PENGGUNAAN TEKNIK APROKSIMASI (METAMODELING) UNTUK MENYEDERHANAKAN MODEL ELEMEN HINGGA DALAM PROSES OPTIMASI
Abstract
Complex modelling using Finite element method in iterative optimisation process often yield expensive computation. One of the solution is to replace the expensive finite element analysis with a mathematically simpler model, that is using an approximate relations (Metamodel). Here, the approach is used to characterise the mechanical properties of an anisotropic plate from their corresponding natural frequencies. Radial basis functions model has been chosen for the Metamodel to approximate the finite element model generated in ANSYS 6.1 software. Characterisation process is regarded as an optimisation problem with the objective is to minimise the accumulation of differences between the target frequencies and the updated ones. The use of Metamodel shorten the computation time as well as increase the likelihood of reaching the global minimum. In general, the overall result is reasonably good to predict the mechanical properties of anisotropic plates.References
Jin, R., W. Chen, and T.W. Simpsom, Comparative studies of metamodelling techniques under multiple modelling criteria. Structural and Multidisciplinary Optimization, 2001. 23(1): p. 1-13.
Meckesheimer, M., et al., Computationally inexpensive metamodel assessment strategies. AIAA Journal, 2002. 40(10): p. 2053-2060.
Bazan, M., M. Aleksa, and S. Russenschuck, An improved method using radial basis function neural networks to speed up optimization algorithms. IEEE TRANSACTIONS ON MAGNETICS, 2002. 38(2).
Olsson, A.M.J. and G.E. Sandberg, Latin hypercube sampling for stochastic finite element analysis. Journal of Engineering Mechanics-ASCE, 2002. 128(1): p. 121-125.
Roux, W.J., N. Stander, and R.T. Haftka, Response surface approximations for structural optimization. Int. J. Numer. Meth. Engng., 1998. 42.
Kleijnen, J., Statistical tools for simulation practitioners. 1987, New York: Marcel Dekker.
Epidemiological and Public Health Interactive Tools. 2000, Mathepi.
Orr, M., Introduction to radial basis function networks. 1996: Edinburgh.
Chen, S., C. Cowan, and P.M. Grant, Orthogonal least squares learning for radial basis function networks. IEEE Trans. on Neural Networks, 1991. 2(2): p. 302 - 309.
Kelley, C.T., Iterative Methods for Optimization. no. 18 in Frontiers in Applied Mathematics, SIAM. 1999, Philadelphia: SIAM.
Sol, H., Identification of anisotropic plate rigidities using free vibration data, Ph.D. Dissertation, Free University of Brussels (VUB), 1986.
De Wilde, W.P. and H. Sol, Anisotropic material identification using measured resonant frequencies at rectangular composite plates, in Composite Structure 4, I.H. Marshal, Editor. 1987. p. 317-324.
Deobald, L.R. and R.F. Gibson, Determination of elastic constants of orthotropic plates by a modal analysis/Rayleigh-Ritz technique. J. Sound & Vibration, 1988. 124(2): p. 269-283.
Papazoglou, V.J., N.G. Tsouvalis, and A.G. Lazaridis, A non-destructive evaluation of the material properties of composite laminated plate. Applied Composite Materials, 1996. 3: p. 321-334.
De Visscher, J., et al., Identification of damping properties of orthotropic composite materials using a mixed numerical experimental method. Applied Composite Materials, 1997. 4: p. 13-33.
Ayorinde, E.O. and R.F. Gibson, Elastic constants of orthotropic composite materials using plate resonance frequencies, classical lamination theory and an optimized three-mode Rayleigh formulation. Composite Engineering, 1993. 3(5): p. 395-407.
Pedersen, P. and P.S. Frederiksen, Identification of orthotropic material moduli by a combined experimental/numerical method. Measurement, 1992. 10(3): p. 113-118.
Frederiksen, P.S., Numerical studies for the identification of orthotropic elastic constants of thick plates. Eur. J. Mechanics, A/Solids, 1997. 16(1): p. 117-140.
Frederiksen, P.S., Experimental procedure and results for the identification of constants of thick orthotropic plates. J. Composite Materials, 1997. 31(4): p. 360-382.
Setiawan, R., Composite plate mechanical characterisation through dynamic tests, Ph.D Thesis, in School of Engineering Sciences, University of Southampton, Southampton, 2004.
