Surface roughness prediction of ground components using a fuzzy logic approach

Abstract In this paper a total of 16 variables, which are most influential on surface roughness in grinding, are considered. The variables are classified into three groups depending on their significance and effect on surface roughness. A three-layer fuzzy model is used to correlate these variables to surface roughness using the fuzzy rules generated based on experimental observations and recommendations from wheel manufacturers. Membership functions, fuzzy rule bases, and a worked example are presented in detail to demonstrate the strength of fuzzy logic in modeling such a complex system in an efficient manner.