基于自适应变步长最小均方算法的磁悬浮高速电机不平衡补偿

Unbalance Compensation for Magnetically Levitated High-Speed Motors Based on Adaptive Variable Step Size Least Mean Square Algorithm

该文针对磁悬浮高速电机的不平衡振动,提出一种基于自适应变步长最小均方误差(LMS)算法的不平衡补偿方法。首先,通过分析步长和频率对LMS算法特性的影响,给出合适的步长计算公式。之后,利用基于变步长LMS算法的自适应滤波器,在线识别转子位移信号中与转速同频的振动分量,并通过反馈对转子的同频振动进行不平衡补偿,以实现最小位移控制。然后,利用广义根轨迹法分析了引入不平衡补偿后转子系统的闭环稳定性。通过切换步长符号,实现转子系统在全转速范围内的闭环稳定运行。最后,分别在频率匹配、频率失配、存在噪声和恒加速四个条件下,将该方法应用于四自由度转子的仿真模型中,并且在磁悬浮高速电机实验平台上进行了相关实验。仿真和实验结果均表明,该方法能够有效地抑制转子的不平衡振动。

For unbalance vibration in magnetically levitated high-speed motors,a modified adaptive variable step-size least mean square(LMS)algorithm was proposed to achieve unbalance compensation.Firstly,a formula to get appropriate step size was given by analyzing the influence of step size and frequency on the characteristics of LMS algorithm.Then,with the help of an adaptive filter based on variable step-size LMS algorithm,the minimum displacement control was achieved by online identifying speed-frequency component in displacement signal and adding feedback control.In addition,the stability of closed-loop system was analyzed using generalized root locus.In order to achieve unbalance compensation within the whole speed range,a varying step-size sign strategy was used.Finally,the proposed method was applied to the simulation model of four-degree-of-freedom rotor under the four conditions of frequency matching,frequency mismatching,noise existing and constant acceleration,and relevant experiments were carried out on the magnetically levitated high-speed motor rotor system test platform.Simulation and experimental results show that the proposed method can effectively suppress the imbalance synchronous vibration.