Algorithmes hybrides pour l’optimisation globale
Application en forgeage
DOI:
https://doi.org/10.13052/REMN.17.303-322Keywords:
genetic algorithm, evolution strategy, clustering, Lisza-Orkisz, forging process, shape optimization, adjoint methodAbstract
We introduce two evolutionnary hybrid optimizers, based on surrogate models which use a limited prescribed number of exact evaluations of the criterion and its gradient. The first algorithm uses a discontinuous ansatz with a clustering technique. The second one uses a Liszka-Orkisz interpolation scheme, and keeps memory of the exactly evaluated individuals of previous generations. These two methods are applied to a 3D forging shape optimization problem. The considered objective combines the total energy cost and a defect criterion. We present numerical results which illustrate the efficiency of the developped algorithms.
Downloads
References
Antonio C. A., Dourado N.M., « Metal forming process optimization by inverse evolutionnary
search », J. of Materials Processing technology, vol. 121, p. 403-413, 2002.
Bäck T., Fogel D.B., Michalewicz Z., Evolutionary computation 1. Basic algorithms and operators,
vol. XXXVIII, IoP, Institute of Physics Publishing, Bristol, 2000.
Berard Y., Désidéri J.A., Habbal A., Janka A., Oulladji L., « Experiments with Hybridized
Genetic Algorithms in Aerodynamics », Evolutionnary Methods for Design, Optimization
and Control with Applications to Industrial and Societal Problems, ECCOMAS, Barcelona,
Spain, September, 15-17, 2003.
Conn A.R., Scheinberg K., Toint Ph.L., « Recent progress in unconstrained nonlinear optimization
without derivatives », Math. Program., vol. 79, p. 397-414, 1997.
Coupez T., Mocellin K., Fourment L., Chenot J.-L., « Toward large scale F.E. computation of
hot forging process using iterative solvers, parallel computation and multigrid algorithms »,
International Journal for Numerical Methods in Engineering, vol. 52, p. 473-488, 2001.
Emmerich M., Giotis A., Özdemir M., Bäck T., Giannakoglou K., « Metamodel assisted evolution
strategies », in , J. M. G. et al. (ed.), Parallel Problem Solving from Nature VII, Int’l
Conf. Granada, September, 2002.
Fourment L., Vieilledent D., Chenot J.-L., « Shape optimization of the axisymmetric preform
tools in forging using a direct differentiation method », Int. J. of Forming Processes, vol. 1,
n° 4, p. 399-423, 1998.
Fourment L., Ward M., « Shape optimization for preform tool design in reverse Superplastic
Forming », VII International Conference on Computational Plasticity, COMPLAS, Barcelonna,
April 7-10, 2003.
Goldberg D.E., Genetic algorithms in search, optimization and machine learning, vol. XIII,
Addison-Wesley, Reading, MA, 1989.
Holland J.H., « Outline for a logical theory of adaptive systems », J. Assoc. Comput. Mach.,
vol. 9, p. 297-314, 1962.
Laroussi M., Fourment L., « The adjoint state method for sensitivity analysis of non-steady
problems. Application to 3D forging », Int. J. of Forming Processes, vol. 7, n° 1-2, p. 35-64,
Liszka T., Orkisz J., « The finite difference method at arbitrary irregular grids and its application
in applied mechanics », Comp. and Structures, vol. 11, p. 83-95, 1980.
PowellM.J.D., « An efficient method for finding the minimum of a function of several variables
without calculating derivatives », Comput. J., vol. 7, p. 155-162, 1964.
Sousa L.C., Castro C.F., « Inverse methods in design of industrial forging processes », J. of
Materials Processing technology, vol. 128, p. 266-273, 2002.
Tho D., Optimisation de forme en forgeage 3D, PhD thesis, Ecole des Mines de Paris, Juillet,
