DEVELOPMENT OF AN ANN MODEL TO PREDICT SURFACE ROUGHNESS DURING CRYOGENIC MACHINING OPERATION

Abstract

This research paper deals with the advanced manufacturing technique which can be achieved by improving machining parameters through cryogenic cooling approach instead of conventional cooling Industrial growth and globalization aim to increase the material removal rate maintaining very good surface finish and high machining accuracy even for harder, difficult to cut materials. Also, it should be possible at lower overall cost as well as maintaining an eco-friendly environment. Cutting tool fails due to high-elevated temperature and plastic deformation. Also the material is removed due to ploughing and rubbing instead of ideal shearing under conventional cooling method. Moreover, for ductile material cutting, heat causes a stickiness which produces material build-up on the cutting edge. This results a bad odor, smoke, health hazards and water penetration into the machine bearings. To improve the surface roughness i.e. to maintain the tool’s form stability for longer time, cutting zone temperature is to be maintained low. Liquid nitrogen used as coolant can reduce the cutting zone temperature to a greater extent and thereby helps to retain the shape of cutting edge of cutting tool for a longer time. Also HSTR alloys, MMC may also be machined economically because of low tool wear rate and easy chip breakability. In this present problem, an ANN model has been developed to predict tool wear during cryogenic machining using back propagation feed-forward network algorithm and four hidden layers feed forward architecture.