Experimental investigation on mechanical properties of reinforced Al6061 composites and its prediction using KNN-ALO algorithms

Metal matrix composites (MMCs) are widely practicing material for improving the surface property. Stir casting is one of the most straightforward processes of manufacturing MMC and attains higher advantages on material processing cost, more comfortable handling of material, size, design and excellent stability of matrix structure. In this research work, MMC of Al6061 with blended MgO and Si3N4 composite mixtures is produced using stir casting process. One of the factors affecting the material homogeneity in the casted material is the tensile rupture, where the proposed composite material subjected to tensile stress and yielding. The structural property of the material tested under universal testing machine and Brinell hardness tester. This paper proposes a novel hybrid approach to evaluate the tensile property of composites. The prediction of the tensile property of the MMC performed by the K-nearest neighbour (KNN) algorithm and ant lion optimisation (ALO) algorithm, which is numerically modelled and experimented in the running platform of MATLAB and compared with decision tree (DT) classifier algorithm for better performance outcome. Predicted test results show that the proposed KNN-ALO is an efficient method for predicting the tensile and hardness properties of stir cast aluminium composites.