Truss optimization with natural frequency constraints using a hybridized CSS-BBBC algorithm with trap recognition capability

• Published in: Computers & structures Vol. 102-103; pp. 14 - 27
• Main Authors: Kaveh, A., Zolghadr, A.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.07.2012; Elsevier
• Subjects: Algorithms; Charged System Search and the Big Bang-Big Crunch (CSS–BBBC); Charging; Exact sciences and technology; Exploration; Frequency constraint structural optimization; Fundamental areas of phenomenology (including applications); Hybridization; Optimization; Physics; Recognition; Resonant frequency; Searching; Solid mechanics; Static elasticity (thermoelasticity...); Structural and continuum mechanics; Truss structures; Trusses; Charged System Search and the Big Bang-Big Crunch (CSS–BBBC); Hybridization; Frequency constraint structural optimization; Truss structures; Charged System Search and the Big Bang-Big; Local search; Constraint; Convex space; Local optimum; Diversity; Constrained optimization; Natural frequency; Trapping; Truss structure; optimization; Crunch (CSS-BBBC); Big bang; Heuristic method; Frequency constraint structural; Non linear effect; Non convex analysis; Structural analysis
• ISSN: 0045-7949; 1879-2243
• DOI: 10.1016/j.compstruc.2012.03.016

► Frequency constraint optimization is performed for highly nonlinear and non-convex search spaces. ► A hybridization of the CSS and BB-BC algorithms is proposed. ► A diversity index is introduced which is employed for trap recognition. ► Five numerical examples are presented to demonstrate the efficiency of the algorithm. Frequency constraint structural optimization includes the exploration of highly nonlinear and non-convex search spaces with several local optima. These characteristics of the search spaces increase the possibility of the agents getting trapped in a local optimum, when using a meta-heuristic algorithm. In this paper a diversity index is introduced which together with a few other criteria, can be employed to recognize such traps. By the use of these concepts, a hybridization of the Charged System Search and the Big Bang-Big Crunch algorithms with trap recognition capability is proposed. Five numerical examples are considered to demonstrate the efficiency of the algorithm.