Network Topology Model for Wear Behavior Prediction of Ti6Al4V Clad Magnesium Substrates

Abstract

In this work, Ti6Al4V alloy was clad on commercially pure magnesium substrate by laser cladding. The laser cladding parameters were laser scan speed and powder feed rate. Wear behaviors of the clad substrates were evaluated by a dry sliding wear testing on Ti-6Al-4V counter material by pin-on-disc method. The wear testing parameters like applied load and sliding velocity were varied by keeping sliding distance as constant. The experimental data were used to develop an optimized neural network model. Network models predicted that the clad deposition affects the wear behavior. The combination of lower laser scan and highest powder feed yields higher deposition compared to higher laser scan and lowest powder feed combination. The higher material deposition improves the wear resistance of the substrates. The experimental data confirms the network models prediction by showing the lower wear rate at 200 mm/min–10 g/min laser parametric condition which has higher clad deposition than lower wear rate at 300 mm/min–5 g/min condition with lower materials deposition. Overall, laser cladding of Ti6Al4V on commercially pure magnesium improves the wear resistance by ~75%.