Assemblability evaluation is a pivotal step in design for assembly (DFA). In this paper, the authors analyze the problem existing in previous assemblability evaluation methods, put forward a new approach-Artificial Ne...Assemblability evaluation is a pivotal step in design for assembly (DFA). In this paper, the authors analyze the problem existing in previous assemblability evaluation methods, put forward a new approach-Artificial Neural Network based Assemblability Evaluation Method (AAE), describe the principles and implementations of the neural network in use for assemblability evaluation. AAE has been put into use in product design. The results of the experiments and analyses show that assemblability is measured more objectively, accurately and completely by using AAE than the conventional methods.展开更多
Design for assembly (DFA) evaluation has preved to be a systematic approach in determining how well a product is designed from the point of view of assembly. Geometric feature, such as symmetry, is an important factor...Design for assembly (DFA) evaluation has preved to be a systematic approach in determining how well a product is designed from the point of view of assembly. Geometric feature, such as symmetry, is an important factor when evaluating the assemblability of a part in it. But the automated detection of symmetry has always been considered to be a difficult problem which requires great effort. Proposed here is a new method for part symmetry automated recognition. The method is based on ray detecting theory and can effectively recognize the symmetry of parts with various shapes. It is of great importance to linking together DFA evaluation procedure with CAD model.展开更多
基金Supported by the National Natural Science Foundation
文摘Assemblability evaluation is a pivotal step in design for assembly (DFA). In this paper, the authors analyze the problem existing in previous assemblability evaluation methods, put forward a new approach-Artificial Neural Network based Assemblability Evaluation Method (AAE), describe the principles and implementations of the neural network in use for assemblability evaluation. AAE has been put into use in product design. The results of the experiments and analyses show that assemblability is measured more objectively, accurately and completely by using AAE than the conventional methods.
文摘Design for assembly (DFA) evaluation has preved to be a systematic approach in determining how well a product is designed from the point of view of assembly. Geometric feature, such as symmetry, is an important factor when evaluating the assemblability of a part in it. But the automated detection of symmetry has always been considered to be a difficult problem which requires great effort. Proposed here is a new method for part symmetry automated recognition. The method is based on ray detecting theory and can effectively recognize the symmetry of parts with various shapes. It is of great importance to linking together DFA evaluation procedure with CAD model.
