Chatter prediction in boring process using machine learning technique

Abstract

Chatter is the main reason behind the failure of any part in the machining centre and lowers the productivity. Chatter occurs as a dynamic interaction between the tool and the work piece resulting in poor surface finish, high-pitch noise and premature tool failure. In this paper, the chatter prediction is done by active method by considering the parameters like spindle speed, depth of cut, feed rate and including the dynamics of both the tool and the workpiece. The vibration signals are acquired using an accelerometer in a closed environment. From the acquired signals discrete wavelet transformation (DWT), features are extracted and classified into three different patterns (stable, transition and chatter) using support vector machine (SVM). The classified results are validated using surface roughness values (Ra). [Received 12 August 2016; Accepted 19 May 2017]