摘要
针对一阶Takagi－Sugeno[以下简称T－S]到模型辨识复杂系统的困难，本文提出了一种新的拟非线性模糊模型．即在一阶T－S模型的基础上，再进行一次非线性映射．这种模糊模型不仅具有较高的辨识精度，而且具有良好的泛化功能．运用改进的FCM（FuzzyC－Means）模糊聚类方法，辨识该模糊模型的结构，与以往的方法比较，极大地简化了结构辨识的复杂性．仿真结果进一步说明了该方法的有效性．
In this paper,a new Quasinonlinear Fuzzy Model(QNFM) is presented to overcome the difficulty of the identification of complex systems using the first order Takagi-Sugeno model. The structure of thefuzzy model is based on the first order Takagi-Sugeno model,then a nonlinear map is carried out. The presented fuzzy model has the advantages of high identification accuracy and good generalization performance. Thestructure of the fuzzy model is identified by the modified FCM fuzzy clustering technique,compared with other existing methods,the procedure for finding the optimal structure of the fuzzy model is simplified. The simulation results show that this method is very efficient.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
1998年第2期286-290,共5页
Control Theory & Applications
