Multi-objective optimization of circular aluminum/ composite hybrid tubes under axial impact load: as mechanical energy absorber

This paper deals with multi-objective crashworthiness optimization of circular hybrid tubes under axial impact load with the aim of maximizing the absorbed energy and load efficiency factor (the ratio of mean load to maximum load) using Non-dominated Sorting Genetic Algorithm-II. A circular aluminum tube is considered which is reinforced by four composite layers with the layup of [۹۰/ϴ۱/ϴ۲/۹۰] . Fiber angle orientations of the middle layers (ϴ۱,ϴ۲) are selected as design variables which are measured from the tube axis. In order to reduce the occupant injury, the mean crushing load is constrained to not exceed from an allowable limit. To reconstruct the response surfaces of objective functions and constraint in terms of design variables, back-propagation neural network is utilized. Training and testing of the neural networks are carried out using obtained results from numerical simulations by ABAQUS finite element code. To validate the numerical simulations, results are compared with the experiments in the literature which shows reasonably good agreement. At last, optimum results are presented as a Pareto frontier.

کلیدواژه ها:

Energy absorption ، Circular hybrid tubes ، Axial crushing ، Optimization ، Genetic algorithm