ANN-RSM synergistic methodology for optimizing machining performance of biocompatible Titanium alloy: A machine learning approach

Non-traditional machining processes are widely used for machining hard and biocompatible materials with high dimensional accuracy. The selection of optimum process parameters is essential to achieve improved surface quality and mechanical properties. The present study investigates the optimization of machining parameters for the biocompatible Ti-6Al-4V alloy using a hybrid Response Surface Methodology-Artificial Neural Network (RSM-ANN) approach. The optimized machining parameters resulted into minimum surface roughness (2.13 µm) and maximum micro-hardness (1657.06 HV). The surface morphology was analysed using Scanning Electron Microscopy (SEM), X-ray Diffraction (XRD) and image analysis software. The microstructural features such as micro-cracks, craters, pits, holes, and recast layer were analysed to evaluate the surface integrity. The image analysis software results revealed that the recast layer occupied a projected area of 311.4 µm 2 (53.32%). Moreover, under higher discharge energy, the thermally affected region occupied the largest projected area of 381.5 µm 2 (65.46%) resulted into wider distribution of the recast layer. Furthermore, the 3D surface topography over a scanned area of 29.29 × 19.78 µm 2 exhibited a maximum surface height of 11.71 µm. This indicated increased crater formation, molten material redistribution and recast layer development. The developed ML-RSM regression models achieved R 2 values greater than 0.95, indicating good agreement between the experimental and predicted results.