Genetic algorithm based design optimization of crashworthy honeycomb sandwiched panels of AA7075-T651 aluminium alloy for aerospace applications

Abstract

Crashworthiness studies, for decades have been an area of tremendous interest in aerospace and automotive sectors, as ensuring the safety of the occupants in the vehicle is a primary goal. Honeycomb sandwich panels are generally used for building floor, ceiling and other important fuselage parts in a commercial airliner to make the structures shock-resistant. In this work, optimization of the design of crashworthy sandwiched honeycomb structure was carried out using a non-dominated sorting genetic algorithm (NSGA) . The stated algorithm had actively monitored the variation of the geometric parameters. To study the shape effect of the cells which absorb impact energy, three cellular shapes namely hex-grid, ortho-grid and iso-grid structures were considered for analyzing their deformation behavior under impact loading. A multi objective function in the genetic algorithm was implemented in MATLAB, where the objectives were to minimize weight and to maximize the absorbed impact energy while constraining the maximum transmitted force and displacement due to crushing. Sandwiched aluminum (AA7075-T651) honeycomb structures were modeled in ABAQUS using python scripts , with the number of core cells and the wall thickness of each core cell being optimized through genetic algorithm in MATLAB using an interactive mechanism. Significant improvements were observed in terms of reduction of weight and increase in the ability to absorb the specific energy during crushing in the sandwiched panels after 450 iterations of optimization cycles. A novel methodology in the domain of crash analysis of aerostructures was established with design optimization of honeycomb sandwiched panels.