Application of Taguchi OA array and Artificial Neural Network for Optimizing and Modeling of Drilling Cutting Parameters

摘要

This paper consists of two phases, in first phase experiments have been conducted with the use of Taguchi FFE (Fractional Factorial experimentation) to find optimal cutting process parameters spindle speed, feed rate and drill diameter with the objective of minimum delamination. Using these data ANN (Artificial Neural Network) based model has been developed for drilling induced delamination while drilling GFRP in the second phase. Delamination in drilling process is important aspect so there is need to minimize delamination while drilling. In the first phase experiments have been conducted by varying spindle speed, feed rate and drill diameter all at three level using high speed twist drills. Using L-9 of FFE, optimal levels of factors have been found out. Three more number of additional experiments has been carried out with input variable values near to obtained optimal levels through Taguchi OA array. Using these total twelve set of experiments (i.e nine + additional three) input-output data ANN has been trained to develop a model in the second phase. Developed model may suit manufacturer to find the lower delamination factor among available choice of process parameters. This developed model has been validated and found to be suitable for predicting delamination for a given spindle speed, feed rate and drill diameter. The R value for training, validation, and test curve was found to be very good.