帶式輸送機傳動裝置的總體設(shè)計【同軸式二級圓柱齒輪減速器】【F=2400Nv=1.2m-s,d=300mm】

帶式輸送機傳動裝置的總體設(shè)計【同軸式二級圓柱齒輪減速器】【F=2400Nv=1.2m-s,d=300mm】,同軸式二級圓柱齒輪減速器,F=2400N,v=1.2m-sd=300mm,帶式輸送機傳動裝置的總體設(shè)計【同軸式二級圓柱齒輪減速器】【F=2400N,v=1.2m-sd=300mm】,輸送,傳動,裝置,總體 CHINESE JOURNAL OF MECHANICAL ENGINEER【NG Vo122，No5，2009 671 DOI：103901CJME200905671，available online at www cjmeneteom；wwwcjmenetcornca Design System of the Twostep Gear Reducer on Casebased Reasoning JI Aimin ， 一，HUANG Quansheng ，XU Huanmin ，and CHEN Zhengming。 1 College ofMechanical and Electrical Engineering,Hohai University,Changzhou 213022，China 2 StateKeyLab ofCAD&CG,Zhiang University,Hangzhou 310058,China 3 College ofComputer and Information Engineering,Hohai University,Changzhou 213022，China Received November 1 8，2008；revised May 22，2009；accepted August 10 ，2009；published electronically August 14，2009 Abstract：The design of the two-step gear reducer is a tedious and time-consuming process For the purpose of improving me efficiencv and intelligence of design process，casebased reasoning(CBR)technology was applied to the design of the twostep gear reducen Firstly, the current design method for the twostep gear reducer was analyzed and the princiFlle of CBR was described Secondly,according to the characteristics of the reducer,three key technologies of CBR were studied and the corresponding methods were provided ，which are as follows：(a)an objectoriented knowledge representation method,(b)a retrieval method combining the nearest neighbor with the induction indexing，and(c)a case adaptation algorithm combining the revision based on rule with artificial revisionAlsofor the purpose of improving the credibility of case retrieval，a new method for determining the weights of characteristics and a similarity formula were presented，which is a combinatorial weighting method with the analytic hierarchy process(AHP)and roughness set theory Lastly,according to the above analytic results，a design system of the two-step gear reducer on CBR was developed by VC+ UG and Access 2003A new method for the design of the twostep gear reducer is provided in this studyIf the foregoing developed system is applied to design the two-step gear reduceg design efficiency is improved，which enables the designer to release from the tedious design process of the gear reducer so as to put more efforts on innovative designThe study result fully reflects the feasibility and validity of CBR technology in the process of the design of the mechanical parts Key words：twostep gear reducecasebased reasoning(CBR)，weights of characteristics，similarity 1 Introduction The traditional design method of the twostep gear reducer is a timeconsuming processEven now CAD is used in its design processrestriction of the traditional design method is not eliminated up to nowFor the sake of changing this kind of situation，many people have done research on the design method of reducer，mainly including the parametric design of reducer and expert system(ES)These two kinds of memods have respective shortcomings：for parametric design，every parts relation need to be fully considered，because a parameters change carl cause a change of a reducers other parameters or structure，which requires designer to acquaint with a reducer and be able to use a design software including the function of the parametric designFurthermore knowledge acquirement is very difficult in reducers ESbecause some expert knowledge js very difficult to express with rules Casebased reasoning(CBR、can improve these two kinds Corresponding authorEmail：iamustcedu This project is supported by National Hitech Research and Development Program of China(863 Program， Grant No 2008AA04Z115)，Science and Technology Program of the Ministry of Construction of China(Grant No2008一K821，Jiangsu ProvinciaI NatI1ra1 Science Foundation of China(Grant NoBK2007042)，and Open Fund of State Key Lab of CAD&CGZhejiang University,China(Grant No A0914、 of shortcomingsA new case is finished on the base of an existed case in CBRso the new case can be achieved by modilying some parts or directly making use of the existed case，and case acquirement is easier than rule acquirement， because the primary knowledge js cases in CBRL JIn f-act case usually provides more information than rule or model_jJ_Many researchers have done a lot of work for applications Of CBR in engineering SUN et alL斗J developed an intelligent fixture design system on CBR KW0NGet al_)Jintroduced a approach to determine proper injection moulding parameters by developed CBR systemLIUet al proposed a retrieval algorithm integrated with the clustering technique to locate the similar cases in the casebase and gave a casebase to illustrate the feasibility of the CBR system in the mechanical design PETERL developed an automated knowledgebased system on CBR for intelligent support of the preprocessing stage of engineering analysis in the contact mechanics domainX10NGet al provided an applied and creative conceptua1 design method based on CBR that embodies the indus仃ial design knowledgeThe system developed abbreviates the conceptual design process，help designers， and provides a base for the following development of productHowever the works described above paid attention to the theoretical research on CBR superior to the combination of CBR technology with mechanical product 672 JI Aimin，et al：Design System ofthe Twostep Gear Reducer on C ase -ba sed Rea soning So，taken the twostep gear reducer as example，the whole process of mechanical parts on CBR will be discussed 2 Key Technologies in CBR The CBR is a kind of similar or analogical methOd When a CBR system solves a new problem，it retrieves one or more cases from the antecedent cases that are the most similar to the new problem，and modifies the cases to sarisfy the new situation The flowchart of the CBR is shown in Fig1According to the flowchart of the CBR， the development of the product design system on CBR needs to solve some problem，including case description， case retrieval，case modification，case study and case base maintenance where case description，case retrieval and case modification are called three key technologies Fig1Flowchart of CBR Case description is the action that cases are coded to data structure accepted by computer with some conventional signsCase can be described by some methods including flame，object，predicate，semantic network and rules，etc， among which，frame and object are most commonly used Case retrieval is a process finding out one interrelated case or more similar cases by characteristic index and similarity JRef91 divides case retrievalinto three parts： characteristic identificationpreliminary matchand best selectionThe paper divides case retrieval into four parts： key characteristics extraction，characteristic identification preliminary match，and best selectionDataset of data mining often includes many characteristic attributesand some attributes are irrelevant to data miningThose irrelevant attributes influence efficiency of data mining Removing those irrelevant attributes can improve efficiency of data mining and make the result of data mining easier to understandThe purpose of key characteristics extraction is to select key characteristics to establish valid index from the case baseThe Durpose of characteristic identification is to select keY characteristics of new caseCharacteristic identificatiOn can be often made reference to key characteristic extractionBecause it includes a plentiful characteristic attributesthe key characteristic extraction of the twostep gear reducer is more importantThese typical methods of case retrieval include nearest neighbor， induction indexing， knowledgeguided，neural indexing on knowledge and template retrievalit0J Case modification is the process modifying the best- match case to meet the new design requirements，andthe most used methods include artificial modification， knowledgeintensive modification and knowledge-lacking modificationIt】 3 Design of the Twostep Gear Reducer on CBR 31 Case base building 3i1 Case&scription The paper USeS a case representation model of object- oriented layer 一 which provides an uniform object oriented data model to the upper so,ware，namely， provides all kinds of objectoriented concept，data structure， maintenance operation and flexible expansion，and provides a relation model of shielding concrete database to bottom，and establishes a transparent object conversion mechanism by mapping principle and carries on reasonable and valid managementMapping principle of object model is a relation of the conversion between a upper obiect oriented data model and a bottom relation model of databaseThe relation among object model layer，database， and other fimction module is shown in Fig2 user lnterface l Knowledge study and Knowledge reasoning and I mle ma ng mcFdu1e explaining module - 。 Description ofobjectoriented concept and its Object，oriented valid maintenance and flexible expansion model Controlling visiting and operation to underlying database by mapping principle J f Underlying 廠) database Fig2Objectoriented module representation The twostep gear reducer is a complicated assemblyfor the sake of convenience of case representation，the twostep gear reducer is divided into five component classes and two part classesThe five component classes are made of high speed gear group，low speed gear group，high speed shaft and bearing，middle shaft and bearing and low speed shaft and bearing，and the two part classes are made of the cover and the housingThese component classes can be broken up part classesFinally，every part class is mapped to its table in the underlying relation databaseTherefore，the objectoriented knowledge model on the twostep gear reducer is given in Fig3 CHINESE JOURNAL OF MECHANICAL ENGINEERING 673 oujectm。ddlayer f Thetw。一step gear reducer I l f + + + l High speed I 1 Low speed 曲speed shaft Middle shaft Low speed sham f geargroup f f geargroup and bearhag andbearhag andbearing 1 ! !J V ! ! + + 、 J s嘞edi Bearing shaft 薯 攀： ： Low ：R?。?： l l L 二 p =卜； ：；， - t V I lTable gear1l Table gear2l Tablehigh shaft l Tablemidshaft l Tablebear2 lTabkbear3ITablexzuof Tablexgaif under1v_m9 rda nn dtahA P f Table bear2 Table CBRf l Tablelow shaft f Fig3Objectoriented knowledge model on the twostep gear reducer 312 Casebasebuilding According to the case representation model of the twostep gear reducer in Fig3，characteristic attribute values of the twostep gear reducers and their components are mapped to the underlying database，thus the case base is builtIn the underlying databaseeach table includes a “case numberfieldThe relation is established by“case numberbetween the tables mapped by every part and by the reducerFig4 shows the storage mode of case base of the twostep gear reducers in the database software of AccessDue to the limited spacethe paper doesnt list successively the partsstorage mode in Access Fig4Storage mode of the twostep gear reducers case base in the database software ofAccess 32 Case retrieval The Paper applies a retrieval method Of the combination of the nearest neighbor and the induction indexing，because the design of the two-step gear reducer is an experience process with a long history,twostep gear reducer includes many casesSoit is better to use the induction indexing to have a rOBgh retrieval。and use the nearest neighbor to have a fine retrieva1Fig5 shows me detailed retrieval process Fig5Flowchart of case retrieval process Now the paper describes the child process and method to case retrieval of the design system of the two-step gear reduceronCBRinturn 321 Kevcharacteristicextraction The characteristics of original case base are pre- processed before the key characteristic extraction (1)Preprocessing of dataThe discrete normalization processing of the quantitative parameters are transformed into the qualitative parameters by equal-frequency- intervals The principle of equalfrequency-intervals is to divide original interval into N small intervals(N is a discrete number given by user)，while each small interval has the almost same number of dataThe first and last intervals are expanded in the paper：the upper limit of the first interval is changed to zero，and the lower limit of the last interval changed to infinity，which ensures that a attribute value of a new case has a corresponding small interval with itbecause CBR is a process of uninterrupted study(the amount of case will uninterruptedly increase1， and avoid that small interval is divided again whenever a new case is added and retrieval of new case is not made beyond the attribute value of original case baseN intervals are identified with 0，1，2，n-1The qualitative characteristic attributes are assigned to0，1 1 by an increasing or decreasing orderFor example，precision grade of reducerhas three options for user to select in the 一 磊 一一一一一一 一一 一 一。一。一 一。一 一。一一 等 螢 一一 674 JI Aimin，et al：Design System ofthe Twostep Gear Reducer on Casebased Reasoning original case base：the class of 7the class of 8 and the class of 9The classes of 7，8，9 are expressed with 1，05，0， respectively according to the above-mentioned method The attribute value of the Boolean attribute is easy to ascertaim the same is 1the other is 0 f21 Key characteristic selectionThese clearly irrelevant characteristies are not directly considered in the data miningHowever many characteristics are difficult to identify their significance in the data miningThese characteristics are selected by the valid strategyThe paper uses decision仃ee to select key characteristics by the size of information gainDecision index tree is established by key characteristiesThe algorithm on the information gain of decision attributes may be referred to Ref1 2 The key characteristic selection on the twostep gear reducer is shown aS followsThe original case base is given inTable 1 Table 1Original case base Firstly，the clearly irrelevant characteristics to data mining are directly eliminatedThe other characteristics by the discrete normalization processing are changed into qualitative attributes， and then selected by the abovementioned algorithm of information gainSecondly the attribute of transmission power is divided into three smal1 intervals according to the abovementioned equalfrequencyintervals：(0，1 0)，1 0，20and(2o，o。) The qualitative numerical ranges are represented respectively with 0，1，2The attribute of transmission ratio 1ifespan and transmission efficiency are respectively divided into several corresponding small intervals (transmission ratio：(0，15)and(15，。)；lifespan：(0，350 ooo)，(350 000，420 ooo)and(420 000，。)：transmission efficiency：(0，094)and(094，1)Each interval numerical range is represented with 0，1， in turnThe attribute of the layout of gear drives is divided into two typesThe developed configuration is represented with 0，and the reverted configuration is represented with 1The result of discrete processing is obtained in Table 2 Table 2Result of discrete processing The at仃ibute of arrangement forrn is viewed as identification attributeand the other attributes are viewed as decision at仃ibuteThe case base is divided into two classes f =21 according to identification attributeThe first class is comprised of six cases whose arrangement forrn is expanding form(rl=6)，and the second arrangement is comprised of other cases whose arrangement form is coaxial form(r2 4、According to the algorithm of information gain，it is easy to acquire the information gains of all attributes as follows：GP：029，6 012，G，=029， G 001；where，Gpis the gain oftransmission power，Gi is the gain of transmission ratio，GL b is the gain of lifespan， G”iS the gain of transmission emciencyThus the attribute of transmission powertransmission ratio and lifespan can be selected as key characteristics from the value of all characteristic attributes 322 Preliminary match The preliminary match is the process that a group of cases interrelated to the current design case are selected from the case baseThe process is realized by index tree built by al1 key characteristics and decision information gain calculationThe index tree is built up as shown in Fig6The preliminary match of case is fulfilled based on theindextree 0 Fig6Index tree of key characteristics 3 23 Best selection The best selection is the process which the best case is selected from the cases acquired by the preliminary match CHINESE JOURNAL 0F MECHANICAL ENGINEER【NG 675 The best case is selected by the nearest neighbor，so it is indispensable for calculating the weight of the kev characteristics and cases similarity f 1、Weight of characteristic The weight of characteristic is used to evaluate significance of characteristicIt can influence the accuracy of reasoning resultAccording to the source of original informationthe method of determining the weight of characteristic is divided into two classes：the subjective method and the objective methodThe information of the subjective method comes from experts，and the information of the oNective is from statistical original dataThe representative approaches of the twoclass method are the analytic hierarchy process(AHP) and roughness set theoryt12J AHP is a kind of decision method that the decision problem is divided into some hierarchies including target，rule，project，etc，and qualitative analysis and quantitative analysis are conductedLet af，be importance degree of the characteristic f compared with the characteristic jThe weight(Dli of characteristic f acquired bv AHP can be calculated by the following formula： 1 i=1，2，z (1) Roughness set theory is a kind of data reasoning method in view of knowledge classification，which mainly applies to analysis of the dependence between reduction of knowledge and characteristic attribute，and solves the problem about the weight of characteristic attribute of similarity measureIt iudges the importance of all characteristic by existent information according to specific classifyingRelevant formulas of the weight calculation are given as foUows： ，D)= fW i曲t ofthe characteristic i acquired by roughness set theory The paper applies a method of combination of AHP and roughness set theory in order to compensate the defect of the two methodsThe combined method is assembled according to the linear superposition principle，The material combined forin is shown in Eq(5)： q=ao)1 +(1一 ) 2f， (5) where fWeight ofcharacteristic acquired by AHP， fweight of characteristic acquired by roughness set theory， Coefficient According to the combined method，the three weights of key characteristics of the reducer is respectively obtained， such as the attribute of transmission power，co1=052；the attribute of transmission ratio，co2=028；the attribute of ljfe-span，w3=O20(suppose a=05) f2、Similarity The characteristic values of the twostep gear reducer are comprised of quantitative parameters and qualitative parametersThe paper uses Eqs(6)and(7)to calculate the similari16J of quantitative characteristic， and the similarity of qualitative and Boolean characteristic have two conditions：0(different)or 1(same)So the paper presents a new algorithm of case similarity，which is expressed as follows Relative distance： , ；。 hv, -vok1 ； Characteristics similarit-y： (6) =1-d(Vf一 )=1-d ； (7) (2) Similarity between two cases SGF(a，C，D)=r(C，J)一r(C一 ，D)， (3) SGF(C一f)，D) 羔SGF(C一i)，D) ( 一， ) I=l i=1，2， ， (4) where r(c，D)-Dependence degree between attribute set C andD， fPOS(G f mber of elements in the union set， I Number ofthe object set， SGF(a，c，D)-Importance ofattribute a attribute setD， a SGF(C-i，D)-Importance degree ofthe characteristic i， k coiSDn sim(n，k)=旦 一 i=1 (8) Where Re1atiVe distance of the ith attribute of case ,and case屯 f-Characteristic value of the ith attribute of case ， -Characteristic value of the ith attribute of casek， -Similarity of the ith attribute of case and case k， sim(nk)-Similarity between the case，z and the case l一” 、 可 口 兀 ，。 676 JI Aimin，et al：Design System ofthe Twostep Gear Reducer Oil Casebased Reasoning 劬一Weight of the ith characteristic， ，2_Number of the characteristic Suppose to design the two-step gear reducer whose design conditions are shown in Table 3，and retrieve two cases which are obtained in Table 4 Table 3Design conditions Table 4Two retrieyed cases The weights of