冲压模具智能设计中的知识发掘及重用技术研究机械制造及其自动化专业论文.docxVIP

AbstractAcquiring Abstract Acquiring and reusing the die design knowledge in an integrated way to reduce the development cost and shorten lead time is the demands of this knowledge age on the stamping industry．It’S also the key to improve the core competitive power for a die—making corporation and the technical basis for innovative design． After reviewing and analyzing the evolution of the intelligent design for stamping，the author believes that it’S of theoretically and practically importance to dig into the field of die design knowledge mining and enhance the knowledge sharing and communication by making the best of latest AI and Web technologies．This paper is composed of three parts：the first one is the technology of mining tacit knowledge from die designs；the second one is the creation and application of a uniform model representing various sources of die design knowledge and the last one is the research and development of a web—based KBE system for die design． Firstly，the paper introduced the fuzzy inference theory for making the tacit die design explicit．Based this theory，the fuzzy rules were created using the similarities between stamping parts and die design，and the belief degree of the rules were decided by the degree of similarities between the stamping parts in one fuzzy set of similar stampings． To widening the sources of knowledge，a uniform knowledge representation model for stamping(UKRMS)was created．The model was aimed to represent all kinds of knowledge from textbook，drawings， journals and CAD models of different platform．In this model，the knowledge was separated from its information carrier and consequently the database and AI technologies could be fully utilized to improve the efficiency and extent of using knowledge． The feature·based stamping model is the foundation and the key to success of knowledge mining．After carefully studying the effect of feature attributes on the die design，a knowledge mining oriented model and its key characteristi