摘要： 提出一种综合控制策略的实数编码遗传算法,用该算法实现对前向网络结构及权值的同时优化设计。用非线性函数的逼近问题作仿真实验。结果表明,该算法能快速有效地确定网络结构及权值。

关键词: 实数编码遗传算法, 综合控制策略, 前向神经网络, 同时优化设计

Abstract: A real-coded genetic algorithms with synthetic control strategy is presented to optimize simultaneously the structure and weights forfeedforward neural networks. Experimented with approximation of nonlinear function, the result shows that the proposed algorithm can acquiresimultaneously topology and weights rapidly and effectually.

中图分类号: 

• TP183