神经网络结合遗传算法优化应用

Combination of GA and BP Networks for Optimization

神经网络特别是BP网络因其函数逼近能力已取得了广泛的应用 ,遗传算法因其解决优化问题的普遍适用性而在现实生活及科研领域获得了广泛应用。本文提出的优化策略是为解决一些工程优化问题 ,即用神经网络及遗传算法结合起来解决此类问题。以BP网络的函数逼近能力隐式地得到问题的函数表达式 ,再用遗传算法优化该网络的输出。

Standard BP networks has been widely used because of its capacity of function approximation and generation. Genetic algorithms capacity of optimization made it widely used as a optimizer. In this paper, the authors proposed an approach of optimization of some engineering optimization problems, through combining GA and BP networks. Using BP networks to approximate the function of the optimized problem, then use GA to optimize the output of the BP networks.