2022年完整word版,StudyonCase-BasedFixtureDesign .pdf

Study on Case-Based Fixture DesignSHI Qi, YANG Hai-chcng, U Yuan, Slate Key Laboratory of CAD/CAM, Northwestern Polytechnical University, Xian 710072, P. R. ChinaAbstract : Fixture is an important manufacturing activity. A fixture design system based on case-based reasoning (CBR) is proposed in this paper. A new method of case representation on the basis of fixture Junction is presented, where the case representation is constituted of workpiece knowledge, processing feature knowledge, and fixture feature knowledge. Running the prototype system shows that the knowledge representation method, using cases, is a better way to transform and explain the design knowledge.Key words: case-based reasoning, fixture design, knowledge representation, case retrieval1. Introduction Fixtures are work holding devices and are used in most of the manufacturing operations such as machining, assembly, inspection, etc. As part of manufacturing planning, fixture design can be a major contributor to the manufacturing, lead-time, and the cost of a product. Owing to the great variety of parts and manufacturing operations, the processes of fixture design become more complex and uncontrolled. Moreover, the design of a fixture relies heavily on the designers expertise and experience as well as the data from manufacturing, processing, technical, and parts . A computer-aided fixture design (CAFD) technique has been developed as part of CAD/CAM integration. Nowadays, CAFD plays an important role in flexible manufacturing systems (FMS) and computer-integrated manufacturing systems ( CIMS). The first thing that must be considered by employing a computer in fixture design is to establish a suitable design planning model. The presentation method of knowledge used in design and the designers experience are essential to establish a successful fixture design system. Compared to classical rule-based reasoning and model-based reasoning, case-based reasoning (CBR) attempts to solve new problems by using previous cases from a case base and adapting the solution of a similar problem according to the parameters of the new situation. The major advantage of this approach over rule-based reasoning or model-based reasoning is that the knowledge used in design is too complex to be reduced to rules or models. The approach holds key benefits with respect to solving mechanical design problems. There are systems that have been documented as design systems in the CBR literature. One system called DEJAVU uses case-based reasoning to assist mechanical design. Another system called CADSYN is a hybrid case-based design model which is proposed to integrate specific domain knowledge. In this paper, cases are used to elaborate the domain knowledge and design experience. A method of case representation of fixture design based on fixture function is proposed. Section 2 describes the concept of case-based reasoning. Section 3 provides the case representation of fixture design and the structure of the case base. Section 4 presents the details of the architecture of the case-based fixture design system. The conclusions follow. 2. Case-based reasoning Case-based reasoning relies on storing solutions as well as problems and adapting these solutions to solve the new similar problems. Design is an ill-structured problem. Design experience and heuristics play important roles in the design process. However, most experience is difficult to articulate as compiled and formulated, so in a rule-based system, knowledge acquisition would be an obstacle. A CBR system, in contrast, solves problems directly by using experience gained from solving similar problems in the past rather than by compiling and generalizing it. A case records not only the result but also how this result 名师资料总结 - - -精品资料欢迎下载 - - - - - - - - - - - - - - - - - - 名师精心整理 - - - - - - - 第 1 页，共 5 页 - - - - - - - - - was obtained such as in what situation, according to what conditions, etc. Some engineers pioneered case-based reasoning as an alternative to the traditional rule-based and model-based reasoning techniques. A typical CBR process includes the steps as shown in Figure 1. First, the problem should be identified and described and this description information is then used to retrieve one or more similar cases from a case base. Case retrieval may be based on some algorithms to improve the accuracy. The result of case retrieval will be shown to the designer. The chosen case is adapted when needed to meet the requirements of the new problem. Finally, the solutions as well as the new problem are stored in the case base as a new case. Figure 1 A case-based design system 3.Representation of design cases and case base organization 3.1 The feature of fixture design In a manufacturing system, fixture design involves three activities: set-up planning, fixture planning, and fixture configuration design as shown in Figure 2. The objective of set-up planning is to determine the operation sequence, the position and orientation of the workpiece in each set-up and the number of the set-ups. Fixture planning determines the locating and clamping points on the workpiece surfaces. The task of fixture configuration design is to make decisions of how to locate and how to clamp. The fixture components are also selected and placed into position to locate and hold the workpiece. Fixture designFigure 2 Fixture design process The main purpose of set-up planning and fixture planning is to accomplish fixture configuration. Two kinds of information are needed during set-up and fixture plannings. One is workpiece information; the other is processing information. Both of these two kinds of information are derived from the original CAD model and the product process planning model. A case in a CBR fixture design system is combining workpiece structure information, processing information and fixture configuration information. 3.2 Abstract case function features The most important issue in case-based reasoning is the information description of design cases. Feature-based technology is used to depict the partsshape, function and structure. Traditionally, it is the limitations of the geometrical model of the CAD system that it can only describe a parts geometrical Setup planning Operation sequence Workpiece orientationFixture planning Locating surface (points) Clamping surface (points)Fixture configuration Element selection Position determinationCase-based reasoningRetrieval mechanismsAdaptaion mechanismsProblem descriptionDesign solutionCase baseProblem spec.Design solutionsDesign Knowledge名师资料总结 - - -精品资料欢迎下载 - - - - - - - - - - - - - - - - - - 名师精心整理 - - - - - - - 第 2 页，共 5 页 - - - - - - - - - information rather than the design experience. Feature-based technology can describe two kinds of knowledge; data relational knowledge such as shape, function, precision, material, and so on, and the knowledge of the relationship among data, such as design domain knowledge, designers experience, and formulas. Cases are aggregates of a set of features. Two parts of knowledge must be kept in cases stored in a case base: description of design problem and design planning. Description of design problem consists of a set of case features from design problems. Design planning is features abstracted from solutions. In fixture design, the fixture model integrates three types of knowledge; fixture solution, workpiece information, and processing information. According to the method of problem reduction in artificial intelligence, a complex design problem can be separated into several less complex sub-problems, and the sub-problems can be separated into more simple sub-sub-problems, and the like. In this system, a design case is organized in a hierarchy that consists of three kinds of feature information; fixture features, workpiece features, and processing features, and in the system the fixture features can be decomposed to function features, behavior features, and structure features. Fixture behavior features show the relationship between fixture functions and structure. Fixture function feature can be further divided into locating features, clamping features, supporting features, and guiding features. The information on a fixture case model is shown in Table 1. The importance of case features is defined in the control knowledge base of the system. Case repre-sentation is done only by summarizing the design features, where the importance and the interrelation of the cases need not be considered. There are two advantages: (1) it simplifies case structure and case storage in the database, (2) a new reasoning strategy independent of case representation and case base organization can be established. CaseinformationFeaturesValue(s)NameTypeAxis, dish, cover, cabinet, gear, and othersWorkpieceQuantityMass, medium, small-lotMaterialSizeLength x width x heightPart drawingFile name and pathMachine nameLathe, milling, planer, grinderContents of processingPlane, outer cylindrical surface, inner hole surface, slotFeed typeLine, circleProcessing featuresProcessing PrecisionFinish machining, rough machiningType of principal axisHorizontal, verticalType of principal locating surfacePlane, outer cylindrical surface, inner hole surface, curved surfacePrecision of principal locating surfaceFinish machining, rough machining, in the roughParts schedule drawingFile name and pathFixture nameFixture typeLathe fixture, milling fixtureLocating modePlane, cylindrical surfaceFixtureLocating deviceTwo-pin, v-block, etc.featuresClamp requirementVertical clamping, horizontal clampingType of damping contactPoint-contact, surface-contactPowerManual clamp, etc.Clamp deviceEccentric wheel, etc.3.3 Case representation The method of case-based knowledge representation is usually not a new method but the higher-level abstract of the common knowledge representation, such as first order logic, production rule, frame, and semantic network. The implementation of case representation is based on those common knowledge 名师资料总结 - - -精品资料欢迎下载 - - - - - - - - - - - - - - - - - - 名师精心整理 - - - - - - - 第 3 页，共 5 页 - - - - - - - - - representation methods. In this paper, an object-oriented knowledge representation is used to describe fixture design cases. Essentially, the object-oriented knowledge representation is combined with a set of common knowledge representation methods in terms of object-oriented technology. The case knowledge can be divided into two parts: one is specific tuple of feature-value pair, and the other is design experience and rules. The representation of the case is similar to the conception of class in object-oriented technology. The feature-value pairs are the attributes in class, which contains the precondition and solution information. Design experience and rules are the methods of class. The contents of a case in object-oriented technology is illustrated as follows. Class / * list of case feature value pair * / Structure (STATIC_STRUCTURESJMPL) / * static structures used to depict specific design function Methods (CASE_FUNCT101VS_LIST) / * list of case function */| ; where,: = CASE_NAME BY DEFAULT) | (CASE_NAME BY USER) FEATURE_VALUE_PAIR_LIST) : = ( FEATURE _TYPE) (FEATURE_NAME)|f (STATIC_STRUCTURESJMPL : = | STRUCTURE ( STRUCTURE_NAME) STRUCTURE_VARIABLES_LIST) | * CASE_FUNCTIONS_LIST) : = (FUNCTION_NAME) FUNCT10N_C0NDITI0N)(FUNCTION_CONCLUSION) The list of case functions contains the method list about the case design experience and design rules. Another problem that should be considered is the structure of the case base. A case base can be structured as a lined list or a frame with a hierarchical structure. We select the latter, which uses a relational database to store cases. The advantage of relational database management consists of three points-. (1) advanced ability to data management; (2) high performance in data retrieval and data maintenance; ( 3) direct operation relationship between relational calculus and logic calculus. 4.Case-based fixture design system Based on the case representation model mentioned above, a case-based fixture design system is built, and its architecture is shown in Figure 3. The system mainly consists of four components, a system control module, a case-based reasoner, fixture design memory, and a design evaluation module. A case-based reasoner is the core of the system, and it performs the case retrieval, case selection , and case transformation. Figure 4 is a screen for the working process. The system input consists of three kinds of information: workpiece general information, such as name, type, material, etc. workpiece processing information, such as machining information, precision and machine type; fixture features information, such as clamp requirement, contact type, power type, etc. The system modules contains the following: Control Module; used to monitor and manage data flow and message flow among other modules; Case Reasoning Module; can be separated into six sub-modules: weight production, judgment of functional similarity, judgment of structure similarity, case retrieval module, fixture case adaptation , and case storage; Case base; used for fixture case storage, establish case index, and perform case maintenance; Design evaluation: evaluate the result of reasoning and submit the solutions. 5.Conclusions The CBR technology is closer to human thought. A fixture design system using this technology is de-scribed in this paper. A new method of case representation on the basis of fixture function is presented, where the case representation is-constituted of workpiece knowledge, processing feature knowledge , and fixture feature knowledge. Running the prototype system shows that the knowledge representation method using cases is a better way to transform and explain the design knowledge. 名师资料总结 - - -精品资料欢迎下载 - - - - - - - - - - - - - - - - - - 名师精心整理 - - - - - - - 第 4 页，共 5 页 - - - - - - - - - References1 Y.CChou.R. A. Srinivas and S. Sujit, Automaticdesign of machining fixtures ? conceptual design.International Journal of Advanced Manufacturing Technology, No. 9 pp. 3 - 12, 1994 2 A. J. Trappey and C. R.U u, A literature survey of fixture design automation. International Journal of Advanced Manufacturing Technology, Vol. 5 , No. 3 , pp. 240 - 255, 1990 3 W. Ma, J. Li and Y. Rong, Development of automated fixture planning systems. International Journal of Advanced Manufacturing Technology, Vol. 15: 171-181, 1999 4 Theodore Bawlasz, Ibrahim Zeid, DEJAVU-. Case-based reasoning for mechanical design. AIEDAM, Vol.7, No.2, pp.111 -124, 1993 5 Mary Lou Maher, D. M. Zhang, CADSYN- A case-based design, process model. AIEDAM. Vol.7, No.2, pp.97 -110, 1993 6 Y. Rong and Y. Bai, Automated generation of modular fixture configuration design. Transactions ASME Journal of Manufacturing Science and Engineering, Vol. 119, pp.208 -219. 7 B. Kumar and B. Raphael, CADREM; A case-based system for conceptual structural design. Engineering with Computers, Vol. 13, pp. 153 164,