关键词: 遗传－算法, 模型－降阶, 神经网络

Abstract:

The model reduction problem of large space structure is solved by direct numerical optimization using genetic algorithms. The mathematic model of the problem and the used genetic algorithms are given. The success of genetic approach is illustrated through examples, including the model reduction of a two bay truss. Results are compared to those obtained by standard internal balancing and by a Hopfield neural network approach. It is concluded that genetic algorithms are a viable solution to the model reduction problem of large space structures.

Key words: genetic-algorithms, model-reduction, neural nets