基于模糊神经网络开关磁阻电动机转矩观测  

Torque Estimation of Switched Reluctance Motor Based on Fuzzy-Neural Networks

作　　者：郑洪涛[1] 李玉榕[2] 乔斌[1] 蒋静坪[1] 

机构地区：[1]浙江大学电气工程学院,杭州310027 [2]福州大学电气系,福州350002

出　　处：《仪器仪表学报》2002年第4期351-353,360,共4页Chinese Journal of Scientific Instrument

基　　金：浙江省科委资助 (0 0 1 1 0 61 2 7);福建省自然科学基金 (A0 0 1 0 0 0 6)项目

摘　　要：转矩脉动是开关磁阻电动机较为突出的缺点 ,出于电机设计和控制的目的 ,常需要实时在线测量电机的动态转矩 ,采用转矩传感器的方法既复杂、昂贵 ,在高速下又不准确。利用一种新型变结构模糊神经网络对开关磁阻电动机转矩特性进行学习 ,之后把它用于动态转矩的实时在线测量。仿真与试验结果表明此方法能够快速。Switched reluctance motor(SRM)produces excessive vibration and torque ripple in low speed area So the realtime and online measurement of SRM's torque is required for control and design purpose However,the measurement method which using torque transducer is complex,expensive and inaccurate This paper uses a novel Fuzzy neural networks(FNN)to approximate the nonlinear torque characteristics of SRMs as a function of position and current,and then the FNN is applied to online measurement of SRM's torque The simulation and experiment results illustrate the validity of the proposed method

关 键 词：开关磁阻电动机 转矩脉动 测量 模糊神经网络 

分 类 号：TM352[电气工程—电机] TP183[自动化与计算机技术—控制理论与控制工程]