外文翻譯---基于事例推理的夾具設(shè)計(jì)研究與應(yīng)用

1、<p>　　華中科技大學(xué)文華學(xué)院</p><p>　　畢業(yè)設(shè)計(jì)（論文）外文文獻(xiàn)翻譯</p><p>　?。?012年12月15日 至2013 年 5月20日） </p><p>　　畢業(yè)設(shè)計(jì)（論文）題目：十字頭的加工工藝及夾具設(shè)計(jì)</p><p>　　學(xué)部（系）：機(jī)械與電氣工程學(xué)部 專業(yè)班級(jí)：機(jī)械設(shè)計(jì)制造及其自動(dòng)化09</p

2、><p>　　學(xué)生姓名： 黃柏程 指 導(dǎo) 教 師： 趙鎮(zhèn)鋒 </p><p><b>　　外文文獻(xiàn)翻譯</b></p><p>　　基于事例推理的夾具設(shè)計(jì)研究與應(yīng)用</p><p>　　摘要:根據(jù)基于事例的設(shè)計(jì)方法,提出采用工序件的特征信息和夾具的結(jié)構(gòu)特征信息來描述夾具的相似性,并

3、建立了包括這2方面主要特征信息為基礎(chǔ)的事例索引碼,設(shè)計(jì)了事例庫的結(jié)構(gòu)形式,創(chuàng)建了層次化的事例組織方式;同時(shí),提出了基于知識(shí)引導(dǎo)的夾具事例檢索算法,以及事例的修改和采用同族事例碼進(jìn)行相似事例的存貯,形成了基于事例推理的夾具設(shè)計(jì).所開發(fā)的原型系統(tǒng)在型號(hào)工程夾具設(shè)計(jì)等項(xiàng)目的設(shè)計(jì)過程中得到了應(yīng)用,并取得了令人滿意的使用效果.</p><p>　　關(guān)鍵詞: 基于事例的推理 夾具設(shè)計(jì) CAD</p><

4、;p>　　夾具是以確定工件安全定位準(zhǔn)確為目的的裝置，并在加工過程中保持工件與刀具或機(jī)床的位置一致不變。因?yàn)閵A具的結(jié)構(gòu)依賴于產(chǎn)品的特點(diǎn)和在企業(yè)規(guī)劃中加工工序的地位,所以它的設(shè)計(jì)是制造過程中的瓶頸,制約著效率的提高. 夾具設(shè)計(jì)是一個(gè)復(fù)雜的過程,需要有從大量的設(shè)計(jì)論文中了解質(zhì)量知識(shí)的經(jīng)驗(yàn)，這些設(shè)計(jì)論文包括工件的結(jié)構(gòu)設(shè)計(jì)、涉及加工工藝，和加工環(huán)境。當(dāng)用這些擅長繪制詳細(xì)設(shè)計(jì)圖的傳統(tǒng)的CAD工具（如Unigraphics、CATIA、Pro/

5、E)時(shí),這仍然是一項(xiàng)非常耗時(shí)的工作，但是利用以往的設(shè)計(jì)經(jīng)驗(yàn)和資源也不能提供一些益處,而這正是提高效率的關(guān)鍵因素. 基于事例推理 (CBR) 的方法適應(yīng)以往個(gè)案解決的辦法,建立一個(gè)新問題的方法,主要有以下四步驟:檢索、利用、修改,并保留.這是一個(gè)比用專業(yè)系統(tǒng)模仿人類思維有用的使用方法，因?yàn)樘岢鲆粋€(gè)類似的情況,和采用一些修改,似乎不言自明,而且比人類更直觀.所以支持不同事例的設(shè)計(jì)工具已經(jīng)在諸多領(lǐng)域中發(fā)展起來，如在注射成型及設(shè)計(jì)、建筑設(shè)計(jì)、模

6、具設(shè)計(jì)投死, 規(guī)劃過程中,還有夾具設(shè)計(jì). 孫用六個(gè)數(shù)字組成代碼參數(shù),包括工件的形狀、機(jī)械部分、軸襯,第一定位裝置,第二定位裝置和夾緊裝置</p><p>　　1事例參數(shù)和事例圖書館的建立</p><p><b>　　1.1事例參數(shù)</b></p><p>　　事例參數(shù)應(yīng)該由工件的所有的特征組成，來區(qū)別不同的夾具. 使用他們能夠使操作方便. 因?yàn)?/p>

7、零件的形狀是多種多樣的, 在生產(chǎn)企業(yè)中制造的技術(shù)要求也不斷發(fā)展,許多特征作被用做事例參數(shù)將會(huì)使搜索速度降低,其主要特征是不重要的,因?yàn)榉峙浣o每個(gè)特征的比重必須減少. 另一方面,事例參數(shù)包含所有的特征是困難的。 </p><p>　　因此,考慮到實(shí)際和快速設(shè)計(jì)的需求,事例參數(shù)要包含工件的主要特征和夾具的結(jié)構(gòu)。事例參數(shù)代碼由16位數(shù)組成：13位數(shù)是事例特征 3位數(shù)是事例識(shí)別數(shù)字。 </p><p&

8、gt;　　前13位數(shù)代表13個(gè)特征。 每個(gè)數(shù)字與特征的一個(gè)屬性相一致,這可能是"*"、"？"、"1"、"2"，…，"A"、"B",…，"Z"，…，等其中的一個(gè)。其中，"*"是指任何一個(gè),"?"代表不確定,"0"代表沒有。 </p>

9、;<p>　　系統(tǒng)規(guī)定:夾具的類型，工件的形狀，位置模式不能是"*"和"?"。在設(shè)計(jì)系統(tǒng)時(shí),三個(gè)項(xiàng)目的屬性信息沒有這些選擇,這就意味著必須選擇確定的屬性。</p><p>　　最后三位數(shù)是事例識(shí)別號(hào)碼,如果事例特征的13位數(shù)是一樣的,這三個(gè)數(shù)字就用來區(qū)別他們。</p><p>　　該系統(tǒng)還規(guī)定:"000"是用于修正的

10、一個(gè)典型事例,其他事例"001"、"002"、…,這些是用于設(shè)計(jì)師查找參考事例的. 如果其中一個(gè)偶爾需要改變成典型事例,首先它必須要求改成"000",前面的自動(dòng)變成參考事例. </p><p>　　事例索引碼的結(jié)構(gòu)如圖1所示。</p><p>　　1—夾具類型； 6—工件重量； 11—夾緊模型；</p>

11、<p>　　2—工件形狀； 7—工件剛度； 12—夾具體；</p><p>　　3—工件材料； 8—加工內(nèi)容； 13—其他；</p><p>　　4—批 量； 9—過程所有物； 14到16—事例識(shí)別碼；</p><p>　　5—工件比例； 10—定位模型；</p><p>　　圖1

12、 事例索引碼的結(jié)構(gòu)</p><p><b>　　1.2事例庫</b></p><p>　　事例庫由許多預(yù)定義的事例組成。事例的描述是基于事例推理的最重要的問題之一。所以由索引碼復(fù)合。</p><p>　　1.3 事例的層次化</p><p>　　夾具的結(jié)構(gòu)相似被認(rèn)為是整個(gè)夾具，成分和內(nèi)容相似。所以，整個(gè)夾具事例庫，成分

13、事例庫，夾具的成分事例庫形成相同。整個(gè)夾具的設(shè)計(jì)資料通常是由工件資料和工件加工資料組成,這就意味著夾具的設(shè)計(jì)應(yīng)滿足特別功能的需求.全部夾具事例是由功能成分組成，它是用功能成分的名字和數(shù)字來進(jìn)行描述的。成分事例代表成員（成分功能和其他結(jié)構(gòu)成分，主要驅(qū)動(dòng)參數(shù)，數(shù)字，和它們的約束關(guān)系）。成分事例（夾具的最低層）是功能成分和和其他成分的結(jié)構(gòu)。在現(xiàn)代夾具設(shè)計(jì)中有很多參數(shù)化準(zhǔn)件和普通非標(biāo)準(zhǔn)件。所以成分事例圖書館應(yīng)記錄特殊參數(shù)和保持它們的方法。<

14、;/p><p><b>　　2事例修改的策略</b></p><p>　　在基于事例的夾具設(shè)計(jì)中，最重要的是相似點(diǎn)的修改，這樣能有助于獲得最相似的事例，以及縮短適應(yīng)時(shí)間。根據(jù)夾具設(shè)計(jì)的需求，事例修改的策略使最接近的事例方法和知識(shí)指導(dǎo)結(jié)合起來。首先在深度上查找，然后在寬度上；知識(shí)指導(dǎo)策略意味著在來自客觀事物根源的知識(shí)規(guī)則上查找，這就要首先查找固定類型，然后查找工件的形狀，第

15、三查找定位方法。例如，如果事例索引碼包括夾具類型的磨削夾具，就只查找所有的磨削夾具，然后查找工件形狀的盒子，第三查找一個(gè)平面兩個(gè)銷的定位方法。如果沒有合適的，就查找深度標(biāo)點(diǎn)，然后回到最上層，然后再找所有與寬度相關(guān)的事例。</p><p><b>　　修改方法：</b></p><p>　　(1)根據(jù)夾具事例庫的事例索引信息，查找有關(guān)事例庫。</p>&l

16、t;p>　　(2)將事例索引碼與事例庫的每個(gè)事例碼匹配，然后計(jì)算相似尺寸的價(jià)值。</p><p>　　(3)整理相似尺寸的次序，最大的架子是最類似的事例。</p><p>　　兩個(gè)事例之間的相似點(diǎn)是基于兩個(gè)事例特征之間的相似點(diǎn)。相似點(diǎn)尺寸的計(jì)算依靠特征的類型。相似點(diǎn)的價(jià)值可以通過數(shù)字化的價(jià)值來計(jì)算，例如比較重量分別是50kg 和 20kg的工件。非數(shù)字化的價(jià)值也能計(jì)算，例如，現(xiàn)在前1

17、3位索引碼都是非數(shù)字化的價(jià)值。一個(gè)夾具的相似尺寸的計(jì)算公式如下：</p><p>　　其中S表示通用夾具的相似尺寸，n表示索引特性數(shù)，表示每個(gè)特性的重量，表示事例庫中特性和相關(guān)夾具的特性的相似尺寸。同時(shí)， ，數(shù)值計(jì)算如下：</p><p>　　其中表示第i個(gè)特征的索引特性值，表示事例庫中第j個(gè)事例的第i個(gè)特征的特性值。</p><p>　　所以有兩種方法選擇相似夾具

18、。一個(gè)方法是建立數(shù)值。如果通用事例的相似尺寸值比給定的數(shù)值小，這些事例就不能選來作相似事例。事例庫最初建立的時(shí)候，只有一些事例，數(shù)值可以建小一點(diǎn)。如果有大量的相似事例，數(shù)值就應(yīng)該建的大一些。另外一個(gè)方法是只建立相似事例的數(shù)字（例如10），這是類型單里相似尺寸的最大值。</p><p>　　3 事例的修改和存儲(chǔ)</p><p><b>　　3.1事例的修改</b><

19、;/p><p>　　夾具設(shè)計(jì)中相似事例的修改包括以下三個(gè)階段：</p><p><b>　　(1)成分的替代</b></p><p>　　(2)保持形式不變，調(diào)整成分的特性</p><p><b>　　(3)模型重新設(shè)計(jì)</b></p><p>　　如果夾具的成分是普通的物品，它

20、們能通過使用工具被修改，代替以及刪除，這些已經(jīng)被設(shè)計(jì)好了。</p><p><b>　　3.2事例的存儲(chǔ)</b></p><p>　　在將一個(gè)新的事例保存到事例庫之前，設(shè)計(jì)者必須考慮保存是否有價(jià)值。如果這個(gè)事例不能增加系統(tǒng)的知識(shí)，就沒有必要把它保存到事例庫里。如果它有價(jià)值的話，設(shè)計(jì)者在保存之前必須分析一下，看看這個(gè)事例是否作為標(biāo)準(zhǔn)事例或參考事例被存儲(chǔ)了。一個(gè)標(biāo)準(zhǔn)事例是

21、一個(gè)描述同族事例主要特征的標(biāo)準(zhǔn)。一個(gè)同族事例是有事例庫中索引碼前13位相同而最后三位不同的那些事例組成的。一個(gè)標(biāo)準(zhǔn)事例的最后三位通常是“000”。一個(gè)參考事例屬于同族標(biāo)準(zhǔn)事例，最后三位用不同數(shù)字區(qū)分。</p><p>　　從被解釋的概念中，可采用以下方法：</p><p>　　(1)如果一個(gè)新的事例和任何一個(gè)存在的事例族一致，和一個(gè)存在的標(biāo)準(zhǔn)事例的前13位數(shù)相同，那么這個(gè)事例就不能存儲(chǔ)因?yàn)?/p>

22、已經(jīng)這種標(biāo)準(zhǔn)事例了?；蛘咧荒茏鳛橐粋€(gè)參考事例保存（最后三位不是“000”，而且和其它的不一樣）在事例庫中。</p><p>　　(2)如果一個(gè)新的事例和任何一個(gè)存在的事例族一致，并且被認(rèn)為代替這個(gè)事例族要比以前的標(biāo)準(zhǔn)事例好，那么這個(gè)標(biāo)準(zhǔn)事例就被這個(gè)新的事例代替，以前的標(biāo)準(zhǔn)事例作為一個(gè)參考事例保存。</p><p>　　(3)如果一個(gè)新的事例和任何一個(gè)存在的事例族不一致，一個(gè)新的事例族將會(huì)自

23、動(dòng)產(chǎn)生，并作為標(biāo)準(zhǔn)事例保存到事例庫中。</p><p>　　4夾具設(shè)計(jì)中基于事例推理的過程</p><p>　　根據(jù)夾具設(shè)計(jì)的特性，夾具設(shè)計(jì)的基本信息，例如夾具的名字，零件，生產(chǎn)和設(shè)計(jì)者等等，必須先輸入。然后，輸入或設(shè)計(jì)工件的模型。輸入有關(guān)工件的細(xì)節(jié)信息，建立事例索引碼，然后CBR開始依靠相似尺寸查找相似事例，選出最相似的事例。如果需要的話，事例要滿足通用性設(shè)計(jì)，再存儲(chǔ)到事例庫中。程序流程

24、圖如圖2所示</p><p>　　圖2 基于事例推理的夾具設(shè)計(jì)流程圖</p><p>　　5基于事例推理的夾具設(shè)計(jì)說明</p><p>　　這是一個(gè)工件如圖3所示。材料是45鋼，底座，形狀為塊狀，生產(chǎn)批量為中批等。需要設(shè)計(jì)成一個(gè)用來旋轉(zhuǎn)孔的旋轉(zhuǎn)夾具。</p><p>　　圖3 需要設(shè)計(jì)夾具的一個(gè)工件</p><p>

25、　?。ㄗ畲蟪叽?0mmx49mmx22mm）</p><p>　　工件的特征值，屬性值，事例索引碼和重量在表1中列出。</p><p>　　表1 工件的事例索引碼和重量</p><p>　　特征名稱 特性值 索引碼 重量</p><p>　　夾具類型 車床夾具 1 100&

26、lt;/p><p>　　工件形狀 塊狀 9 90</p><p>　　工件材料 中碳鋼 3 70</p><p>　　批量 中批 2 60</p><p>　　工件比例 小

27、 5 60</p><p>　　工件重量 輕 5 60</p><p>　　工件剛度 硬度強(qiáng) 1 60</p><p>　　加工內(nèi)容 孔 3 80</p><p>　　程序要求

28、 完成加工 3 70</p><p>　　定位方法 三個(gè)平面 1 100</p><p>　　夾緊方法 不確定 ？ 90</p><p>　　夾具體 復(fù)合 4 80</p><

29、p>　　其他 沒有 0 60</p><p>　　通過查找和計(jì)算相似點(diǎn)，最相似的事例的事例索引碼是19325513321402000，細(xì)節(jié)信息在表2中列出。</p><p>　　表2 最相似事例的事例索引碼</p><p>　　特征名稱 特性值 索引碼</p>

30、<p>　　夾具類型 車床夾具 1 </p><p>　　工件形狀 塊狀 9 </p><p>　　工件材料 中碳鋼 3 </p><p>　　批量 中批

31、 2 </p><p>　　工件比例 小 5 </p><p>　　工件重量 輕 5 </p><p>　　工件剛度 硬度強(qiáng) 1 </p>

32、<p>　　加工內(nèi)容 孔 3 </p><p>　　程序要求 完成加工 3 </p><p>　　定位方法 三個(gè)平面 1 </p><p>　　夾緊方法 不確定

33、 ？ </p><p>　　夾具體 復(fù)合 4 </p><p>　　其他 沒有 0 </p><p><b>　　相似點(diǎn)的計(jì)算如下：</b></p><p>　　所以夾具

34、的相似尺寸值是0.806，這是在事例庫中用于設(shè)計(jì)的最相似的事例，最相似的事例的結(jié)構(gòu)如圖4所示</p><p>　　圖4 最相似的夾具</p><p>　　當(dāng)成分替代，修改定位模型和夾緊模型，以及調(diào)節(jié)相關(guān)尺寸之后，新的夾具被設(shè)計(jì)出來，圖形如圖6所示</p><p>　　圖5 需要設(shè)計(jì)的新夾具</p><p>　　因?yàn)樵谑吕龓熘袥]有相似

35、夾具，新夾具被儲(chǔ)存到事例庫中。事例索引碼是19325523311402000。</p><p><b>　　6 結(jié)論</b></p><p>　　基于事例推理，作為一個(gè)問題解決的方法，是一個(gè)比模仿人類思想的專業(yè)系統(tǒng)更有效的方法，已經(jīng)在很多難獲取知識(shí)的領(lǐng)域里得到發(fā)展?；谑吕评淼膬?yōu)點(diǎn)如下：它和人類的思想很相似；一個(gè)事例庫通過保存新事例獲得自學(xué)能力，它比有慣例庫更快更容

36、易，它可以更好的傳遞和解釋新的知識(shí)，這和慣例庫有很大的不同?；谑吕评碇刑岢龅囊粋€(gè)夾具設(shè)計(jì)的框架已經(jīng)被實(shí)行了，使用的是支持基礎(chǔ)數(shù)據(jù)的VC++，UG電腦繪圖軟件。這個(gè)框架也已經(jīng)和普通成分庫和典型夾具庫結(jié)合起來。這個(gè)發(fā)展的標(biāo)準(zhǔn)系統(tǒng)，用于航空項(xiàng)目，幫助夾具設(shè)計(jì)者提高設(shè)計(jì)效率和重新使用先前的設(shè)計(jì)資源。</p><p>　　Application and development</p><p>　

37、　Of case based reasoning in fixture design </p><p>　　Abstract: Based on the case based designing (CBD) methodology, the fixture similarity is in two respects: the function and the structure information. Then

38、, the computer aided fixture design system is created on case based reasoning (CBR),in which the attributes of the main features of work piece and structure of fixture as case index code are designed for the retrieve of

39、the similar cases, and the structure and hierarchical relation of case library are set up for store. Meanwhile, the algorithm </p><p>　　Keywords: case based reasoning ;fixture design; computer aided design(C

40、AD)</p><p>　　Fixtures are devices that serve as the purpose of holding the work piece securely and accurately, and maintaining a consistent relationship with respect to the tools while machining. Because the

41、 fixture structure depends on the feature of the product and the status of the process planning in the enterprise, its design is the bottleneck during manufacturing, which restrains to improve the efficiency and lead-tim

42、e. And fixture design is a complicated process, based on experience that needs compreh</p><p>　　1 Construction of a Case Index and Case Library</p><p>　　1.1 Case index</p><p>　　The

43、case index should be composed of all features of the work piece, which are distinguished from different fixtures. Using all of them would make the operation in convenient. Because the forms of the parts are diverse, and

44、the technology requirements of manufacture in the enterprise also develop continuously, lots of features used as the case index will make the search rate slow, and the main feature unimportant, for the reason that the re

45、lative weight which is allotted to every feature must di</p><p>　　Therefore, considering the practicality and the demand of rapid design, the case index includes both the major feature of the work piece and

46、the structure of fixture. The case index code is made up of 16 digits: 13 digits for case features and 3 digits for case identification number.</p><p>　　The first 13 digits represent 13 features. Each digit

47、is corresponding to an attribute of the feature, which may be one of“*”, “？”, “1”, “2”,…,“A”,“B”,…, “Z”,…, etc. In which, “*” means anyone, “?” uncertain, “0” nothing.</p><p>　　The system rules: fixture type

48、, work piece shape, locating model cannot be “*”or“?”. When the system is designed, the attribute information of the three items does not have these options, which means the certain attribute must be selected. </p>

49、<p>　　The last three digits are the case identification number, which means the 13 digits of the case feature are the same, and the number of these three digits is used for distinguishing them.</p><p>

50、;　　The system also rules: “000” is a prototype case, which is used for retrieval, and other cases are “001”,“002”,…, which are used for reference cases to be searched by designers. If occasionally one of them needs to b

51、e changed as the prototype case, first it must be required to apply to change the one to “000”, and the former is changed to referential case automatically.</p><p>　　The construction of the case index code i

52、s shown in Fig.1.</p><p>　　1.2 Case library</p><p>　　The case library consists of lots of predefined cases. Case representation is one of the most important issues in case based reasoning. So co

53、mpounding with the index code,.</p><p>　　1.3 Hierarchical form of Case</p><p>　　The structure similarity of the fixture is represented as the whole fixture similarity, components similarity and

54、component similarity. So the whole fixture case library, components case library, component case library of fixture are formed correspondingly. Usually design information of the whole fixture is composed of work piece in

55、formation and work piece procedure information, which represent the fixture satisfying the specifically designing function demand. The whole fixture case is made up o</p><p>　　2 Strategy of Case Retrieval<

56、;/p><p>　　In the case based design of fixtures ,the most important thing is the retrieval of the similarity, which can help to obtain the most similar case, and to cut down the time of adaptation. According to

57、the requirement of fixture design, the strategy of case retrieval combines the way of the nearest neighbor and knowledge guided. That is, first search on depth, then on breadth; the knowledge guided strategy means to sea

58、rch on the knowledge rule from root to the object, which is firstly searched by </p><p>　　Retrieval algorithms:</p><p>　　1)According to the case index information of fixture case library, search

59、 the relevant case library;</p><p>　　2)Match the case index code with the code of each case of the case library, and calculate the value of the similarity measure;</p><p>　　3)Sort the order of s

60、imilarity measure, the biggest value, which is the most analogical case.</p><p>　　Similarity between two cases is based on the similarity between the two cases. features. The calculation of similarity measur

61、e depends on the type of the feature. The value of similarity can be calculated for numerical values, for example, compare</p><p>　　Work piece with the weight of 50kg and 20kg. The value can also be calculat

62、ed between non numerical values, for example, now the first 13 digits index code is all non numerical values. The similarity measure of a fixture is calculated as follows:</p><p>　　where S is the similarity

63、measure of current fixture, n is the number of the index feature, is the weight of each feature, is the similarity measure of the attribute of the i2th feature with the attributeof relative feature of the j-th case in

64、 the case library. At the same time, , the value counts as follows:</p><p><b>　　.</b></p><p>　　Where is the value of the index attribute of the i-th feature, and is the value of a

65、ttribute of the relative i-th feature of the j-th case in case library.</p><p>　　So there are two methods to select the analogical fixture. One is to set the value. If the values of similarity measure of cur

66、rent cases were less than a given value, those cases would not be selected as analogical cases. When the case library is initially set up, and there are only a few cases, the value can be set smaller. If there are lots o

67、f analogical cases, the value should get larger. The other is just to set the number of the analogical cases (such as10), which is the largest value of si</p><p>　　3 Case adaptation and Case Storage</p>

68、;<p>　　3.1 Case adaptation</p><p>　　The modification of the analogical case in the fixture design includes the following three cases:</p><p>　　1) The substitution of components and the co

69、mponent;</p><p>　　2) Adjusting the dimension of components and the component while the form remains; </p><p>　　3) The redesign of the model.</p><p>　　If the components and component

70、 of the fixture are common objects, they can be edited, substituted and deleted with tools, which have been designed.</p><p>　　3.2 Case storage</p><p>　　Before saving a new fixture case in the c

71、ase library, the designer must consider whether the saving is valuable. If the case does not increase the knowledge of the system, it is not necessary to store it in the case library. If it is valuable, then the designer

72、 must analyze it before saving it to see whether the case is stored as a prototype case or as reference case. A prototype case is a representation that can describe the main features of a case family. A case family consi

73、sts of those cases </p><p>　　From the concept that has been explained, the following strategies are adopted:</p><p>　　1) If a new case matches any existing case family, it has the same first 13

74、digits as an existing prototype case, so the case is not saved because it is represented well by the prototype case. Or is just saved as a reference case (the last 3 digits are not “000”, and not the same with others) in

75、 the case library.</p><p>　　2) If a new case matches any existing case family and is thought to be better at representing this case family than the previous prototype case, then the prototype case is substit

76、uted by this new case, and the previous prototype case is saved as a reference case.</p><p>　　3) If a new case does not match any existing case family, a new case family will be generated automatically and t

77、he case is stored as the prototype case in the case library.</p><p>　　4 Process of CBR in Fixture Design</p><p>　　According to the characteristics of fixture design, the basic information of the

78、 fixture design such as the name of fixture, part, product and the designer, etc. must be input first. Then the fixture file is set up automatically, in which all components of the fixture are put together. Then the mode

79、l of the work piece is input or designed. The detailed information about the work piece is input, the case index code is set up, and then the CBR begins to search the analogical cases, relying on the </p><p>

80、;　　5 Illustrating for Fixture Design by CBR</p><p>　　This is a work piece (seeFig.4). Its material is 45# steel. Its name is seat. Its shape is block, and the product batch size is middle, etc. A fixture is

81、turning fixture that serves to turn the hole, which needs to be designed.</p><p>　　The value of feature, attribute, case index code and weight of the work piece is show n in Tab.2.</p><p>　　Thro

82、ugh searching, and calculating the similarity, the case index code of the most similar case is 19325513321402000, and the detailed information is show n in Tab. 3.</p><p>　　The similarity is calculated as fo

83、llows:</p><p>　　So the value of similarity measure of the fixture which needs to be designed with the most analogical case in case library is 0.806, and the structure of the most analogical case is shown in

84、Fig.5.</p><p>　　After having been substituted the component, modified the locating model and clamp model, and adjusted the relative dimension, the new fixture is designed, and the figure is show n in Fig.6.&

85、lt;/p><p>　　As there is not the analogical fixture in the case library, the new fixture is restored in to the case library. The case index code is 19325513311402000.</p><p>　　6 Conclusion</p>

86、<p>　　CBR, as a problem solving methodology, is a more efficient method than an expert system to simulate human thought, and has been developed in many domains where knowledge is difficult to acquire. The advantag

87、es of the CBR are as follows: it resembles human thought more closely; the building of a case library which has self learning ability by saving new cases is easier and faster than the building of a rule library; and it s