Development of an Intelligent Apprentice System for Extrusion Die Design and Process Simulation

Document Type

Conference Proceeding

Publication Date

1985

Find this in a Library

Catalog Record

Abstract

The costly build-and-test methods used by the experts today for optimizing extrusion die design increase lead time before production which, in turn, adversely affects operation of the aerospace factory. The success of such designs is strongly dependent upon having an experience base available. Also, the recent work on the computer aided engineering (CAE) approach to extrusion of difficult-to-extrude materials, carried out by the current investigators, reveals that the build-and-test methods are not feasible for extrusion of new aerospace alloys. To overcome these difficulties a prototype intelligent apprentice system was developed using an approach which synthesizes the available techniques from software engineering (SE), data base management systems (DBMS), operating systems (OS), and artificial intelligence (AI) to exploit the power of existing analytical techniques with the help of heuristic rules. The system offers a potential capability for prompting and aiding the design engineer in his task of finding effective solutions to complex problems. In this paper, the extrusion die design criteria and methodology, the approach and the various steps used in the development of the system, and the results of the validation are discussed. The results clearly indicate that the current engineering approach to using AI is more practical and beneficial for solving immediate problems in automation of die design than a pure AI approach.

Comments

Presented at the 1985 ASME International Computers in Engineering Conference and Exhibition, Boston, MA.

Catalog Record

Share

COinS