• Johnson, Nevan Nicholas
• Ali, Syed Mansoor
• Sahu, Jambeswar
• Anand, Suya Prem
• Srivas, C. S.
• Madhavadas, Vaishnav

Abstract

<jats:title>Abstract</jats:title><jats:p>The surface finish of ground samples is highly influenced by the grinding parameters, grinding conditions and the type of grinding wheel. This paper emphasizes on the effect of various grinding factors such as the grinding conditions, the type of grinding wheel and operating process parameters like depth of cut and table speed on the surface roughness of the ground samples. Two types of grinding wheels alumina (Al<jats:sub>2</jats:sub>O<jats:sub>3</jats:sub>) and cubic boron nitride (CBN) were used for grinding AISI D3 tool steel under dry and wet conditions. The material removal rate and surface roughness were evaluated for all the ground samples. The results showed that wet grinding outperformed dry grinding and provided a better surface finish while using both grinding wheels. Machine Learning was implemented to optimize the grinding parameters. Multi-objective optimization using genetic algorithm was done and a Pareto frontier chart was made to help determine what values for the input parameters would achieve the required outputs such as material removal rate and surface roughness. Two different approaches Genetic Algorithm and Principle Component Analysis were then compared for multi-objective optimization. The type of grinding wheel used had a dominant effect on the surface roughness of ground samples while the depth of cut had a lesser effect.</jats:p>

Topics

• impedance spectroscopy
• surface
• nitride
• tool steel
• Boron
• machine learning
• wet grinding