An On-line Process Control Based on Geometrical Parameters from Measured Data in Machining Process

Abstract

This paper is intended as a development of new process control method of making high quality of products in machining process Conventional process control methods are for off-line control and cannot be determined optimal control limits according to cutting state in machining process. The proposed method is applicable to on-line process control and determination of optimal control limit according to cutting state of machining process in short time. By introducing new parameter of actual measured geometrical deviation (AMGD), it becomes clear that it can obtain control limit based on geometrical relation. The proposed method is constructed reasoning networks by learning fuzzy neural network and generalize the control limit by fuzzy reasoning. The control limits of geometrical tolerances are shown by simulation.