A new method for online monitoring when grinding Ti-6Al-4V alloy

Abstract

This study proposes a grinding process from a control perspective to improve surface quality and precision in the machining process and speed up intelligent production. This method provides a cognitive decision-making process. On this basis, a new approach was developed to measure the amount of grinding wheel wear (GWW) and predict surface roughness (SR) in accordance with the compressed air measuring head and hybrid algorithms fuzzy neural networks (ANFIS)–Gaussian regression function (GPR) and Taguchi empirical analysis. A series of experiments were conducted in various processing conditions. The results showed the efficiency of the grinding process in measuring the amount of GWW and predicting the SR of the Ti–6Al–4V alloy accurately. The proposed model is able to predict SR values with 99.69% precision and 98% confidence interval. This study laid the foundation for monitoring GWW and SR in actual industrial environments.