首页 | 官方网站   微博 | 高级检索  
     

行波型超声电机基于神经网络的逆模型辨识
引用本文:张新良,谭永红.行波型超声电机基于神经网络的逆模型辨识[J].控制与决策,2009,24(2).
作者姓名:张新良  谭永红
作者单位:1. 上海交通大学,自动化系,上海,200240
2. 上海师范大学,机电学院,上海,201814
基金项目:国家自然科学基金,上海师范大学前瞻性项目 
摘    要:行波型超声电机的动态特性受定子压电陶瓷迟滞和接触层非线性摩擦力的影响,表现出复杂的多值映射特征.通过引入动态迟滞逆算子,将存在于超声波电机逆系统中的多值映射在新的扩张输入空间上,转换为一一映射;然后使用神经网络建立超声波电机的逆模型,对迟滞和非线性摩擦力的影响进行补偿.所建立的模型结构简单,可以在线调整适应电机参数的非线性变化.实验仿真结果验证了该方法的有效性.

关 键 词:行波型超声电机  迟滞  神经网络  动态迟滞逆算子

Neural networks inverse model for travelling wave ultrasonic motor
ZHANG Xin-liang,TAN Yong-hong.Neural networks inverse model for travelling wave ultrasonic motor[J].Control and Decision,2009,24(2).
Authors:ZHANG Xin-liang  TAN Yong-hong
Affiliation:1.Department of Automation;Shanghai Jiaotong University;Shanghai 200240;China;2.College of Mechanical and Electronic Engineering;Shanghai Normal University;Shanghai 201814;China
Abstract:Due to the hysteresis in pizoelectric ceramic and the friction force from the contact interface,the travelling wave ultrasonic motor(TWUSM)exhibits heavy nonlinear dynamics which are characterized of multi-valued mapping.In this paper,an inverse model based on neural networks for the TWUSM is proposed to compensate the effect of the hysteresis and friction.By introducing a dynamic inverse hysteresis operator,the multi-valued mapping in the inverse system of the motor is decomposed into one-to-one mapping.Th...
Keywords:Travelling wave ultrasonic motor  Hysteresis  Neural networks  Dynamic inverse hysteresis operator  
本文献已被 CNKI 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-23

京公网安备 11010802026262号