面向工艺规划的零件分类和识别方法研究

基于成组技术,提出了一种面向工艺规划的零件分类与识别方法,首先,基于现有零件的工艺信息,在对传统生产流程分析法零件相似系数进行改进的基础上,利用聚类分析方法进行零件族构造;然后,采用零件族累加矩阵来构建零件族的相似性标准,并建立了基于零件族累加矩阵的零件模式识别方法,该方法把编码分类法与零件-零件族的相似系数有机结合起来,确保新零件准确、自动地归属于相应的零件族.     

基于人工神经网络的零件族智能搜索算法

采用生产流程分析法对零件分组时,通常难以为新零件确定所属零件族,故提出了基于神经网络的零件族智能搜索算法。该方法将工艺与零件族的关系转化为零件编码与零件族之间的关系,基于现有零件的编码、工艺过程和零件族划分情况,用神经网络构建零件编码与零件族之间的智能非线性映射模型,从而实现新零件的零件族智能搜索。给出了算法的基本原理、神经网络设计和训练过程,并结合实例验证了算法的可行性和有效性。     

面向知识重用的角盒零件分类编码及识别方法EI北大核心

为满足角盒类零件工艺知识重用的分类编码识别需求,提出一种分面集特征匹配的零件分类编码识别方法。对整体零件形状和加工特征进行了分析定义,建立了基于加工特征的分类编码原则;利用型腔壁边面集深度差异,获取零件面深度0-1特征矩阵,分解提取壁边面集并判断零件基本类型;设计了面集几何拓扑矩阵和特征面边环关系矩阵,根据零件基本类型与面集属性选取典型特征矩阵进行匹配,缩小典型特征匹配范围以提高识别效率。结合实例说明了典型特征库的提取及其与编码类型的映射匹配,验证了分类编码识别算法的有效性与可行性。     

排序聚类分析在成组技术中的应用

为了找出工艺相似的零件组,在实践中,曾提出两种主要方法,一种是单纯凭借分类编码系统;另一种是单纯利用生产流程分析。只有将两种方法有机地结合,才能使问题得到完满的解决。     

基于模糊综合评判工艺决策的CAPP系统研究

针对工艺决策在实现C A PP智能化的重要性,在分析了C A PP专家系统的智能工艺决策的基础上提出一种基于层次分析的模糊综合评判的方法。基于典型工艺数据库,建立零件各个因素间模糊关系,结合零件编码加权相似系数检索法,通过采用一级编码和一级编码匹配率以及二级编码和二级编码匹配率的方式分配零件因素的权重。利用模糊综合评判技术来衡量新零件与模板工艺库中某一个实例零件的相似程度,实现简化零件加工工艺决策,为C A P P智能化提供了方法。     

基于实例检索的轴类零件工艺设计

轴类零件在机械产品中广泛应用,以轴类零件为对象,研究了基于实例相似性检索的轴类零件工艺规划。设计了特征编码结构进行实例分组,通过特征编码矩阵的匹配在实例库中检索相似的实例,以此获得待加工零件的加工工艺。基于实例技术的工艺设计方法具有较好的通用性,同时也提高了设计效率和质量。     

面向复杂零件成组编码系统及其与CAPP集成的研究

以一般零件和复杂零件为基础建立了工艺与设计相结合的成组编码系统。采用固定码对一般零件进行编码,而用固定码 柔性码,对复杂零件进行编码。制定了零件相似性标准和零件组的特征矩阵,并划分了零件组。构造了典型零件图库和典型零件工艺库。建立了基于成组技术的派生式CAPP系统。用VC 语言编制了零件分组查找程序,通过成组编码和所编制的程序零件可以快速、准确定位到所在的典型零件组,实现了编码系统与CAPP系统的集成。     

智能三坐标测量机的零件位姿识别

提高三坐标测量机对零件检测的智能化,实现零件的位姿识别是非常必要的。提出了一种新的视觉识别零件及位姿的方法,该方法首先比较全面地提取实际零件的三维特征信息,然后根据零件三维特征在虚拟空间中寻找匹配组合,并根据匹配组合计算旋转平移矩阵,最后根据实际和虚拟的零件成像轮廓的相似度进行位姿识别。实验表明新方法能够对零件及位姿进行识别,并且最大误差不超过1mm。     

零件分类成组的聚类分析法

本文是根据用于机械加工成组技术的分类原则,以聚类分析的方法,对零件进行无编码分类成组的一种新方法。文中详细介绍了用聚类分析法对零件进行分类成组的方法和步骤。这一方法可一次形成零件的多组分类,并以动态聚类图的形式进行描述,表达清晰、直观,并且便于应用计算机进行零件的分类成组。     

基于聚类算法及BP网络的零件族构造方法研究

为解决多品种小批量产品的零件成组设计与加工的问题,提出了基于编码的K-means聚类算法和有效性指标的已有零件成组,以及利用BP网络法实现新零件归族的方法来构造零件族的方法。建立零件聚类成组的数学模型,利用函数指标来检验零件成组的有效性并得出最佳聚类族数。利用神经网络算法来进行零件分组的BP网络训练,通过训练后的网络对新的零件进行仿真,从而实现新零件的匹配。通过实例证实该方法可准确构建相似件的零件族。     

Strategy and method of part reengineering

Reducing part redundancy because of design arbitrariness is an effective measure of improving production volume, of which part reengineering is a reasonable and effective approach. The problem was formalized, then similarity degree was measured by Euclidean proximity through considering similar rules of parts based on properties of isostructure, geometry, precision, material and performance indices. A similarity matrix could be constructed based on similarity degree, and part reengineering arithmetic was advanced based on fuzzy clustering. Finally, analysis and adjustment of part reengineering scheme was discussed. The strategy and method are effective in an instance of a motorcycle-hydraulic-disk brake disk reengineering.     

聚类程序新算法及其在油液分析软件中的应用

研究了一种新的聚类谱系图生成的程序算法,这种方法只需要计算一次距离矩阵,利用设计的排序数组,就可以实现谱系图中关键的输出位置计算。该方法能很好地集成在开发的油液分析软件中。应用该算法,程序简单高效,并能够用聚类方法来分析油液的状态,进而得知机械设备的运行状态,为油液分析方法提供了一个可行的新思路,有助于提高诊断结果的可信度。     

A neural information retrieval system

Group technology is an approach to manufacturing that attempts to enhance production efficiency by grouping similar activities and tasks together. The results of this process are then used in the execution of similar tasks and activities. This concept can be applied to a variety of activities such as design retrieval, purchasing, sales, and process planning [1]. Traditionally, classification and coding has been used to implement group technology. In this paper, however, we discuss a novel approach using neural networks, a technology noted for its powerful pattern-matching capability. Although this approach can be applied to the entire spectrum of group technology applications, we focus on an application to the retrieval and reuse of engineering part designs.Neural networks based on adaptive resonance theory (ART) are being developed for application in the retrieval and reuse of engineering designs. Two-dimensional representations of engineering designs are input to ART-1 neural networks to produce groups or clusters of similar parts. These representations, in their basic form, amount to bit maps of the design, and can become very large when the design is represented in high resolution. We describe a neural information retrieval system (NIRS) under development. This system demonstrates the feasibility of training an ART-1 network to first cluster designs into families, and then to recall a family of similar parts when queried with a new part design. This application is of large practical value to industry because it aids in the identification, retrieval, and reuse of engineering designs.     

固定盘压形冲孔落料级进模的设计与探讨

着重分析固定盘的压形冲裁生产工艺特征,重点论述工位设计、排样图设计、步距精度计算、模具结构设计及关键零件设计等关键技术。探讨和论证其排样方案选择的正确性,在设计过程中,把先压出固定盘底部凹坑作为设计重点,从而解决了材料局部流动问题;为避免已成形的工件变形,采用双斜面的凹模刃口。通过生产实践证明：工艺方案合理,模具结构安全,性能可靠,被加工工件质量好,生产效率高。这对此类零件的级进模设计有着极其重要的参考价值。     

Formation of machine groups and part families in cellular manufacturing systems using a correlation analysis approach

The important step in the design of a cellular manufacturing (CM) system is to identify the part families and machine groups and consequently to form manufacturing cells. The scope of this article is to formulate a multivariate approach based on a correlation analysis for solving cell formation problem. The proposed approach is carried out in three phases. In the first phase, the correlation matrix is used as similarity coefficient matrix. In the second phase, principal component analysis (PCA) is applied to find the eigenvalues and eigenvectors on the correlation similarity matrix. A scatter plot analysis as a cluster analysis is applied to make simultaneously machine groups and part families while maximizing correlation between elements. In the third stage, an algorithm is improved to assign exceptional machines and exceptional parts using respectively angle measure and Euclidian distance. The proposed approach is also applied to the general group technology (GT) problem in which exceptional machines and part are considered. Furthermore, the proposed approach has the flexibility to consider the number of cells as a dependent or independent variable. Two numerical examples for the design of cell structures are provided in order to illustrate the three phases of proposed approach. The results of a comparative study based on multiple performance criteria show that the present approach is very effective, efficient and practical.     

基于零件方位特征建立工艺模型

建立零件工艺模型是CAPP系统运行的前提条件.一般的CAPP系统使用的零件工艺模型无法满足CIMS系统的需要,在CAD与CAPP之间产生的信息断层使得集成工艺设计自动化迟迟不能解决,建立一个统一的CAD/CAPP零件工艺模型是非常重要的.基本实体 方位特征描述法是一种既适合于回转体零件也适合于非回转体零件的零件工艺模型,该模型为计算机处理零件信息提供可靠基础,它不但能满足工艺过程设计要求,而且为CAD/CAPP/CAM的集成奠定了基础.     

Learning group-technology part families from solid models by parallel distributed processing

This paper presents a new approach to part classification in group technology. It advocates the introduction of a feature-based solid-modelling scheme for part representation which, in turn, helps in identifying features of interest. The extracted features of the part are then used to determine the part family to which the part belongs. A parallel distributed processing (PDP) model has been utilised in developing a learning module for the part-classification problem. The proposed model has been implemented in the Unix environment of a Sun work-station. The usefulness of the proposed model has been validated with an example of 16 parts in three part families.     

Intelligent support system for continuous improvement in a dynamic manufacturing environment

In a dynamic competitive environment where consumer demands are getting more sophisticated and technology advancement is very fast, a firm that can quickly improve its workforce's skill level to match the introduction of advanced manufacturing technology will gain a competitive edge. This paper discusses how expert system technology should be modified and integrated with conventional analytical techniques like statistical estimation, pattern recognition, parameter estimation, etc., to provide on-line assistance to less experienced production operators, as well as to provide off-line support of continuous learning and the accumulation of experiences. A new paradigm that integrates the fault-tree (AND-OR graph) and the flow chart is developed. It is shown how such a paradigm can be applied to various applications in the production process, and how it can play an important role in the continuous learning process which will increase the speed of the workforce's skill improvement and production performance.     

Automatic generation of group technology families during the part classification process

Initial part family formation and subsequent part classification are two important problems to be addressed in applying the group technology principle. Although these two problems are closely related, they have been treated separately. As an aggregate problem, the automatic creation of new part families during the classification process, is investigated. A two-layer neural network using the adaptive resonance theory is adopted. The capability of this neural network model of dealing with the stability-plasticity dilemma is utilised in classifying the parts into families and creating new families if necessary. A heuristic algorithm using the neural network is described, with illustrative examples.     

选择性激光烧结新型扫描方式的研究及实现

选择性激光烧结（Ｓｅｌｅｃｔｉｖｅ　ｌａｓｅｒ　ｓｉｎｔｅｒｉｎｇ，ＳＬＳ）快速成形技术是近１０年来发展起来的一种先进制造技术，采用此项技术可以显著地缩短产品投放市场的周期，降低成本，提高产品质量，增强企业的竞争能力。而扫描方式是ＳＬＳ成形过程中的一项关键技术，在分析了现有扫描方式及其对ＳＬＳ制件精度、强度以及成形速度影响的基础上，指出了现有扫描方式的缺陷。为了克服这种缺陷，特提出并实现了一种新的分区变向扫描方式及其分区算法，它通过将扫描线段进行分组，以避开截面内孔和凹槽。这种新的分区变向扫描方式及其算法已经成功地应用到华中科技大学开发的ＨＲＰＳ系列ＳＬＳ系统中。实际应用表明：这种新型扫描方式能大幅度地提高烧结成形效率和减小烧结体的翘曲变形量。