A Data-Dependent Systems Approach to Optimal Microcomputer Control Illustrated by EDM

Abstract

A new approach to microcomputer control of manufacturing processes is proposed and illustrated by application to optimal control of electro-discharge machining (EDM). Based on a recently developed methodology called Data Dependent Systems (DDS), it is capable of providing off-line modeling and analysis under working conditions to suggest the best possible strategy for the existing design, an on-line optimal control algorithm, adaptable fine-tuning of the controller to enhance its effectiveness, and finally, design improvements aimed at further increase in productivity and quality. All these features are illustrated by means of DDS models for pulse current and voltage in EDM, one of the least understood and most difficult to control manufacturing processes.