品的參數(shù)化設計技術之間的融合發(fā)展起來的。設計系統(tǒng)的框架在細節(jié)上表現(xiàn)為主控模塊，初始化模塊，三維和二維的更新管理模塊，知識管理模塊和知識庫。本文對系統(tǒng)的關鍵技術設計比如知識挖掘與分類、自動變換結構設計、建立數(shù)據(jù)庫、自動設計變結構與車床設計導航進行了進一步研究。最后，本文對自動設計系統(tǒng)的界面和

2、設計過程進行了詳細的介紹。</p><p>　　關鍵詞：車床；知識重組；自動設計系統(tǒng)</p><p><b>　　一、導言</b></p><p>　　新一代的數(shù)碼解決方案使技術創(chuàng)新能在不同的環(huán)境下進行，并且減少整個產(chǎn)品的開發(fā)周期、節(jié)省時間和提高產(chǎn)品質(zhì)量。目前，在消費產(chǎn)品的研發(fā)中采用不同的方法進行快速設計并取得了成果，然而對于商業(yè)產(chǎn)品卻很少受到

3、注意，特別是機床，目前尚未對其進行研究。我們的第二個目的是衡量不同的人群在同一個測試中表現(xiàn)出來的不同的敏感性。該階段的研究是很詳盡的，包括：對描述的選擇，對圖像的選擇和對所參與人群的選擇。結果表明，這些技術也適用于機床的設計，不同的群體對機器的認識也是不同的，在某些方面的差異，并不僅限于用戶與專家在相關的行業(yè)中這種差異依然存在，在相關行業(yè)中通常是根據(jù)技術規(guī)格來決定機器的購買或使用。這些技術規(guī)范是可以測量的，但機器一些重要的參數(shù)，如便捷性

4、，安全性，耐用性等卻不那么容易衡量和比較。本文表明，語義差別的方法是用來衡量這些性能的一種可能途徑。</p><p>　　車床的設計是復雜的過程，在設計的過程中很多知識都是必要的。在生產(chǎn)的過程當中，不同的知識在不同產(chǎn)品不同開發(fā)階段中起不同的作用。許多新產(chǎn)品是在原有產(chǎn)品的基礎上設計發(fā)展起來的。車床設計行業(yè)的設計成果中，幾乎60％是由組件設計的重組制作而成的，30％是根據(jù)過去的產(chǎn)品修訂的。在本文中，我們提出了一個車床

5、設計技術，采用NX平臺，建立了車床設計的專門知識庫，并融入新產(chǎn)品的設計理念和成熟的設計經(jīng)驗，運用先進的開發(fā)工具，如UG/ Open API，UIStyler，MenuScript 。</p><p>　　這種新方法可以有效地提供設計質(zhì)量保證，縮短產(chǎn)品設計周期，提高設計效率和車床水平，并且對汽車玻璃行業(yè)有重大影響力。</p><p>　　車床設計系統(tǒng)應用所學到的知識來完成車床快速設計。設計系

6、統(tǒng)過程中所需要應用的知識包括成熟的車床產(chǎn)品，車床設計標準的有關技術數(shù)據(jù)，專業(yè)知識，豐富的設計經(jīng)驗和最新的研究成果。在產(chǎn)品設計周期中，嵌入式系統(tǒng)需要設計論證和設計指導機制。而且設計師只需要輸入?yún)?shù)并由人機界面通過索引的幫助下符合客戶要求即可。這些參數(shù)包括車床厚度，鉗板的長度等，推理機將與每一個規(guī)則所規(guī)定的條件進行比較，如果它們匹配，則該規(guī)則的結論將作為一個新的事實被添加到設計知識庫。根據(jù)發(fā)展過程中的需求，該系統(tǒng)分為以下功能模塊。系統(tǒng)開發(fā)的

7、目的是通過知識重組來實現(xiàn)自動車床的設計，提高產(chǎn)品的設計質(zhì)量和效率。以檢查軌距為導向的車床設計系統(tǒng)的設計過程可分為：需求分析階段，初步設計階段，詳細設計階段和功能的改進、測試階段。在需求分析階段，經(jīng)過仔細分析和對話，系統(tǒng)能夠根據(jù)用戶設置的鉗板曲線設計車床形狀，實現(xiàn)自動車床重點部分的設計，自動創(chuàng)建一套車床。</p><p>　　在選擇開發(fā)工具方面，我們考慮了用戶的要求以及對一些工具的熟練程度，最終決定采用UG NX5

8、.0和Visual Studio 2005作為開發(fā)平臺。在詳細設計階段，我們先在UG互動的環(huán)境一步一步進行設計，然后由應用程序最終完成。該方案的主體部分僅僅是在前文提到的上述階段實現(xiàn)的，但是在應用過程中一些細節(jié)沒有完全考慮進去，在提高產(chǎn)品性能的階段，我們做了詳細的測試，一些錯誤已經(jīng)得到糾正。這有助于使系統(tǒng)更加穩(wěn)定和完善。</p><p><b>　　A.設計知識表示</b></p&g

9、t;<p>　　有效的知識表示方法是進行知識融合和知識重組的關鍵。目前，常見的知識表示方法，包括動詞的邏輯表示，產(chǎn)生式規(guī)則，幀表示，面向對象的表示等，本文中生成型車床的設計規(guī)則說明如下：</p><p>　　<Rule>∷=（IF{<Condition>} THEN {<Result>}）</p><p>　　< Condition

10、>∷=(<Expression>∣<Fact>)</p><p>　　< Expression >∷=({<Variant><Verb><Attribute>})</p><p>　　< Fact >∷=({<Fact item>})</p><p>　　<

11、Verb >∷=(< > = + - * / sin、cos、tg、log …)</p><p>　　< Attribute value>∷=({< Variant >∣<Value>})</p><p>　　<Conclusion>∷=({<Presentation>∣<Operation>})&l

12、t;/p><p>　　< Presentation >∷=({< Presentation item>})</p><p>　　<Operation>∷=({<Operating function>})</p><p>　　例如，在嵌入塊成型，兩平面之間的關系（平行或交叉）（P1和P2）的車床設計是由法線向量（V1和V2）

13、來判斷。在這個例子中，三個規(guī)則可以從規(guī)則庫選取，有如下幾點：</p><p>　　規(guī)則1：如果V1×V2=0而且P1的矢量方向和P2是相同的，</p><p>　　則P1∥P2和P1和P2具有相同的向量方向。</p><p>　　規(guī)則2：如果V1×V2=0和P1的矢量方向和P2是相反的，</p><p>　　則P1∥P2和

14、P1和P2有相反的向量方向。</p><p>　　規(guī)則3：如果V1×V2≠0，P1的矢量方向和P2是不同的，</p><p>　　則P1和 P2的彼此相交。</p><p>　　車床設計的框架表示。幀，是另外一種知識表示形式的數(shù)據(jù)結構，每幀作為一個知識單元，由馬文?明斯基在1975年提出。詳細的形式在下面的例子中說明：</p><p&g

15、t;<b>　　<Frame></b></p><p><b>　　……</b></p><p>　　Slot name I：Flank name i1 (Value i11, Value i12, …)</p><p>　　Flank name i2 (Value i21, Value i22, …)<

16、/p><p>　　Slot name J：< Frame name></p><p><b>　　…</b></p><p>　　Slot name K：<Additive course></p><p>　　車床設計案例檢索。最創(chuàng)新的產(chǎn)品設計往往都基于以前的經(jīng)驗和理論，因此，新產(chǎn)品許多地方繼承了原有

17、產(chǎn)品的設計。由于大多數(shù)產(chǎn)品的設計不是原來的，但以前的車床設計規(guī)則或流程仍可能會有所幫助。在本文中，我們提出了使用基于案例的推理方法來實現(xiàn)產(chǎn)品的車床設計的理念。</p><p>　　我們采用動態(tài)存儲模型來表示系統(tǒng)中的案例。這意味著各種案件根據(jù)它們的一般特征來組織，并根據(jù)他們之間的分歧指數(shù)來區(qū)分它們。案例庫通過起關鍵作用的關系數(shù)據(jù)庫技術的產(chǎn)品ID建立起來。 SQL可作為初步檢索句子。例如，如果我們要查詢案例，可用下面

18、的偽代碼。</p><p>　　SELECT case number FROM case base</p><p>　　WHERE product ID =‘input value’</p><p>　　為了作進一步的分析，車床設計要求能夠對其進行修改，以便適應自動探索設計空間。如果修改只影響某些參數(shù)的設置值而不觸及問題的結構，我們就說它是一個參數(shù)化設計。對于過程流

19、相關的任務級應用，進程中沒有結構性的變化，即，結構設計是被限制的。對于序列中不同步驟的不同流程之間的比較，被映射到的一個給定步驟的影響在某些參數(shù)值的調(diào)節(jié)方面存在差異。例如，如果是兩個不同的進程選項之間的比較，對其中一個植入額外的步驟，另一沒有此步驟，那么在植入的步驟參數(shù)值設置非常低的情況下就能很方便地建模，而在實際應用中完全沒有這一步。</p><p>　　為了實現(xiàn)一般的參數(shù)化設計的產(chǎn)品，我們首先需要對產(chǎn)品進行分

20、析，并確定組件之間的相互關系，包括：車床設計的關系、幾何關系和參數(shù)的關系。下一步，在數(shù)據(jù)，幀，條件語句，以及先進的程序語言的形成過程中，需要對相關工程知識進行收集、整理、一般化處理、提取和存儲。然后，需要對所有對產(chǎn)品性能有影響的工程參數(shù)加以界定，這可以用來采用模糊分析的方法通過工程參數(shù)對訂單要求的其他類型的自動車床進行檢驗。圖1顯示了車床的不同知識表達方式。</p><p>　　基本知識：基本車床知識包括：1）客

21、戶對車床的使用要求;2）對客戶潛在需求的預測;3）相關的部件和組件設計的國際國內(nèi)標準;4）以往的設計案例;5）在特殊的環(huán)境下對產(chǎn)品性能的要求;6）對不同的部件和組件的設計要求;7）車床不同零部件之間的設計聯(lián)系和幾何位置關系;8）不同材質(zhì)和規(guī)格零件的特性。通過對自動車床基本知識的分析和研究，我們已經(jīng)制定出總體結構的主要控制參數(shù)和大體的設計規(guī)則以及產(chǎn)品設計知識庫。</p><p>　　產(chǎn)品結構知識：該系統(tǒng)采用自上而下

22、的設計方法來開發(fā)產(chǎn)品型號。在建立產(chǎn)品模型之前就應該確定出產(chǎn)品的結構，因此應該對影響產(chǎn)品結構的知識進行確認、分析。產(chǎn)品結構的知識，主要包括尺寸，形狀，車床的設計尺寸和不同零部件之間的位置關系。據(jù)NX平臺的特點，產(chǎn)品的整體框架知識包括基準面產(chǎn)品結構知識，車床零部件之間的設計限制，以及基于功能的組件的布局等。組成車床檢驗設計的控制結構建立在NX平臺上的產(chǎn)品草圖拓撲之上。</p><p>　　產(chǎn)品設計過程的知識：產(chǎn)品設計

23、過程的知識主要包括產(chǎn)品設計流程、鑒定原則等，反映了基本知識和產(chǎn)品結構知識之間的相互依存關系。在基本知識和產(chǎn)品結構知識間聯(lián)系的基礎上，采用產(chǎn)品開發(fā)工具UFun來實現(xiàn)產(chǎn)品設計的智能導航和自動可變產(chǎn)品結構設計。</p><p>　　B.可變結構的自動設計</p><p>　　作為一個典型的串行轉化產(chǎn)品，自動車床的結構形式隨用戶需求的變化而變化。本文采用一些先進的技術解決可變結構問題，如維度的驅動

24、技術，限制驅動技術，幾何對象驅動技術和數(shù)據(jù)制約技術。</p><p>　　畢業(yè)設計（論文）外文翻譯原文</p><p>　　Research on Lathe Automatic Design System</p><p>　　Abstract—A lathe digital design system was developed through the integ

25、ration between knowledge-engineering-related technology and product parametric design technology. The frame of the design system was expressed in detail such as main control module, initialization module, update manageme

26、nt module of 3D and 2D, knowledge management module and knowledge library. The key technologies of the system design such as Knowledge mining and classification, automatic design of transform structures, establ</p>

27、<p>　　Keywords-lathe; knowledge reuse; automatic design system</p><p>　　I. INTRODUCTION</p><p>　　The new age of digital solutions empowers innovation through different environments and tr

28、ansforms the entire product cycle development to reduce waste and lead time, all involved with the aiming for quality. Nowadays, different approaches to rapid design are being applied to consumer products with successful

29、 results, but commercial products have generally received less attention and machine tools in particular have not yet been studied. [1-3] Our second objective is to measure the different sen</p><p>　　Lathe d

30、esign is complicated procession in which lots of knowledge is necessary. The knowledge plays different parts on the different steps of product development. Many new products are developed on the base of existed design kn

31、owledge and effects. In lathe design industry, almost 60% of the design works are based on the reuse of existed component design and 30% of them are based on the revision of past parts [4-8]. In this paper we put forward

32、 an lathe design technique that adopts NX platform, e</p><p>　　This new way can effectively provide design quality supports, shorten product design period, improve the design efficiency and level of lathe, a

33、nd have a significant influence on auto glass industry.</p><p>　　II. PREPARE YOUR PAPER BEFORE STYLING</p><p>　　The lathe design system applies acquired knowledge to complete rapid design of lat

34、he. The knowledge that is applied in the design system includes mature lathe products, relevant engineering data of lathe design standard, expertise, accumulated design experience and latest research effects. During the

35、product design period, the system is embedded with design reasoning and design course guide mechanism. And designers only need to input parameters demanded by the customers through human-computer in</p><p>　

36、　In terms of choosing development tools, we have considered user’s requirements and proficiency degree toward some tools, and have decided to adopt UG NX5.0 and Visual Studio 2005 as the development platform to develop t

37、he system at last. During the detailed design phase, we get it done in the UG interactive environment step by step first, and then by applying program. The main body of the program is only achieved during the phases ment

38、ioned above and some details in applying course have not been</p><p>　　A. Design knowledge representation</p><p>　　Effective knowledge representation method is the key in implementing knowledge

39、fusion and knowledge reuse. At present, common knowledge representation methods include verb logic representation, generating type rules, frame representation, object-oriented representation, etc. In this paper generatin

40、g type lathe design rules are described as follows:</p><p>　　<Rule>∷=（IF{<Condition>} THEN {<Result>}）</p><p>　　< Condition >∷=(<Expression>∣<Fact>)</p>

41、<p>　　< Expression >∷=({<Variant><Verb><Attribute>})</p><p>　　< Fact >∷=({<Fact item>})</p><p>　　< Verb >∷=(< > = + - * / sin、cos、tg、log …)

42、</p><p>　　< Attribute value>∷=({< Variant >∣<Value>})</p><p>　　<Conclusion>∷=({<Presentation>∣<Operation>})</p><p>　　< Presentation >∷=({&l

43、t; Presentation item>})</p><p>　　<Operation>∷=({<Operating function>})</p><p>　　For example, in the lathe design of embedding block and molding, the relationship (parallel or inte

44、rsecting) of the two planes (P1 and P2) are judged by the normal vectors (V1 and V2). In this example, three rules can be procured from the rules library and they are as follows:</p><p>　　Rule1: IF V1×V

45、2=0 AND the vector directions of P1 and</p><p>　　P2 are the same.</p><p>　　THEN P1 ∥ P2 AND P1 and P2 have the same vector</p><p>　　directions.</p><p>　　Rule2: IF V1

46、15;V2=0 AND the vector directions of P1 and</p><p>　　P2 are opposite.</p><p>　　THEN P1 ∥ P2 AND P1 and P2 have the opposite vector</p><p>　　directions.</p><p>　　Rule3:

47、IF V1×V2≠0 AND the vector directions of P1 and</p><p>　　P2 are different.</p><p>　　THEN P1and P2 intersects with each other.</p><p>　　Lathe design Frame Representation. Frame,

48、put forward by Marvin Minsky in 1975, is another data structure of knowledge representation. Every frame is looked at as a knowledge unit. The detailed form is shown in the following example:</p><p><b>

49、;　　<Frame></b></p><p><b>　　……</b></p><p>　　Slot name I：Flank name i1 (Value i11, Value i12, …)</p><p>　　Flank name i2 (Value i21, Value i22, …)</p>&l

50、t;p>　　Slot name J：< Frame name></p><p><b>　　…</b></p><p>　　Slot name K：<Additive course></p><p>　　Lathe design Cases Retrieval. Most innovative products a

51、re designed based on previous experience and theory, thus, many parts of the new product design are inherited from the old product design. Since most design of products are not original, previous lathe design rules or pr

52、ocesses may be helpful. In this paper we present using a case-based reasoning method to implement product lathe design.</p><p>　　We adopt dynamic storage models to represent cases in the system. This means a

53、ll kinds of cases are organized based on their general characteristics and distinguished according to the index of their differences. The case library is built through a relation database technique where the product ID a

54、cts as the key. There SQL is available as the preliminary retrieval sentence. For instance, if we want to inquire a case, the following pseudo code is available.</p><p>　　SELECT case number FROM case base<

55、;/p><p>　　WHERE product ID =‘input value’</p><p>　　For further analysis, a design has to be able to be modified in order to be subjected to automatic exploration of the design space. If the modific

56、ation affects only the value of certain parameter settings while leaving the structure of the problem untouched, we speak of a parameterized design [9-12]. Forprocess-flow related task-level applications, no structural c

57、hanges in the process flow, i.e., the structure of the design, are admitted. In the case where comparisons between process flows with </p><p>　　To realize the general parameterized design of the product, we

58、first need to analyze the product, and determine the interrelation of the components, including: lathe design relationship, geometric relationship, and parametric relationship. Next, the relevant engineering knowledge ne

59、ed to be collected, arranged, generalized, abstracted and stored in the forms of data, frame, conditional sentence, and advanced program languages. [15-19]. Then, all influential engineering parameters of the product <

60、;/p><p>　　??Basic knowledge: The Basic knowledge of lathe includes: 1) Customer’s personal using requirement to lathe; 2) Customer’s potential demands prediction; 3) International and domestic standards for rel

61、ated parts and component design; 4) Previous design case; 5) Products’ performance requisition under a special environment; 6) Design requisition of different parts and component; 7) The lathe design relationship and geo

62、metric position among different parts and components; 8) The characters of parts</p><p>　　??Product structure knowledge: This system adopts topdown- based design method to develop product model. The product

63、structure should be determined before the establishment of product model, so the knowledge that influences product structure should be analyzed and determined. Product structure knowledge mainly includes out-shape dimens

64、ions, lathe design dimensions and position relationships between different parts and components. According to the characters of NX platform, the whole framework kn</p><p>　　Knowledge of product design proces

65、s: The knowledge of product design process mainly includes product design flow, appraisement principles, etc. and reflects the interdependence relationship between basic knowledge and product structure knowledge. On the

66、base of the interlinkage of basic knowledge and product structure knowledge, product developing tool UFun is used to achieve product design intelligent navigation and automatic design of variable product structure.</p

67、><p>　　B. Automatic design of variable structure</p><p>　　As a typical serial transforming product, the structure form of auto lathe changes with requirement change of users. In this paper some adv

68、anced techniques are adopted to tackle the variable structure issues, such as the dimension-driving technique, restriction-driving technique, geometric objects driving technique and data restraining technique.</p>