Neuro-Optimizer, Its Application to Discrete Structural Optimization
Some discrete optimization problems can be programmed and solved on artificial neural networks. The NEURO-OPTIMIZER attains good (not necessarily optimal) solutions for general nonlinear discrete optimization problems through neurons state transitions. The simulated annealing method is introduced to escape from local minima. In this paper the number representation of the discrete variable by using neurons is investigated, and a mapping technique is proposed for irregularly discrete variables.
Keywordsdiscrete optimization neural network structural optimization number representation
Unable to display preview. Download preview PDF.
- 1.Hopfield J.J., Electronic Network for Collective Decision Based on Large Number of Connections between Signals, United States Patent, 1987, No. 4, 660, 166.Google Scholar
- 2.Kishi M., Suzuki T., Hosoda R., Structural Design by Neuro- Optimizer, in Practical Design of Ships and Mobile Units, eds. J.B. Caldwell and G. Ward, Vol.2, Elsevier Applied Science, 1992, 940–952.Google Scholar
- 4.Piatt J. C., Barr A. H., Constrained Differential Optimization, in Neural Information Processing Systems, ed. D.Z. Anderson, American Inst, of Physics, 1988, 612–21.Google Scholar
- 6.Pedersen P., Optimal Joint Positions for Space Trusses, J. Struct. Div., ASCE, Vol. 99, St10, 1973, 2459–2477.Google Scholar