Composites communications on AW2024/E-glass/TiC hybrid material by data-driven technique using linear regression model

Soft Computing tools gaining a popular and effective decision-making technique for most of the complicated problem involved in engineering application. It also provides prediction of experimental tested data by eliminating traditional trial-error method which consume more time and cost. Herein, this paper makes use of Linear Regression (LR) model to predict the mechanical properties such as hardness and yield strength of AW2024/E-glass/TiC hybrid composite. Highest hardness and yield strength was found to be 73.4 VHN and 161.2 N/mm 2 for AW2024/6%TiC/5%E-Glass hybrid composite. It is seen that approximately 58% improvement in hardness and yield strength is noticed. Taguchi’s L16 orthogonal array is used with two factors (E-glass and TiC) and its 4 level each. It is examined that the titanium carbide has major significance for both hardness and yield strength properties. Furthermore, microstructure, residual plots and visualization of fracture specimen were discussed in detail. • Study on the AW2024/E-glass/TiC hybrid composite has been done by fabrication of stir casting route by varying 1%, 3%, 5% and 7% E-glass and 2%, 4%, 6% and 8% TiC. • Addition of 6% of TiC content on AW2024/5%E-glass mixture contributed significant improvement of 58% of hardness and yield strength. • Microstructure study, revel the clear grain boundaries, agglomeration, particle separation, surface scratches and uniform dispersion of secondary reinforcement. • Regression analysis revels that the average of percentage error found to be 3.48% and 3.38% for hardness and yield strength of AW2024/E-glass/TiC hybrid composite. • Midst fracture with clear cleavage pattern, facets surface and rough fine grain dimple structure at nearby location along with inclusions for AW2024/E-glass composite. • AW2024/E-glass/TiC hybrid composite indicated hard-rough dimple fracture resulting ductile type of failure.