基于T-S模糊模型的神经网络的系统辨识

A System Identification for T-S Model Based on Neural Network

基于T-S模糊模型,提出了利用神经网络实现非线性系统的辨识。首先,利用一种无监督的聚类算法分析输入输出数据生成初始的结构模型,确定系统的模糊空间和模糊规则数,构造神经网络辨识模型前提参数,使前提参数自适应变化,有较好的自学习能力和优化能力,采用最小二乘法取得结论参数。仿真结果验证了该方法是有效和可行的。

In accordance with T-S fuzzy model, the paper proposes the identification of nonlinear dynamic system based on neural network. First, a clustering algorithm is utilized to determine the structure of system and obtains the fuzzy space and the number of fuzzy rides by analyzing the sample. Further, the parameters of model are identified by the neural network and LS with better selflearning and generalization ability .The simulation results demonstrate the effectiveness of proposed method.