基于动态模糊神经网络的非线性系统辨识

在分析模糊神经网络辨识特点及现状的基础上,设计了一种适用于非线性多输入系统的辨识模型。本模型将T-S模糊模型与5层动态模糊神经网络结构相结合,通过参数学习算法优化辨识结构,对辨识模型进行反馈调节,得到的辨识精度较高。另外,对输入数据采用归一化的方法进行预处理,加快了网络的辨识速率。最后,通过仿真实例证明了该设计的有效性,为模糊神经网络辨识结构的设计提供了一种新的思路和方法。

Identification of Nonlinear System Based on Dynamic Fuzzy Neural Network

The paper analyses the characteristics and the situation of dynamic fuzzy neural network identification, and designs a suitable identification model for muhivariable nonlinear system. The model combines T-S fuzzy model with 5-layer dynamic neural network, thus the identification structure can be optimized by using a kind of parameters learning algorithm and the identification precision can be improved. In addition, it gains quickly identification veloc- ity by the input-data preconditioning. Finally, the simulation proved the effectiveness of the model, which provides a new idea and method for designing a fuzzy neural network identification model.