Fabrication of Al8090 foam via friction stir processing: Compressive behaviour analysis and parametric optimization using Taguchi's approach with artificial neural network-improved Grey Wolf optimizer validation

Abstract

Metal foam is one of the high-strength and low-weight materials having high energy absorption capacity during the impact because of its porous structure. These properties make it particularly useful in the automobile sector. There are various techniques for the fabrication of metal foam but in recent years friction stir processing (FSP) grabbed more attention for the development of metal foam. In the current study, Al8090 is taken as a parent material and titanium hydride is considered as foamable and powdered alumina (Al₂O₃) is taken as stabilizing agent. Mixing of the reinforced particles has been controlled by the tool rotational speed, tool traversing speed, tilt angle, heat treatment temperature and time. The quality and uniformity of the porous material depends upon the process parameters of the FSP. Optimization of process parameters is done by using design of experiment in which Taguchi L27 orthogonal array is used. A special-purpose fixture is designed to produce tilt in the base plate while doing FSP on a vertical machining centre. Artificial neural network and improved Grey Wolf optimizer technique have been used for the validation of the results. The mixing of reinforced particles enhances compressive strength and energy absorption capacity of the aluminium alloy. Compressive strength tests have been performed for the metal foam and its value ranges between 11.2 MPa and 37 MPa. Scanning electron microscopy and energy-dispersive spectroscopy tests are performed to validate the mixing of the foamable particles in the base material.