高性能感应电机智能动态滑模位置控制  

Intelligent Dynamic Sliding Mode Control of High Performance Induction Motor

作　　者：祁瑒娟 于洋[2] QI Yangjuan;YU Yang(Department of Mechanical and Electrical,Baotou Railway Vocational and Technical College,Baotou 014060,Nei Monggol,China;Jilin Railway Technology College,Jilin 132000,Jilin,China)

机构地区：[1]包头铁道职业技术学院机械与电气系,内蒙古包头014060 [2]吉林铁道职业技术学院,吉林吉林132000

出　　处：《电气传动》2022年第6期9-13,32,共6页Electric Drive

摘　　要：为提高感应电机(IM)伺服驱动系统的控制性能,抑制电机参数变化、外部扰动和未建模动态等不确定性因素对系统的影响,提出一种基于径向基神经网络(RBFN)的智能动态滑模控制(IDSMC)方法。首先利用动态滑模控制(DSMC)方法削弱抖振,提高系统的跟踪精度。但由于DSMC中切换函数所需的不确定性边界值无法获知,因此将RBFN不确定性估计器与DSMC相结合,设计IDSMC方法进一步提高系统的鲁棒性。RBFN可通过自适应学习算法估计不确定性因素值并在线训练调整网络参数,以确保系统在不确定性因素存在时仍能高性能运行。最后,通过TMS320C31 DSP控制核心验证所提方法的有效性。实验结果表明,IDSMC不但可以保证系统精准的响应能力,还有较强的鲁棒性。In order to improve the control performance of induction motor(IM)servo drive system,and to restrain the influence of uncertain factors such as motor parameter change,external disturbance and un-modeled dynamics,an intelligent dynamic sliding mode control(IDSMC)method based on radial-basis function neural network(RBFN)was proposed.Firstly,dynamic sliding mode control(DSMC)method was used to reduce chattering and improve the tracking accuracy of the system.However,the uncertainty boundary value of switching function in DSMC can not be obtained,so combining RBFN uncertainty estimator with DSMC,the IDSMC method was designed to further improve the robustness of the system.RBFN can estimate the value of uncertainty factors by adaptive learning algorithm and adjust the network parameters by online training,so as to ensure that the system can still run with high performance when the uncertainty factors exist.Finally,the effectiveness of the proposed method was verified by TMS320C31 DSP control core.The experimental results show that IDSMC can not only ensure the accurate response ability of the system,but also has strong robustness.

关 键 词：感应电机 动态滑模控制 径向基神经网络 鲁棒性 

分 类 号：TM272[一般工业技术—材料科学与工程]