【作者】 刘宁；

【导师】 夏长亮；

【作者基本信息】 天津大学 ， 电气工程， 2018， 硕士

【摘要】 永磁同步电机伺服驱动系统以其效率高、性能好、功率密度大等优点,被广泛应用于牵引驱动、航空航天、机器人、电动汽车等诸多工业领域。然而,由于死区效应、磁链谐波和电流采样误差等非理想因素会引入周期性扰动,从而造成转速波动,严重削弱了永磁同步电机驱动系统的稳态控制性能。针对这一问题,本文从传统比例-积分(PI)控制和线性二次最优控制两个角度开展了转速波动抑制算法的研究。针对采用传统PI控制的永磁同步电机驱动系统中转速波动问题,利用卡尔曼滤波对各主要频次转速波动分量进行估算,然后把这些转速波动分量与PI控制器并联,嵌入到转速控制环的前向通道中,并根据系统对周期性扰动转矩的衰减程度,选取各主要频次转速波动分量对应的比例增益补偿值,从而对传统PI型转速控制器进行改进。实验结果表明,该算法能够实现对周期性扰动转矩的抑制,提高稳态控制性能。为了改善基于传统线性二次最优控制在永磁同步电机系统上的稳态性能,降低周期性扰动的影响,本文根据永磁同步电机系统周期性扰动特点及分布,构建包含周期性扰动信息的增广系统状态空间模型,实现对永磁同步电机系统更精确的描述。在此基础上,基于线性二次最优理论求取永磁同步电机系统状态反馈控制律。然后,将扰动观测器与卡尔曼滤波理论相结合设计周期性扰动观测器,用于估算关于周期性扰动的增广状态变量,从而使状态反馈控制得以实现。实验结果表明,与传统线性二次最优控制相比,本文提出的新型线性二次最优控制算法在不影响动态性能的同时能够获得更好的稳态性能。更多还原

【Abstract】 Permanent magnet synchronous motor(PMSM)drives has been widely applied in aerospace,traction drive,robots,electric vehicles and other industrial fields because of the advantages of high efficiency,good control performance,high power density and so on.The non ideal factors such as dead time effect,flux harmonic and current sampling error can cause the speed ripples,which seriously weakens the steady-state control performance of the PMSM drives.To solve this problem,this paper studies the speed ripple minimization algorithm from two aspects of traditional proportional integral(PI)control and linear quadratic optimal control.Aiming at the problem of speed ripples in PMSM vector control system based on PI controller,this article estimates the main order speed ripple components by Kalman filter,and embeds these estimated components in forward channel of speed loop by paralleling them with the PI controller.Then,according to attenuation degree of periodic disturbance torques,the values of proportional gain compensation corresponding to these speed ripple components are selected.Thus these periodic disturbance torques are suppressed.The experimental results show that the proposed strategy can efficiently improve the steady-state performance of the PMSM drives.In order to improve the steady-state performance and reduce the influence of non ideal factors for PMSM drives based on the traditional linear quadratic optimal control,this article establishes an augmented state space model containing information of periodic disturbances according to the characteristics and distribution of periodic disturbances in PMSM drives.Then,based on the model,the linear quadratic optimal control law of state feedback controller is solved.Because it is incapable to directly measure the augmented state variables related to periodic disturbances in the augmented state space model,this paper designs a periodic disturbance observer to estimate these augmented state variables.Thus the state feedback control is realized.The experimental results show that the proposed method can efficiently improve the steady-state performance of the PMSM drives without sacrificing dynamic-state performance.更多还原