基于自适应网络模糊推理系统的开关磁阻电机建模方法

提出一种开关磁阻电机(switched reluctance motor，SRM)数学建模的新方法：在已知开关磁阻电机静态电感曲线和矩角特性曲线的基础上，将自适应网络模糊推理系统(adaptive network based fuzzy inference system，ANFIS)用于SRM的整体建模中。该模糊推理系统由5层网络构成，将模糊推理与神经网络有机结合起来，利用它的自学习功能计算出模糊系统的隶属度函数以及相应的模糊规则，形成一个结构简单、紧凑的网络来实现电机绕组电流、转子位置角与电感和转矩的非线性映射关系，然后离线训练得到电感与转矩模型。把这种基于ANFIS的电感和矩角模型应用于SRM的系统建模中，以550 W、6/4极SRM为例，进行了仿真与实验比较，结果表明此建模方法能够较好的反映SRM的实际工作状况，从而为SRM系统的建模分析与设计提供一种新的有力的工具。

Modeling For Switched Reluctance Motor Based on Adaptive Network-based Fuzzy Inference System

This paper proposes a novel mathematic model for switched reluctance motor (SRM): adaptive network based fuzzy inference system(ANFIS) is used to model for SRM based on the static phase inductance and torque-angle characteristics. This fuzzy inference system is consists of five levels of networks，combined neural network and fuzzy inference，and using the system's self-learning function to calculate the fuzzy membership functions and the corresponding fuzzy rules. So that the ANFIS is built with a much simpler and tighter structure to form an efficient nonlinear map. It realizes the current，the rotor position and the nonlinear relationship between flux linkage and torque. Then the flux linkage and torque models are trained off line. The inductance and torque models based on the ANFIS are utilized to model of the SRM system. Taking a 550W 6/4 poles SRM as example, the simulation and experimental test are carried out to show that this method is proved to be more accurate. It provides the application of SRM with a new powerful tool.