RSM and ANN modeling of the energy absorption behavior of steel thin-walled columns: a multi-objective optimization using the genetic algorithm

Abstract

In many industries, it is necessary to use structures that exhibit a proper stability against the design loads and depreciate the energy in a controlled manner. In this study, the energy absorption characteristics of thin-walled structures with rectangular cross sections are investigated under the quasi-static loading. The section of structures has a different aspect ratio, and in all of them, an elliptical cutout with a different diameter ratio exists on the larger side. In all instances, the area of the cross section and cutout is constant. Hence, an experimental design with two design parameters consisting of the shell aspect ratio and the diameter ratio of the cutout was conducted by applying the central composite design method. Energy absorption parameters were modeled using the artificial neural network and the response surface method. A systematic crashworthiness study was carried out with a multi-objective optimization design using the genetic algorithm. The results showed that the optimal amount of the specific energy absorption was 14.48 kJ/kg and the optimal amount of the peak crushing load was 37.77 kN which was obtained in the aspect ratio of 1 and the diameter ratio of 0.7. The validity of the results was confirmed by empirical experiments.