Global optimization of laminated cylindrical panels based on fundamental natural frequency

This investigation presents an optimization of laminated cylindrical panels based on fundamental natural frequency. Also, trends of change in optimum stacking sequence while the proportions of structures vary, are studied which can be insightful for design purposes. A displacement based finite element model is used, in order to extract fundamental natural frequencies of T300/5208 Carbon/Epoxy cylindrical panels. To obtain optimum designs, the Globalized Bounded Nelder–Mead (GBNM) algorithm is employed. Predictions are compared with the results of Genetic Algorithm (GA) method and show faster and more accurate convergence to the global optimum, while variables are continuous in GBNM and discrete in GA. Moreover, verification of novel convergence criteria to ameliorate local searcher in GBNM is examined.