基于核岭回归的直线电机内模控制系统研究

发布时间：2018-04-29 18:07

本文选题：直线电机 + 核岭回归　；参考：《福州大学》2014年硕士论文

【摘要】：在许多工业控制领域中,常常需要控制被控对象做直线运动,但是由于直线驱动装置发展还不成熟,使得不得不通过旋转电机的旋转运动经机械变换来获得直线运动。直线电机可以直接实现直线驱动,而无需中间机械转换环节,因此具有许多独特的优点。然而也是因为没有中间环节,使得负载扰动、参数摄动以及推力波动等不确定性,无任何缓冲或削弱环节,都直接作用到电机上,将影响系统的伺服性能,难以实现高精度高速的伺服驱动要求。内模控制以其结构简单、设计直观简便、在线调节参数少以及调整方针明确等特点在工业控制领域获得大量应用。然而传统内模控制存在着内模精度差、内模和控制器无法实现良好匹配等缺点,无法实现直线电机的高精度跟踪控制。本文针对直线电机伺服系统对鲁棒性和抗干扰能力的要求,在经典内模控制理论的基础上,设计新型的内模控制系统。本文所设计的内模控制具有如下两个特点：第一,以核岭回归方法建立永磁同步直线电机的动态模型作为内模控制系统的内模,进而通过分析内模的结构,将对逆模控制器的求取过程转换为对非线性方程根的求取过程,以弦截法迭代来实现,这样可使得内模和逆模控制器能够获得很高的匹配精度,并对非线性内模控制系统进行稳定性及稳态误差分析。仿真结果显示,对比传统的基于采样数据来直接构造逆模控制器的内模控制方法,所提出的算法具有更好鲁棒性和更高的控制精度,并对阶跃输入信号实现无稳态误差跟踪。第二,由于直线电机直驱的结构特点,当直线电机跟踪周期性输入信号时,由端部效应造成的与直线电机位移相关的周期性的推力波动以及建模误差会降低系统的伺服性能。针对这一问题,依据第四章所得出的结论：跟踪误差与内模和被控对象的匹配精度直接相关,在第五章将推力波动以及建模误差均看作内模与被控对象的不匹配度,进而设计了基于重复控制原理的补偿器加入到原有的内模控制系统中,对之进行补偿。仿真结果表明,这种控制方案有效地提高了内模控制系统对于周期性输入信号的跟踪精度。
[Abstract]:In many industrial control fields, it is often necessary to control the controlled object to make straight line motion, but the linear drive device is still immature, so it is necessary to obtain linear motion through mechanical transformation of rotating motor. Linear motor can directly realize linear drive without intermediate mechanical conversion link, so it has many unique advantages. However, because there is no intermediate link, the uncertainties such as load disturbance, parameter perturbation and thrust fluctuation, without any buffer or weakening link, all act directly on the motor, which will affect the servo performance of the system. It is difficult to achieve high precision and high speed servo drive. Internal model control (IMC) has been widely used in industrial control field because of its simple structure, simple design, less on-line adjustment parameters and clear adjustment policy. However, the traditional internal model control has some shortcomings, such as poor precision of internal model, bad matching between internal model and controller, and the high precision tracking control of linear motor. Based on the classical internal model control theory, a new type of internal model control system is designed to meet the requirements of robustness and anti-interference ability of linear motor servo system. The internal model control designed in this paper has the following two characteristics: first, the dynamic model of permanent magnet synchronous linear motor (PMSM) is established by using the Kernel regression method as the internal model of the internal model control system, and then the structure of the internal model is analyzed. In this paper, the inverse mode controller is transformed into the solution of the root of the nonlinear equation, which is iterated by the chord section method, so that the internal model and the inverse model controller can obtain high matching accuracy. The stability and steady-state error of nonlinear internal model control system are analyzed. The simulation results show that compared with the traditional internal model control method based on sampling data, the proposed algorithm has better robustness and higher control accuracy, and realizes unsteady error tracking of step input signal. Secondly, due to the structural characteristics of linear motor direct drive, when linear motor tracks periodic input signal, the periodic thrust fluctuation caused by end effect and modeling error will reduce the servo performance of the system. According to the conclusion of Chapter 4, the tracking error is directly related to the matching accuracy of the internal model and the controlled object. In Chapter 5, the thrust fluctuation and the modeling error are regarded as the mismatch between the internal model and the controlled object. Furthermore, a compensator based on the principle of repetitive control is designed to be added to the original internal model control system to compensate it. The simulation results show that this control scheme can effectively improve the tracking accuracy of the internal model control system for periodic input signals.
【学位授予单位】：福州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM359.4

【相似文献】