一种用于非线性系统辨识与控制的自组织模糊神经网络  

A Self-organizing Fuzzy Neural Network for Identification and Control of Nonlinear Systems

作　　者：寇增前 张建华[1] 王如彬[2] 

机构地区：[1]华东理工大学信息科学与工程学院,上海200237 [2]华东理工大学理学院,上海200237

出　　处：《华东理工大学学报（自然科学版）》2016年第6期835-844,862,共11页Journal of East China University of Science and Technology

基　　金：国家自然科学基金(61075070);国家自然科学基金重点项目(11232005)

摘　　要：提出了一种自组织模糊神经网络(Self-Organizing Fuzzy Neural Network,SOFNN),采用了误差反向传播算法与带遗忘因子的递推最小二乘法相结合的混合优化算法优化系统的模糊规则库及其参数,此外,也引入SRIC(Schwarz&Rissanen Information Criterion)准则设计模糊系统。将本文提出的方法应用于非线性系统的辨识与控制,并讨论了阈值参数对该方法性能的影响。仿真结果表明,本文方法能有效地防止模糊模型过拟合,提高模糊系统的泛化能力,进而提高控制性能。In this paper we design a self-organizing fuzzy neural network （SOFNN） by structure learning and parameter learning. A hybrid learning algorithm, by integrating back propagation and recursive least-squares （RLS） algorithm with forgetting factor,is used to learn the optimal parameters of the SOFNN. Furthermore,a fuzzy system is constructed and evaluated under the Schwarz ＆ Rissanen information criterion （SRIC）. Finally, several simulation examples of identification and model-reference tracking control of nonlinear systems are presented and analyzed to demonstrate the effectiveness of the proposed method, and the effect of the threshold parameter in fuzzy rule learning algorithm is also discussed. The simulation results show that this method can effectively prevent the system from overfitting,improve the generalization ability of the system, and acheive the control performance of the system.

关 键 词：模糊控制 模糊系统 自组织模糊神经网络 非线性控制 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置]