Research On Ball Screw CAPP System Based On Feature Coding And Production Rule

Ball screw is widely used in aviation, aerospace, military, medical, automotive and other high-tech industries as a kind of typical precision drive components, but the defect of the traditional craft of ball screw is long design cycle, high cost and high labor strength, it is difficult for modern enterprises to adapt to the new trend of product diversification, high quality and high efficiency. So the ball screw CAPP system is designed based on feature coding and production rule, the derived and generative process of ball screw is automatically generated, the process design cycle is shorten, the work intensity is lowered.This article firstly introduced the background and significance of the research and the develop level of domestic and foreign, the framework and workflow of the ball screw CAPP system is designed through analyzing the function of the system.The database function and the data model of ball screw CAPP system is analyzed, the two-dimensional relational database is established, the physical of database is also designed. The connection between VB and database is realized by setting up ODBC data source and using the ADO Data control, the display, addition, modification, deletion and sort of the ball screw mechanical process is realized through making use of the static and dynamic Access methods synthetically.The feature coding module of ball screw is designed, the similarity coefficient calculation is realized on the basis of the coding; the 3D parametric modeling is completed on account of the secondary development of SolidWorks based on dimension driven principle. The task time of ball screw are analyzed; The production rules is expressed by using the two-dimensional relational database, the rule element tables and element relational tables are established, a traversal query is proposed to lay the theoretical foundation for the generative reasoning.A friendly human-computer interaction is designed. Through programming and running instance, the ball screw process retrieval is achieved. If the retrieval is failing, the derivation process is achieved on the basis of similarity coefficient by encoding for the failing instance. If the derivation is failing, the feature information should be input and saved, and according the feature information, the generative process reasoning is achieved based on production rules.