项目名称: 面向协同的设计重用启发模型
项目编号: No.51305052
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 机械、仪表工业
项目作者: 孙长乐
作者单位: 大连海事大学
项目金额: 25万元
中文摘要: 通过基于内容的3维CAD模型相似性识别进行设计重用,可以提高设计的成熟度、缩短产品研制周期,在实现手段上,特征法相对几何法具有建模意图不丢失、目标模型可修改的优势。但当前的特征法对由不同特征构建的相同模型,鲁棒性不强,且未考虑拓扑的几何性及各特征对模型相似度贡献的区别。针对上述问题,本项目拟通过研究模型相似机理,利用特征生成面交集因子判别特征相似性,采用方向关系判别拓扑依赖的几何相似性,改善特征法的鲁棒性、提高拓扑关系识别的准确性;采用复杂度与显示度计算特征相似权重,建立基于特征相似加权索引树的模型相似度算法,区分具体特征对模型相似的贡献,提高相似度的计算速度与准度;基于分类/聚类方法与提示优先级算法,构建面向协同的设计重用启发模型,支持在零件造型与零件关系两个层面进行设计重用。本项目研究可以改善特征法的鲁棒性及相似度计算的准确性,实现零件关系重用,为在产品研制中进行设计重用提供方法支持。
中文关键词: 模型检索;模型重用;相似度;模型方向;导入接口
英文摘要: Design reuse by content-based 3D CAD model similarity recognition can improve the maturity of the design and shorten the product development cycle. Feature-based method has some advantages over geometry-based method, e.g., modeling knowledge will not be lost and the target model can be modified. But current feature-based methods are not robust to the same model built by different features, and not consider the geometry of topology and the model similarity contribution difference between features. Aiming at the above problems, this project intends to study the similarity of models, use surface intersection of feature to discriminate the feature similarity, and adopt the orientation relation to discriminate topology dependent geometric similarity, which will enhance the robustness of feature-based method and promote the accuracy in the recognition of the topology relation. This project will also calculate feature similarity weight by introducing complexity and size, and then establish a model similarity algorithm based on the feature similarity weighted index tree to distinguish specific feature contribution to model similarity, which will improve the calculation speed and accuracy. And, this project will establish a collaboration-oriented design reuse heuristic model based on classification/clustering method and
英文关键词: Model retrieval;Model reuse;Similarity;Model orientation;Import interface