首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于DSP的无刷直流电动机模糊神经网络调速系统
引用本文:王晓远,宋鹏,苏轶,田亮,习贺勋.基于DSP的无刷直流电动机模糊神经网络调速系统[J].天津大学学报(自然科学与工程技术版),2006,39(2):149-154.
作者姓名:王晓远  宋鹏  苏轶  田亮  习贺勋
作者单位:[1]天津大学电气与自动化工程学院,天津300072 [2]北京电力公司工程管理中心,北京100062
基金项目:天津市自然科学基金资助项目(023602911).
摘    要:为有效降低直流无刷电动机固相电流换相而引起的转矩脉动,设计出以TMS320LF2407为核心的全数字调速系统.给出了硬件电路的设计,驱动电路简单、高效;分别对参数自调节模糊控制器和BP神经网络控制器做出了数学解释;提出以系统超调量为判据实现切换的双模控制系统;论述了在换相过程中保持PWM占空比等于1来减小换相转矩脉动的方法.实验结果表明,该系统具有明显优于PI调节器的转矩脉动抑制效果、较强的鲁棒性及抗干扰能力,且超调更小,响应时间更快(超调量减少12%,响应时间缩短30%)。

关 键 词:模糊理论  神经网络  无刷直流电动机  速度  调节器
文章编号:0493-2137(2006)02-0149-06
收稿时间:2004-09-17
修稿时间:2004-09-172005-03-30

Fuzzy Neural Network Speed Regulator of Brushless DC Motor Based on DSP
WANG Xiao-yuan,SONG Peng,SU Yi,TIAN Liang,XI He-xun.Fuzzy Neural Network Speed Regulator of Brushless DC Motor Based on DSP[J].Journal of Tianjin University(Science and Technology),2006,39(2):149-154.
Authors:WANG Xiao-yuan  SONG Peng  SU Yi  TIAN Liang  XI He-xun
Institution:1. School of Electrical and Automation Engineering, Tianjin University, Tianjin 300072, China; 2. Management Centre for Project, Beijing Electric Power Corporation, Beijing 100062, China
Abstract:A speed feedback structure is put forward as a solution to minimizing ripple of torque produced by commutation of phase current of brushless DC motor. The design for hardware circuit are given; a parameter self-correction fuzzy controller and a neural network controller are mathematically interpreted; a method to switch between both controllers according to overshoot is set forth; and the way to keep PWM duty cycle equal to 1 to reduce ripple of torque is demonstrated. Experimental results show significant reduction in overshoot and response time by 12% and 30% respectively. So it's more applicable to minimize ripple of torque compared to a PI regulator.
Keywords:fuzzy theory  neural network  brushless DC motor  speed  regulator
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号