Optimization using genetic algorithm of tribological behaviour of WC tool material

Abstract

In this investigation we have used a heuristic approach to optimize the process parameters in terms of tool wear rate. We have used the L8 orthogonal array design of experiments with three input parameters set at two levels. We have carried out the experimentation on two different processes viz. dry sliding and dry turning processes. An attempt has been made to achieve and validate the results obtained from these processes to check the repeatability of values in the same experimental environment. The tool material chosen for tool insert is Tungsten Carbide which is used in the manufacturing industries. We have optimised the results obtained on tribometer under the dry sliding process through a modern optimization technique i.e. genetic algorithm. The response surface methodology model (L8 orthogonal array) formed the basis for the development of genetic algorithm model through which we have defined the conditions. We have used the conditions of minimum tool wear for turning process, minimum coefficient of friction and minimum surface roughness for sliding process on a pin-on-disc test rig. It has been inferred that the sliding and turning processes under the conditions of no lubrication yielded analogous results. We have verified the same results practically by performing confirmation experiments on lathe machine for turning operation under the same experimental conditions.