中南大学学报（自然科学版）2009，Vol.40Issue(5)：1367-1373,7.
基于神经元的异步电机直接转矩控制系统速度辨识
Speed identification in direct torque control system of induction motor based on neuron
摘要
Abstract
In order to avoid solving the problems of obtaining feedback gain matrix in traditional model reference adaptive system (MRAS), the asynchronous motor mathematical model was established in stator α-β coordinate system based on direct torque control (DTC) theory, discrete space vector modulation (DSVM)-DTC system and stator current equation were gained according to this model and then the equation was transformed into neural element with five input nodes, and an output node as stator current adaptive identification modules. According to speed identification formula, the speed of motor was real-time estimated and transferred back to adjustable model whose parameters were modified and made close to the real value. Finally, the simulation was made in the Matlab/Sinulink using parameters of 37 kW asynchronous motors. The results show that the identified speed can accurately track the actual speed and perform good dynamic response. When system parameters change, the system has good robustness and is easy for application.关键词
神经元/转速辨识/DSVM/直接转矩控制Key words
neuron/ speed identification/ discrete space vector modulation (DSVM)/ direct torque control分类
信息技术与安全科学引用本文复制引用
高金生,高宏洋,喻寿益..基于神经元的异步电机直接转矩控制系统速度辨识[J].中南大学学报（自然科学版）,2009,40(5):1367-1373,7.基金项目
国家高技术研究发展计划("863"计划)项目(2006AA042181) （"863"计划）
