Surface Roughness Prediction in Grinding Ti using ANFIS Hybrid Algorithm

Stephen, Deborah Serenade; Sethuramalingam, Prabhu

Abstract

Intelligent manufacturing is needed, and many techniques and tools have been developed with this in mind. Over time, many of these techniques have been combined, and hybrid approaches have provided better results in shorter times, leading to a more precise prediction of outcomes when compared to the use of individual tools. This research focused on grinding Ti-6Al-4V workpiece material with a Carbon nanotube (CNT) incorporated grinding wheel. The Adaptive Neuro-Fuzzy Inference System (ANFIS) was used to predict surface roughness which was taken as the output of choice for this study. A new hybrid of ANFIS with Genetic Algorithm (ANFIS-GA) was then proposed to see if this prediction method could obtain greater precision. The regression analysis predicted the experimental model’s linear relationship to surface roughness, and the effect of grinding process parameters on surface roughness was analysed using the sensitivity analysis method.

Keyword(s)

ANFIS, CNT Grinding wheel, Fuzzy logic, Regression analysis, Sensitivity analysis, Taguchi analysis

Full Text: PDF (downloaded 506 times)

Refbacks

  • There are currently no refbacks.
This abstract viewed 806 times