基于UKF的无速度传感器永磁同步电机矢量控制系统的研究

发布时间：2018-04-26 16:26

本文选题：扩展卡尔曼滤波 + 无迹卡尔曼滤波　；参考：《昆明理工大学》2017年硕士论文

【摘要】：永磁同步电机就转动惯性而言相对较小,在提高运行效率的同时还可以获得高动态性能等特点,使得这类电机广泛的应用在控制系统中[1]。针对永磁同步电机的控制策略中因为矢量控制技术可以对电流励磁和转矩分量存在的耦合性进行解耦,使得该控制技术被广泛应用。但是机械传感器的存在无疑会增加系统的运行成本。与此同时为了检测数据的精确度,要求高精度传感器与永磁同步电机的转轴具有较高的同轴度,这就为安装和维护带来了极大的考验。实际生产控制过程中的环境对高精度传感器有着重要的影响。针对以上问题,无速度传感器技术的出现可以很好的解决以上问题并且该技术在实际控制生产过程中已经得到初步的实现。该技术的理论基础就是获得输入电机绕组端的电压和电流等信号,再对电机转子端的位置和速度等信号采用固定的估计方法在线计算,上述经过估计方法计算出的信号通过坐标变换技术处理后通过反馈环节,最后经过脉宽调制技术处理生成控制系统所需要的控制信号,以实现控制永磁同步电机的目的[2]。本文首先利用扩展卡尔曼滤波算法(EKF)的思想搭建永磁同步电机的矢量控制系统模型以实现该控制系统的无速度传感器技术。该模型能够验证该控制系统在不同的控制时段以及不同的控制条件下,该控制系统均能够完成对电机转速的实时控制。但是该控制系统的实际转速和估计转速之间存在着较大的转速误差而且转速误差平方根也是很大。这种情况下,提高控制系统的滤波精度以达到减小两者之间的转速误差成为研究的重点。针对以上问题,提出基于无迹卡尔曼滤波算法的矢量控制模型,仿真结果验证该控制系统相比EKF的矢量控制系统能够很好的实现减小实际转速与估计转速的转速误差。这就表明该控制系统可以提高系统的转速跟踪效果从而可以很好的提高该控制系统的控制性能。但是在研究基于UKF的矢量控制系统模型时,发现其算法程序单次运行时间过长从而使得控制系统容易导致滤波发散的问题。为了解决以上问题,建立基于AISR-UKF算法的矢量控制系统仿真模型。通过仿真结果验证该算法能够在有效降低算法程序单次运行时间的同时保证系统滤波精度较高。因为实际转速与估计转速之间的误差在各个时段相较UKF更小,算法程序单次运行时间比EKF短从而说明该算法计算复杂性较低运算量较小。
[Abstract]:The permanent magnet synchronous motor (PMSM) is relatively small in rotation inertia. It can also obtain high dynamic performance and so on while improving operation efficiency. This kind of motor is widely used in the control system for the coupling of current excitation and torque component in the control strategy of [1]. for permanent magnet synchronous motor because of vector control technology. The control technology is widely used, but the existence of the mechanical sensor will undoubtedly increase the operating cost of the system. At the same time, in order to detect the accuracy of the data, the high precision sensor and the rotating shaft of the permanent magnet synchronous motor have a high coaxiality, which has brought a great test for the installation and maintenance. The environment in the process of production control has an important influence on the high precision sensor. In view of the above problems, the emergence of the speed sensorless technology can solve the above problems well and the technology has been preliminarily realized in the actual control production process. The theoretical basis of this technology is to obtain the voltage and the voltage of the input motor winding end. The signals, such as current and other signals, are calculated online by fixed estimation methods, such as the position and speed of the rotor end of the motor. The signals calculated by the estimation method are processed through the feedback link through the coordinate transformation technology, and the control signals needed by the control system are generated by the pulse width modulation technology to achieve the control of permanent magnet. The purpose of synchronous motor [2]. is to build a vector control system model of permanent magnet synchronous motor by using the idea of extended Calman filter algorithm (EKF) to realize the speed sensorless technology of the control system. This model can verify that the control system can be used in different control periods and under different control conditions. The real time control of the motor speed is completed. However, there is a large speed error between the actual speed and the estimated speed of the control system and the square root of the speed error is very large. In this case, it is the key to improve the filtering precision of the control system to reduce the speed error between the two. The vector control model based on the untracked Calman filter algorithm is proposed. The simulation results show that the control system can reduce the actual speed and speed error of the estimated speed compared to the vector control system of EKF, which shows that the control system can improve the speed tracking effect of the system so that the control system can be improved well. But in the study of the model of vector control system based on UKF, it is found that the single operation time of the algorithm is too long so that the control system can easily lead to filtering and divergence. In order to solve the above problems, a simulation model of vector control system based on AISR-UKF algorithm is established. In order to effectively reduce the single operation time of the algorithm, the accuracy of the system is higher, because the error between the actual speed and the estimated speed is smaller than the UKF in each period, and the single operation time of the algorithm is shorter than that of the EKF, which shows that the computational complexity of the algorithm is smaller.

【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM341;TP273

【参考文献】