Finite Element Crushing Analysis, Neural Network Modelling and Multi-Objective Optimization of the Honeycomb Energy Absorbers

Abstract

The thin-walled honeycomb structures are one of the most common energy absorber types. These structures are of particular use in different industries due to their high energy absorption capability. In this article, the finite element simulation of honeycomb energy absorbers was accomplished in order to analyze their crushing behavior. 48 panels with different hexagonal edge length, thickness and branch angle were examined. In the following, the amounts of mean stresses versus the geometric variables using neurotic lattices were considered. Comparison between the finite element results and the obtained neural network model verified the high accuracy of the obtained model. Then the model was optimized by one of the efficient genetic algorithm methods called “Multi-objective Uniform-diversity Genetic algorithm”. The obtained optimum results provide practical information for the design and application of these energy absorbers regards to designer requirement. It was observed that honeycomb energy absorbers with 11.07 mm hexagonal edge length, 0.078 mm wall thickness and 123-degree branch angle have the maximum energy absorption over the panel mass.

Keywords

• energy absorbers
• honeycomb
• multi-objective optimization
• neural-network modeling