基于改进遗传算法的
递归神经网络非线性系统辨识

基于改进遗传算法的递归神经网络非线性系统辨识

摘要: 将递归内时延神经网络应用于非线性动力学系统辨识中,描述了其动力学方程,并引入改进遗传算法作为其学习算法,通过非线性动力学SISO和MIMO系统的辨识仿真研究,验证了内时延递归网络结构和改进遗传算法的有效性。
- 非线性系统 /
- 辨识 /
- 递归神经网络 /
- 遗传算法
Abstract: An internal time-delay recurrent neural network (RNN) is used for identification of nonlinear dynamic systems, and its dynamic equations are described. As a learning algorithm, an improved genetic algorithm is applied to train the RNN. Two identification simulation examples for nonlinear dynamic SISO and MIMO plants validate the efficiency of the internal time-delay RNN and the improved GA.
- nonlinear system /
- identification /
- recurrent neural networks /
- genetic algorithms

点击查看大图