【作者】 郑惠江；

【导师】 王太勇；

【作者基本信息】 天津大学 ， 机械制造及其自动化， 2010， 博士

【摘要】 加工质量信息的获取、表达、传递和利用是数字化制造技术的重要内容。在机检测系统的使用能够有效的提高高档数控装备的使用效率,降低生产成本。本文围绕在机检测中曲面拓扑特征重建和检测点分布的关键技术展开了深入研究,并构建了复杂空间型面在机检测系统。对三角面片进行拓扑重建和曲面划分是解决在机检测系统中零件模型表面特征识别的前提条件之一。本文在建立点-边拓扑关系的基础上,提出了基于关联-散列结构三角网格拓扑重建方法,在滤除冗余数据的同时完成拓扑关系的建立,并对STL(stereo lithographic)模型的基本拓扑关系进行拓展,提出了一种基于无向图结构关系模型的曲面划分方法。通过计算实例验证了上述算法的高效性和有效性。根据工件的几何特征及其参数确定采样策略,并进行采样点的选择和布局是在机检测系统中重要的环节之一。本文根据三角面片法向量在高斯球面的映射图像判定自然二次曲面几何特征,将快速聚类法和Gauss-Newton法相结合,提出了一种针对STL模型数据文件格式特点的自然二次曲面特征参数提取方法。对于一般随机采样方法的局限性,分析了CVT(Centroidal Voronoi Tessellations)结构采样法和可展曲面的特点,提出了基于CVT结构的可展曲面采样策略。针对曲面采样点布局算法对于离散数据环境下的复杂曲面测点布局的局限性,通过引入局部网格曲率特性参数,构建了基于三角面片微分几何特性的网格简化方法,提出了基于三角形折叠的复杂曲面离散数据检测点布局策略;并构造了连续LOD(Level Of Detail)模型,建立了不同层次网格之间距离值相对于简化三角形数量的拟合曲线方程,验证了曲率权值优化方法在离散数据环境下对提高复杂曲面采样点布局精度的有效性。最后,在对在机检测系统进行功能需求分析的基础上,搭建了在机检测系统的原型实验平台,提出了复杂空间型面在机检测系统的层次化结构功能模型,并采用计算机建模和仿真技术开发了在机检测过程仿真模块,结合规范化报告输出技术,建立了以上述方法为核心的复杂空间型面在机检测系统。更多还原

【Abstract】 Access to quality information processing, expression, transfer and use is an important part of the digital manufacturing technology. The On Machine Verification (OMV) system can enhance the availability factor of the high-grade NC equipment in effect and reduce the cost of production. The investigation on the key technology of the topological feature reconstruction and sampling point distribution in OMV system was carried out. The OMV system for complex curved surfaces was developed.The topological reconstruction and surface division for the triangular faces is one of the preconditions for the feature recognition for some surfaces of the part model in the OMV system. On the base of the topological relation between vectors and edges, a new algorithm of topological reconstruction for triangular mesh based on Correlation-Hashing was proposed. The redundant data were filtered in the algorithm, meanwhile, the topological relations were built. A new method based on the relation model of undirected graph for surface division, which is on the base of the basic topological relations expansion in STL(stereo lithographic) models, was proposed.The calculation examples demonstrated that the algorithms were efficient and stable. A determination of sampling strategy trajectory planning of measurement according to the geometric features and their parameters of work pieces is one of the important steps in OMV system. According to the mapping images of the triangular faces’normal vectors on the Gauss sphere, the geometric features of natural quadric surfaces were judged. The characteristic parameters were calculated by fast clustering algorithm method and Gauss-Newton. A new method for the characteristic parameter extraction, which is suitable for the STL format, was given. The characteristics of the sampling strategy based on CVT(Centroidal Voronoi Tessellations) and the distinguishing features of developable surfaces are analyzed and the sampling strategy based on the CVT for developable surfaces were proposed aiming at the limitation of common random sampling strategy.Aiming at the limitation of sampling point distribution algorithms for that of complex curved surfaces in dispersed data environment, the curvature characteristic parameters of local grid were introduced to construct the method of mesh simplication based on the differential geometric properties of triangles. On the basis of this, the sampling distribution strategy of complex curved surfaces based on triangle collapse was proposed. The LOD(Level Of Detail) model was constructed. The Fitting curve equations were established according to the relationship of the distance between the different levels of meshes to the amount of simplification triangles. The effectivity of curvature weights optimization in improving precision of the sampling point distribution strategy based on triangle simplification in dispersed data environment was verified.At last, on the basis of requirement analysis for the OMV system, the prototype experiment environment was constructed. The hierarchically functionalmodeling of the OMV system for complex curved surfaces was proposed. The OMV process simulation module was developed using computer modeling and simulation technology. Combining technology of standardized report outputting, the OMV system for complex curved surfaces whose key technology was composed of that methods was developed.更多还原