A Hybrid Grey-Fuzzy Technique For Optimizing Multiple-Responses In Centerless Grinding

Author(s): M. Karthikeyan, L. Praveen Kumar, N. Saran Raj, A. Sriram

Abstract: This analysis focuses on finding an optimum set of machining parameters in external centerless grinding process by adopting grey relational analysis, coupled with fuzzy logic approach. This hybrid technique is used to determine the grey-fuzzy reasoning grade from the calculated multi-characteristics grey relational grade of surface roughness and roundness error to reduce the fuzziness in output. Experiments are designed using Taguchi’s Design of Experiments (DoE), for three parameters varied through three levels an L9 (33) Orthogonal Array (OA) is selected. The optimal level values are determined from the response table and main effects plot and the individual effect of one parameter over another parameter is determined using the interaction plot. Confirmation experiment conducted with the optimum input parameters obtained from grey-fuzzy reasoning grade shows a reduction in surface roughness and roundness error values.