Statistical analysis and ANN modelling of tribological performance of aluminium hybrid composites fabricated by PM method

This study represents the wear behaviour of the aluminium hybrid composites developed through powder metallurgy. The pure Al-matrix powder was blended with Si3N4/ZrO2 ceramic powders in a high-centrifugal ball mill and then compacted at three different compaction pressure. The sintering of the composites was done in an environment-controlled tube furnace at 500 degrees C for 1 h. The effect of wt% of the reinforcement, compaction pressure and applied load was investigated on the wear results performed on the pin-on-disc setup. The Analysis of variance (ANOVA) and artificial neural network (ANN) was used for the wear loss results prediction. The ANOVA results revealed that load is a highly influential parameter for wear loss followed by reinforcement and compaction pressure. Furthermore, an ANN model using feed-forward backpropagation analysis was built which found that the predicted and experimental results are close.

Automated summary

This study investigates the wear behavior of aluminium hybrid composites developed through powder metallurgy. The effects of reinforcement content, compaction pressure, and applied load on wear results were analyzed. The study utilized Analysis of variance (ANOVA) and artificial neural network (ANN) for wear loss prediction, finding that load is the most influential parameter followed by reinforcement and compaction pressure.