摘要
为解决永磁直线同步电机(PMLSM)伺服系统位置跟踪精度易受参数变化、负载扰动、摩擦力等不确定性因素影响的问题,该文提出一种基于径向基函数(RBF)神经网络反推终端滑模控制方法。首先,建立含有不确定性的PMLSM动态数学模型。然后,采用反推终端滑模控制将系统状态在有限时间内收敛到平衡点,提高系统的响应速度;为了进一步削弱抖振现象,利用双曲正切函数与边界层厚度相结合来设计饱和函数,以取代符号函数;并且利用RBF神经网络去逼近系统中存在的不确定性,进而获得快速的跟踪性能和较强的抗扰能力。最后,实验结果表明,所提出的控制方法不仅改善了系统的跟踪性和鲁棒性,而且明显削弱了抖振问题。
The position tracking accuracy of permanent magnet linear synchronous motor(PMLSM)servo system is susceptible to parameter variation,load disturbance,friction and other uncertain factors.Therefore,a backstepping terminal sliding mode control method based on radial basis function(RBF)neural network is proposed.Firstly,the PMLSM dynamic mathematical model with uncertainty is established.Then,the state of the system converges to the equilibrium point in a finite time and the response speed of the system is improved by using the backstepping terminal sliding mode control.In order to further weaken chattering phenomenon,the saturation function is designed combined the hyperbolic tangent function with boundary layer thickness to replace the signum function.Moreover,RBF neural network is used to approximate the uncertainties in the system,and then fast-tracking performance and strong immunity ability are obtained.Finally,the experimental results show that the proposed control method not only improves the tracking and robustness of the system,but also significantly weakens the chattering problem.
作者
付东学
赵希梅
Fu Dongxue;Zhao Ximei(School of Electrical Engineering Shenyang University of Technology,Shenyang 110870 China)
出处
《电工技术学报》
EI
CSCD
北大核心
2020年第12期2545-2553,共9页
Transactions of China Electrotechnical Society
基金
辽宁省自然科学基金计划重点项目(20170540677)
辽宁省教育厅科学技术研究项目(LQGD2017025)资助。
关键词
永磁直线同步电机
反推终端滑模控制
径向基函数神经网络
抖振
鲁棒性
Permanent magnet linear synchronous motor
backstepping terminal sliding mode control
radial basis function(RBF)neural network
chattering
robustness
