TOOL WEAR PREDICTION IN MACHINING AISI 1040 STEEL UNDER MQL

Abstract — Productivity and quality are critical issues in any manufacturing industry. In order to minimize the idle time associated with tool setups, estimation of tool wear is important. Several models are available in literature to estimate and predict tool wear, however, very few papers are available on machining using cutting fluids. In the available literature, flood lubrication is generally considered. In view of stringent environmental restrictions and drive towards green manufacturing, Minimum Quantity Lubrication (MQL) is gaining momentum. But, since the studies are in incubating stage, not much literature can be found on tool wear estimation in MQL. In the present work, AISI 1040 steel was machined and different parameters like cutting forces, surface roughness, temperatures, etc. were recorded. The data was used to build two models, viz., mathematical regression and Artificial Neural Networks (ANN). The models were tested to estimate their efficacy. It was found that ANN model could predict tool wear with less than 8% error