A two-archive multi-objective multi-verse optimizer for truss design

Multi-objective structure optimization is a complex design issue that involves dealing with multiple conflicting objectives and various constraints. Although metaheuristics have been successful in resolving such issues, their stochastic nature and limitations can make implementation challenging. To address this challenge, an efficient and powerful optimizer called the multi-objective multi-verse optimizer (MOMVO) has been proposed in this study. The MOMVO algorithm is based on a twoarchive concept focused on convergence and diversity, respectively, and is termed MOMVO2arc. The effectiveness of MOMVO2arc was evaluated by applying it to five conventional planar and spatial real-world structural optimization problems. These problems had conflicting objectives, such as minimizing structural mass and minimizing maximum nodal deflection, while imposing safety and size constraints related to stress on components and discrete cross-sectional areas. The results were then compared with those of seven advanced optimization techniques, including MOEA/D and NSGA-II, using four globally accepted performance indicators, including Hypervolume, Inverted Generational Difference, Spacing to Extent matrices, and the qualitative behavior of the best Pareto-front plots that each algorithm can produce. The comparative analysis and average Friedman rank test revealed that MOMVO2arc was more effective in solving large structure optimization problems with less computation time. The suggested approach not only found and maintained more Pareto-optimal sets but also achieved good diversity and convergence in both the decision and objective spaces. Therefore, MOMVO2arc can be a promising tool for resolving complex multi-objective structure optimization problems.

Automated summary

Multi-objective structure optimization is a complex design issue that involves dealing with multiple conflicting objectives and various constraints. A powerful optimizer called MOMVO2arc has been proposed and evaluated for solving large structure optimization problems with less computation time.