Multi-objective lightweight design of the container S-beam based on MNSGA-II with grey relational analysis

Abstract

A novel bottom corrugated cross-beam (S-beam) structure improved the dynamic and static performance of a container based on the combination of a modified non-dominated sorting genetic algorithm (MNSGA-II) and grey relational analysis. First, a parametric model was established and used to verify the structure’s validity through static physical testing. Eight design variables for the S-beam container structure were also defined according to the parametric model technology. Second, MNSGA-II was used for the multi-objective lightweight optimization design of an S-beam container to obtain the optimal combination of design parameters that are considerably affected by weight reduction under peak bending stress and peak loading deflection as well as first-order natural frequency variations within the allowable range. A set of non-dominated solutions was used to obtain a multi-objective optimization design. Finally, grey relational analysis and grey entropy theory are applied to rank all solutions and determine the best compromise solution. The comparison of the technique for the order of preference by similarity to ideal solution method with grey relational analysis demonstrates the extraordinary reliability and superiority of the latter. In addition, the combined method can achieve a weight reduction of up to 23.54%, which can enhance the utilization of materials and demonstrates the superiority of the combined method relative to the initial model.

Multi-objective optimization for multi-stage sequential plastic injection molding with plating process using RSM and PCA-based weighted-GRA

“…In other multi-objective optimization studies, GRA was combined with modified nondominated sorting genetic algorithm (MNSGA-II), and the optimized results were compared with similarity to ideal solution (TOPSIS). 28,29 In another multi-objective optimization studies, a radial basis function (RBF) neural network was employed to map the relation between GRG and process parameters 30,31 and further the PSO algorithm was applied on the RBF prediction model to find the optimal value of GRG. For multi-objective optimization, Taguchi method, GRA, and RSM were also integrated to predict the optimal process condition.…”