第19卷第7期 2011年7月
文章编号1004-924X(2011)07-1643-08
光学精密工程
Optics and Precision Engineering
Vol. 19No.7
Jul.2011
模糊神经网络在自适应双轴运动控制系统中的应用
陈向坚，李迪，白越，续志军”
（中国科学院长春光学精密机械及物理研究所，吉林长春130033）
摘要：为了更好地提高工艺加工平台的精确度，结合自适应控制与区间二型模糊神经网络理论设计了一个双轴运动控制系统。该系统通过控制两个场向永磁同步电机来定位X-Y双轴运动转子的位置，从面跟踪预设的蝶形曲线。针对由区间二型模糊神经网络的有限规则产生的不可避免的通近误差和优化的参数向量等集中不确定性因索，设计了自适应集中不确定性估计律，并通过在线鲁棒补偿器处理集中不确定性的值。最后，通过TMS320C32数字信号处理器运行了本文提出的控制算法。实验结果验证了基于区间二型模棚神经网络设计的双轴运动控制系统的轨迹跟踪精确度较高。与
一型模糊神经网络控制系统相比，区间二型模棚神经网络控制系统具有更好的控制性能，鲁棒性更强。关键调：区间二型模棚神经网络；双轴运动控制系统，永磁同步电机；Lyapunov稳定性理论
中图分类号：TP183;TP273.2
文献标识码：A
doi:10.3788/OPE.20111907.1643
Applicationoftype-IIfuzzyneuralnetworkto adaptive double axis motion control system
CHENXiang-jian，LIDi,BAIYue,XUZhi-jun
(Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,130033,China)
*Correspondingauthor,E-mail:ruzj538@ciomp.ac.cn
Abstract: An adaptive double axis motion control system to improve the accuracy of processing plat-forms was designed by combining the adaptive technology and the interval type-Il fuzzy neural net-work theory. The system controled two field oriented permanent magnet synchronous motors to locate the X-Y double axis motion rotor to track the butterfly contour. Meanwhile, a robust compensator was proposed to confront the Lumped uncertainty, including the inevitable approximation error due to the finite rules of the interval type-Il fuzzy neural network, optimal parameter vectors and so on. Fi-nally, the proposed control algorithm was implemented in a TMS32oC32 digital signal processor. The experimental results indicate that the butterfly contour tracking performance of the double axis motion control system is improved significantly,and the control system based on the interval type-Il fuzzy neural network is more robust than that based on the type-I fuzzy neural network for different uncer tainties.
Key words: interval type-II fuzzy neural network; double axis motion control system; permanent-mag
net synchronous motor; Lyapunov stability theorem
收稿日期：2010-12-13；修订日期：2011-02-15
基金项目：国家自然科学基金资助项目（No.50905174）