Keywords: composite, cylindrical panel, optimization, free vibration, genetic algorithm, neural networks.

The use of composite laminated structures, beside their high values of strength to weight ratios, allows the designer to choose between many possible structural layouts of the material, in order to obtain high structural performance. Among these structural components, multi-layered cylindrical panels are a structural topology extensively used in the aerospace field in aircraft fuselages.

The optimal design of these structures with respect to the changes in the orientation of fibres (stacking sequence optimization) is studied by different authors. Kere and Koski [1] have optimally designed composite laminates subjected to multiple loading conditions, considering the laminate failure margins as the criteria. In recent research, the use of evolution strategies, appropriate for procedures contributed with discrete variables, such as the genetic algorithm has been developed. Different researchers have used the G.A. in the optimization of composite laminates [2,3]. More particularly, this algorithm has been used for the stacking sequence optimization of laminates for maximum natural frequency by Shakeri [4] and also by other researchers with different objective functions [5,6,7].

In this study, a newly developed optimization procedure is used for the design of laminated composite cylindrical panels under free vibration requirements, in which neural computing as a global approximation strategy plays a significant role in the sense of speeding up the optimization process. The applicability of neural nets in the optimization of structural components is explained in [8]. This method is implemented in the optimal design of stiffened cylindrical panels in the post-buckling field in [9] and of helicopter components in the crash phenomenon in [10].

Here, first a system of artificial neural networks able to reproduce the free vibration behaviour of the structure is developed. Training and test sets are generated by finite element analyses. Changing the stacking sequence of the panels with different number of layers and thicknesses, corresponding natural frequencies are obtained from the response surfaces generated by the neural system. Then to find the optimal solution, a genetic algorithm is implemented and the optimum layout of the structure for maximum natural frequency is proposed.

Finally, the results of optimization are compared with the case in which the objective function evaluation is done directly by the finite element analyses. By comparison, with an acceptable error, computational time of the optimization process is reduced by a considerable amount.

1

P. Kere, J. Koski, "Multi-Criterion Stacking Sequence Optimization scheme for composite Laminates Subjected to Multiple Loading Conditions", Composite Structures, 54, 225-229, 2001. doi:10.1016/S0263-8223(01)00092-7

2

Muc, W. Gurba, "Genetic Algorithm and Finite Element Analysis in Optimization of Composite Structures", Compostite Structures, 54(2-3), 275, 2001. doi:10.1016/S0263-8223(01)00098-8

3

V.B. Gantovnic, et al, "A Genetic Algorithm with Memory for Optimal Design of Laminated Sandwich Composite Panels", Composite Structures, 58, 513-520, 2002. doi:10.1016/S0263-8223(02)00128-9

4

M. Shakeri, M.H. Yas, M. Ghasemi Gol, "Optimal Stacking Sequence of Laminated Cylindrical Shells Using Genetic Algorithm", Proceeding of 9th EASEC, Bali, Indonesia, 2003.

5

B. Liu, et al, "Permutation Genetic Algorithm for Stacking Sequence Design of Composite Laminates", Computer Methods in Applied Mechanics and Engineering, 186,357-372, 2000. doi:10.1016/S0045-7825(99)90391-2

6

G. Soremakun, et al, "Stacking Sequence Blending of Multiple Composite Laminates Using Genetic Algorithms", Composite Structures, 56, 53-62, 2002. doi:10.1016/S0263-8223(01)00185-4

7

D.B. Adams, et al, "Genetic Algorithm Optimization and Blending of Composite Laminates by Locally Reducing Laminate Thickness", Advances in Engineering Software, 35, 35-43, 2004. doi:10.1016/j.advengsoft.2003.09.001

8

M. Papadrakakis, et al, "Structural Optimization Using Evolution Strategies and Neural Networks", Comput. Methods Appl. Mech. Engng, 156, 309-333, 1998. doi:10.1016/S0045-7825(97)00215-6

9

C. Bisagni, L. Lanzi, "Post-Buckling Optimization of Composite Stiffened Panels Using Neural Networks", Composite Structures, 58, 237-247, 2002. doi:10.1016/S0263-8223(02)00053-3

10

C. Bisagni, et al, "Optimization of Helicopter Subfloor Components under Crashworthiness Requirements Using Neural Networks", J. of Aircraft, 39(2), 296-304, 2002. doi:10.2514/2.2927

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £140 +P&P)