Publication

Abstract : Titanium alloys have been extensively utilized in the aerospace and biomedical industries because of higher corrosion resistance and their good strength to weight ratio. In spite of several advantages, titanium alloys are difficult to machine because of their poor thermal conductivity and high chemical reactivity. Identification of suitable machining conditions is the key to get the good surface finish. Chatter during machining brings adverse effects in surface quality, dimensional accuracy and in tool life. The objective of this work is to identify chatter free machining conditions for machining titanium alloys and to predict the chatter occurrence with the help of machine leaning algorithms. During machining of titanium alloy, the vibration signals are captured for various machining conditions using accelerometer. From the raw signal statistical features are extracted and decision tree algorithm is used in selecting the dominant features. By monitoring the dominant features, chatter occurrences are predicted using Decision Tree (DT), Artificial Neural Network (ANN) and Support Vector Machines (SVM).