Optimization of Cutting Conditions in Milling Process to Minimize Tool Wear Using Selected Soft Computing Techniques: A Comparative Study

Abstract

The life of a tool is very significant in metal cutting because production time is a waste whenever a tool is replaced or reset. Because of prolonged use, cutting tools blunt and their effectiveness decreases over time. It is necessary to replace, index or resharpen, and reset the tool after certain time. Tool life is a measure of the length of time a tool will cut effectively. The life of a cutting tool is mostly affected by many factors, such as, the tool wear, material removal rate, microstructure of the material being cut, setup rigidity, and cutting fluid quality. In this paper, up-milling and down-milling are compared with respect to different input parameters mainly cutting speed, feed, and number of teeth, rake angle, and helix angle. A relationship between tool wear and these input parameters is established by using regression analysis and soft computing techniques are applied to minimize the tool wear. Five selected soft computing techniques viz. (i) Genetic Algorithm (GA), (ii) Simulated Annealing (SA), (iii) Hybrid GA with Pattern Search (PS), (iv) Particle Swarm Optimization (PSO), and (v) Threshold Acceptance Algorithm (TAA), are used for optimization of tool wear. The results obtained from each of them are also analyzed. The effect of input parameters on tool wear, the way it is different for up-milling and down-milling is addressed for each case with the help of five selected soft computing techniques. The results indicate that the tool wear is least in case of down-milling when compared to up-milling and hybrid GA gave the best results.