A Multi Layer Perceptron (MLP) Based on Back Propagation Technique for Flank Wear Analysis of Single Point Cutting Tool Inserts

摘要

Wear of cutting tools is very important in manufacturing industry. Study and analysis of each tool is extremely useful in the tool condition monitoring process. Determination of optimal cutting conditions and prediction of expected tool life is gaining momentum. Various measurable parameters and their correlative results are essential in taking steps to make the tool health monitoring process on-line and minimize downtime in productive environments. Condition monitoring of the cutting tools is very important as the process capability and process parameters may be evaluated based on the condition of the tool. Wear of the tool tip generates poor surface finish and an unexpected tool failure may damage the tool, work piece and the machine tool. Advanced manufacturing demands an optimal machining process with proper control of machining parameters. Efforts are on to make control of most of the machining parameters on-line. Many problems that affect optimization are related to the diminished machine performance caused by worn out tools. One of the most promising tool condition monitoring technique is based on the analysis of acoustic emission (AE) signals. Generation of the AE signals in the cutting zone makes them sensitive to changes in the cutting process. Experiments were conducted on select specimens of steel using select varieties of cutting tools inserts on a high speed lathe for some specific cutting conditions. The AE signal analysis is carried out using, ring down count (RDC), rise time (RTT), event duration (EDT), energy (ENT) and peak amplitude (PA).