有铁芯直线电机推力波动的分析与补偿方法研究

发布时间：2018-04-13 12:23

本文选题：有铁芯直线电机 + 推力波动　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：当今时代,精密运动平台在各行各业中都受到广泛应用,例如IC制造业、航天航空业、精密机械加工业等。其中,永磁直线电机在高速度、高精度、高载荷的运动控制中应用最为广泛。直线电机不需要机械传动装置实现从旋转运动到直线运动的转换,因此大大降低了电机系统非线性和摩擦力扰动的影响,但同时也带来了更多的模型不确定性和更强的外部扰动的问题,如推力波动问题。因此,若要获得高速度、高精度的运动控制,无论是通过恰当地结构设计还是有效的系统控制,降低上述问题带来的影响都是至关重要的工作。本文以尖端精密设备——光刻机中的双工件台系统为研究背景,针对该系统宏动台X向有铁芯直线电机存在的推力波动问题,进行了深入的理论分析和扰动补偿方案研究。首先,本文从电机结构层面对产生直线电机推力波动的主要原因进行了理论分析。并对研究对象——工件台宏动X向电机进行扫频实验,通过频率特性得出了较为准确的被控对象模型,并针对该模型进行了反馈控制系统设计,为后续工作打下基础。其次,本文利用已有实验平台结构对定位力和纹波扰动进行了数据采集实验,并利用获得的数据进行了模型辨识,得到了二者基于RBF神经网络的辨识模型。利用辨识模型进行查表补偿,实验结果表明针对研究对象设计其它有效补偿方案是必要的。再次,本文通过理论分析、仿真研究和实验研究,对迭代学习补偿方法进行了深入研究。实验结果表明,迭代学习补偿方法对周期性推力波动具有较好的抑制能力,但其不能解决部分呈现非周期性的推力波动。最后,本文深入研究了扰动观测器补偿方法,并提出了一种改进策略。实验结果验证了改进扰动观测器对扰动抑制的有效性和优异性。最后还提出了一种扰动观测器+迭代学习补偿策略,可以缩短迭代周期的同时提高系统跟踪精度。
[Abstract]:Nowadays, precision sports platform is widely used in various industries, such as IC manufacturing, aerospace industry, precision machinery processing industry and so on.Among them, permanent magnet linear motor is widely used in motion control with high speed, high precision and high load.The linear motor does not need mechanical transmission to realize the conversion from rotating motion to linear motion, so the influence of nonlinear and friction disturbance of the motor system is greatly reduced.But it also brings more model uncertainty and stronger external disturbance, such as thrust fluctuation.Therefore, in order to obtain high speed and high precision motion control, it is very important to reduce the influence of the above problems, whether through proper structural design or effective system control.In this paper, based on the research background of double workpiece table system in the sophisticated equipment-lithography machine, a deep theoretical analysis and disturbance compensation scheme are carried out to solve the problem of thrust fluctuation in the X direction linear motor with iron core on the macro platform of the system.Firstly, the main causes of thrust fluctuation of linear motor are analyzed theoretically from the motor structure level.The frequency sweep experiment of workpiece macro moving X direction motor is carried out, and a more accurate controlled object model is obtained by frequency characteristic, and the feedback control system is designed for the model, which lays a foundation for the follow-up work.Secondly, this paper makes use of the existing experimental platform structure to carry on the data acquisition experiment to the localization force and the ripple disturbance, and uses the obtained data to carry on the model identification, obtains the identification model based on the RBF neural network.The experimental results show that it is necessary to design other effective compensation schemes for the object of study.Thirdly, through theoretical analysis, simulation and experimental research, the iterative learning compensation method is deeply studied.The experimental results show that the iterative learning compensation method has a good ability to suppress periodic thrust fluctuations, but it can not solve the partial non-periodic thrust fluctuations.Finally, the disturbance observer compensation method is studied and an improved strategy is proposed.The experimental results show that the improved disturbance observer is effective and excellent for disturbance suppression.Finally, an iterative learning compensation strategy for disturbance observer is proposed, which can shorten the iteration period and improve the tracking accuracy of the system.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273

【参考文献】