感应电动机矢量控制系统转速辨识算法的研究

发布时间：2018-01-04 06:11

本文关键词：感应电动机矢量控制系统转速辨识算法的研究　出处：《哈尔滨工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：基于矢量控制的感应电机交流调速系统由于其调速范围宽、动态响应快等特点,越来越多的被运用在各种行业中。在矢量控制系统中需要进行转速闭环控制,以达到对转速的精确控制。但位置传感器具有价格高、不易维护等缺点,因此无速度传感器技术就能很好的解决这些问题。然而对感应电机的速度辨识各种算法都存在不同的优缺点,为此本文围绕三相感应电动机转速辨识算法做了如下工作:根据感应电机不同坐标系下的数学模型,分析坐标变换和磁场定向原理,设计定子电压解耦电路,从而实现定子励磁电流分量和转矩电流分量进行解耦,再设计带位置传感器的感应电机矢量控制系统,矢量控制系统良好的控制特性为研究感应电机转速辨识提供基础。研究传统的EKF和MRAS电机转速辨识算法,对比两种算法在转速辨识稳态精度和动态特性,得出MRAS算法效果更好。在此基础上,研究改进型MRAS算法,针对当参考模型为电压模型时,存在纯积分环节会导致误差累积等问题,研究基于混合磁链模型和基于定子电流模型的MRAS算法,跟传统MRAS算法相比,基于混合磁链模型的算法对定子电阻变化和直流偏置有更好的鲁棒性,而基于定子电流的MRAS算法除了对定子电阻影响比较小外,在转速辨识过程中,动态跟随性能更好,然后将改进后的算法运用在无位置传感器的矢量控制系统中,仿真验证系统具有跟带位置传感器一样的控制效果。设计出以TMS320F28335为核心的硬件实验平台,在软件开发平台上,设计整个算法的主程序与中断程序,最终在实验平台上进行转速稳态和动态实验,验证了研究的算法能够准确地辨识出电机转速。
[Abstract]:The vector control of induction motor system due to its wide speed range based on fast dynamic response characteristics, are used in a variety of industries more and more. The vector control system of speed closed-loop control, to achieve the precise control of speed. But the position sensor has the disadvantages of high prices, not easy to maintain. Therefore, no speed sensor technology can solve these problems very well. However, the speed identification algorithms of induction motor have different advantages and disadvantages, this paper around the speed identification algorithm of three-phase induction motor as a work: according to the mathematical model of induction motor in different coordinates, the coordinate transformation and the analysis of the principle of field orientation, design the stator voltage decoupling circuit, thus realizing the stator excitation current component and torque component of current decoupling, design of induction motor vector control with position sensor System control characteristic vector control system provides the basis for good speed identification of induction motor. The study of traditional EKF and MRAS motor speed identification algorithm, comparing the two kinds of speed identification algorithm in steady state accuracy and dynamic characteristics, the MRAS algorithm has a better effect. Based on the research of improved MRAS algorithm, for the reference model when the voltage model exists, the pure integrator will lead to error accumulation problem based on hybrid flux model and MRAS algorithm based on stator current model, compared with the traditional MRAS algorithm, robust hybrid flux model algorithm is better on the variation of the stator resistance and DC bias based on the MRAS algorithm based on the stator current in addition to the impact the stator resistance is relatively small, in the process of rotor speed identification, dynamic tracking performance is better, and then use the improved algorithm in sensorless vector control In the system, the simulation system has the same effect as a control with position sensor is designed. The hardware platform with TMS320F28335 as the core, in the software development platform, the main program design of the algorithm and interrupt program, final speed steady state and dynamic experiments in the experimental platform, verify the algorithm can accurately identify the motor speed.

【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】：2016
【分类号】：TM346;TP273

【相似文献】