The demand for production of fine micro-grooves with least heat-affected zone (HAZ), recast layer thickness, microcrack, and microcavity are increasing as Ti6Al4V is a readily used orthopedic implant material. This study presents a groove quality optimiser called response surface methodology (RSM) for fiber laser micro engraving on Ti6Al4V to improve microgroove quality and functionality. The critical process factors' (number of laser beam passes, scan speed, and pulse frequency) for each response characteristic are identified through the RSM optimiser by analysing the experimental data sets. The optimised data sets are beneficial in maintaining groove quality characteristics (width, depth, HAZ, recast layer thickness, and microhardness). Number of passes and scanning speed has been recognised as crucial process parameters that primarily regulate the groove response characteristics. A 3-20-5 neural network (NN) model is developed and trained through the experimental input-output data sets. Performance of 3-20-5 model is found as 1.63e(-27), and regression score is significantly closer to 1. It has been recognised that ANN model predicts more accurate results with a low error percentage than RSM model even when the input factor level settings are beyond the defined boundary. Under these experimental settings, the microgroove characteristics are enhanced in terms of low HAZ, recast layer thickness, microhardness, and spatter. X-ray diffraction analysis (XRD) is also performed on base metal and grooved surfaces to determine diffraction peaks, crystallite phase composition, size, and lattice strain. XRD analysis revealed the presence of both HCP-alpha-Ti and BCC-beta-Ti in both base metal grooved surfaces.
展开▼
