一类非线性系统神经网络控制的稳定性分析

从理论上研究将神经网络用于非线性系统控制，通过对神经网络的训练，实现一类非线性系统的定点跟踪。证明了神经网络学习算法的收敛性不仅与系统的动态特性有关，而且与网络的初始条件有关。仿真结果表明，适当选取网络的初值和加权的调节速率，可以实现非线性系统的定点跟踪。

Stability Analysis of a Class of Nonlinear Systems Using Neural Network Control

neural network controller is constructed for set-point tracking in a class of nonlinear systems.By training the neural network using the proposed algorithm, the set -POint tracking of nonlinear systemscan be achieved. The convergence of the proposed algorithm is proyed to be governed by not only the dynamiccharacteristics of the system but also the initial weights of the neural networks. Simulation results show thatthe asymptotic set -point tracking of the systems can be guaranteed by selecting the proper initial weightsand the updating rate of the weights of the networks.