零件分类成组系统设计

论文总字数：21185字

摘 要

本文简要地介绍了计算机辅助制造技术及成组技术，系统地介绍了零件分类成组的主要方法。并且采用单链聚类分析法和排序聚类分析法两种方法，用C 语言设计出了一个零件分类成组MFC程序，输入为零件-机床信息，输出为零件分组信息。

程序在算法实现的过程中进行了创新。在单链聚类分析法中，在程序运行前预先输入一个相似系数，在程序运算过程中判断相似系数是否达到分组的要求，从而避开了生成聚类树状图这一繁琐复杂的过程。在排序聚类分析法中，创建一个函数判断矩阵每次行列变换后是否与上个矩阵相同，从而巧妙地解决了程序何时跳出行列变换循环的核心问题。

最后通过大量数据分析实验，得出结论，当相似系数为0.5时，单链聚类法语排序聚类法分类结果基本一致，当相似系数越大，单链聚类分析法分的组数越多，相似系数越小时分的组数越多。而排序聚类法分组数与给定相似系数没有关联。所以，一般情况下单链聚类法和排序聚类分析法都可以满足分类要求，当对分组数有具体要求时，可以选择单链聚类分析法，通过调整相似系数，得到更加符合生产要求的分组。

关键词：计算机辅助制造；成组技术；零件分类；聚类分析法

Parts classification group system design

02012129 Gang liu

Supervised by Jie cheng

Abstract： Group technology is a kind of idea in manufacturing industry. it uses the similarity of parts to divide the parts together, to achieve the higher level of integration of design and manufacturing. The classification of parts is one of the keys to group technology. The development of computer aided group technology, makes the parts grouping can use computer software and hardware technology assisted, gradually developed the production flow analysis method, coding and classification method and potential function method fast, reasonable and accurate classification of parts grouping method.

This paper briefly introduces the computer aided manufacturing technology and group technology, and introduces the main methods of parts classification. And using single linkage cluster analysis method and rank order cluster analysis method, using C to design a part classification group MFC program, input the parts - machine information and output the parts grouping information.

Program in the process of the realization of the process of innovation. The chain clustering analysis method, before running the program pre input a similarity coefficient, in the process of program operations to determine similarity coefficient is to achieve the requirements of packet so as to avoid the generated dendrogram the cumbersome and complex process. In order to create a function to determine whether the matrix is the same as the last matrix, the core problem of the program when the program will jump out of the loop is solved.

Finally through a large amount of data analysis experiment, draw the conclusion, when the similarity coefficient of 0.5, single linkage cluster analysis method and rank order cluster analysis method computational classification results are basically the same, when the greater the similarity coefficient, single linkage cluster analysis method of the group number, similar coefficient is small of the time group number. And there is no correlation between the number of clusters and the given similarity coefficient. So, generally single clustering and ordination and cluster analysis method can meet the requirements of classification, when the number of groups have specific requirements, you can choose single linkage cluster analysis method, by adjusting the similarity coefficient, more in line with the requirements of the production group.

Key words: Computer-aided manufacturing；Group Technology；Parts classification；Chaser analysis

目 录

一、引言 6

1.1 CAM简介 6

1.2 成组技术简介 6

1.3 零件分类成组的意义和目的 7

1.4 国内外研究现状 7

1.5 课题研究内容 7

1.6本章小结 7

二、零件分类成组方法 7

2.1 视检法 7

2.2生产流程分析法 8

2.2.1关键机床法 8

2.2.2顺序分枝法 9

2.2.3聚类分析法 11

2.3 编码分类法 12

2.3.1编码分类的原理 12

2.3.2编码分类的方法 12

2.4本章小结 13

三、零件分类成组系统设计与分析 14

3.1 系统需求分析 14

3.2 系统结构与功能 14

3.3 分类方法的选择和设计 15

3.3.1软件概况 15

3.3.2程序开发 15

3.4 系统测试 25

3.5本章小结 29

四、结论 30

4.1系统质量评价 30

4.2总结与展望 30

致 谢 31

参考文献： 32

一、引言

剩余内容已隐藏，请支付后下载全文，论文总字数：21185字