Publication

Abstract : Magnesium alloys are well known for their light weight and low density, which facilitates their utilization in a wide range of engineering applications. When softer magnesium alloys are reinforced with hard ceramic particles, the machinability of magnesium alloys is expected to improve. In the current work, limestone powder (LSP) was used as a bioceramic reinforcement in AZ31 magnesium alloy, and the composites were fabricated through gravity stir casting technique. LSP particles were characterized by x-ray diffraction (XRD). Magnesium composite specimens were initially subjected to tensile tests to identify the best composite configuration. Composites with 6% LSP possessed high tensile strength and only these composites were subjected to drilling analysis. Generalized regression neural network (GRNN) has been used to forecast the surface roughness of AZ31 magnesium alloy composites during drilling. Outputs of drilling operations were utilized to train the neural network model and the values predicted by the GRNN technique were ascertained using confirmation experiments. It was concluded that bioceramics improved the machinability of the magnesium alloy composites and the GRNN application towards prediction of surface roughness showed satisfactory results exhibiting less deviation from the experimental results even for any value considered within the limits of the parametric levels.