神经网络技术在大系统递阶优化中的应用研究

Study on Application of Neural Networks to Hierarchical Optimal Control of Large-Scale Systems.

本文给出了一种应用神经网络技术解决大系统递阶优化问题的新方法。文中以解决线性离散大系统的递阶优化问题为例设计出一个二级互联神经网络模型，并对其寻优过程进行了介绍。最后，给出的计算机仿真结果证实了该方法的可行性和有效性。

A new method of applying neural network in solve the hierarchical optimal control problem in large scale systems was given. A two level connectional network of novel architecture was designed for optimization problems of large scale linear discrete time systems. The optimal procedure is described technically in detail, and simulation results demonstrated that this new method is feasible and effective.