基于变结构模糊神经网络的开关磁阻电动机非线性模型

Nonlinear Modelling of Switched Reluctance Motor Based on Variable Structure Fuzzy-Neural Networks

开关磁阻电动机驱动系统是一个时变、非线性系统 ,若采用通常的建模方法 ,无法获得精确的数学模型。给出了开关磁阻电机变结构模糊神经网络非线性模型 ,基于Takagi Sugeno模糊神经网络 ,提出可变结构的变步长学习算法。仿真结果表明此法比BP神经网络具有更高的精度和更快的收敛速度。

Switched Reluctance Motors(SRM) are almost always operated within the saturation region for a very large operation region.This yields very strong nonlinearity,which makes it very difficult to derive a comprehensive mathematical model for the behavior of the machine.This paper presents the variable structure fuzzy neural network model of SRM.Based on the Takagi Sugeno fuzzy neural networks a variable structure and step learning arithmetic was presented.Simulation results show that this method is more precise and less time consuming for convergence than BP neural network model.