Multiple objective crashworthiness optimization of circular tubes with functionally graded thickness via artificial neural networks and genetic algorithms

Adil Baykasoğlu,Cengiz Baykasoğlu
DOI: https://doi.org/10.1177/0954406215627181
2016-01-28
Abstract:The objective of this paper is to develop a multiple objective optimization procedure for crashworthiness optimization of circular tubes having functionally graded thickness. The proposed optimization approach is based on finite element analyses for construction of sample design space and verification; artificial neural networks for predicting objective functions values (peak crash force and specific energy absorption) for design parameters; and genetic algorithms for generating design parameters alternatives and determining optimal combination of them. The proposed approach seaminglesly integrates artificial neural networks and genetic algorithms. Artificial neural network acts as an objective function evaluator within the multiple objective genetic algorithms. We have shown that the proposed approach is able to generate Pareto optimal designs which are in a very good agreement with the finite element results.
What problem does this paper attempt to address?