一株东祁连山耐低温纤维素降解菌株的分离、鉴定及产酶特性.pdfVIP

第五届全国振动利用工程学术会议暨第四次全国超声电机技术研讨会  The Fifth Conference on Vibration Utilizing Engineering in China The Fourth Workshop on Ultrasonic Motor Techniques in China （CVUEC WUMTC2012 ）   32 基于自回归神经网络和 Lyapunov 稳定性的超声波 电机伺服控制研究 傅平 闽江学院物理学与电子信息工程系，福州350108 The servo control research of ultrasonic motor Based on regression neural network and Lyapunov stability FU Ping Department of Physics and Telecommunication Engineering ，Minjiang University ，Fuzhou 350108 China 摘要：超声波电机是强非线性时变系统，用常规的 PID 难以取得满意的效果。针对超声波电机的速度/位置伺服控制，本文 提出使用基于自回归神经网络和Lyapunov 稳定性的控制算法，可以在一定程度上反映超声波电机的输入输出非线性关系。 由于自回归神经网络可以在线调整其参数值，因此当电机运行参数发生变化时，可以通过改变神经网络辨识参数来实现速 度和位置的快速跟踪，并通过Lyapunov稳定性进行控制参数调整。研究表明，使用此算法的伺服系统对超声波电机进行伺 服控制，可以达到快速响应和良好的控制效果。 主题词：超声波电机；自回归神经网络；Lyapunov稳定性；伺服控制； 中图分类号：TM355 ABSTRACT: Ultrasonic motor(USM) is a system of severe non-linearity and parameter variations. Due to that, common PID control methods can’t meet practical demands. This paper introduces a novel control system which is based on autoregression neuron network and combined with theory of Lyapunov stability. Because neuron controller can change its parameters online when operation conditions are changed, it can also change its control characteristic automatically to realize speed and position control by Lyapunov stability. The research results show that USM can complete rapid response speed and high precise position control while using this kind of controller based on neuron algorithm. Keyword ：Ultrasonic motor ；Autoregression nerve network ；Lyapunov st