A KNOWLEDGE BASED SYSTEM FOR WELDING PROCEDURE- GENERATION OF 2.25 Cr-lMo STEEL

Abstract

There has been a rapid growth over the past few years in the development of Knowledge Based Systems for engineering applications. Welding as a separate branch of engineering has got its recognition as a high technology.- Recently a great potential of the knowledge based systems in the field of welding is realised. The main objective of this work was to utilize artificial intelligence techniques in formulating welding procedure. The -various variables affecting the welding procedure were identified in detail and accordingly a methodology was devised for the selection of welding process, consumables, edge preparation, preheat temperature, postweld heat treatment etc.. The relevant knowledge has been gathered from different welding experts from industry and selective literature. This knowledge based approach has been utilized for the development of a knowledge based system -named "CROMOEX",-for welding procedure. generation for 2.25Cr-1Mo steel. Welding defect avoidance module has provided an additional diagonostic nature to the CROMOEX. It has been found that the advice offered by CROMOEX is theoretically justifyable and-practically interacts suitably with different modes of ,the problem domain i.e. welding procedure generation of 2.25Cr--1Mo steel. The system results are in-good agreement with the welding procedures established by the experts in the industry.