异步电机直接转矩控制技术的理论研究

【作者】 李佳钰；

【导师】 徐中；

【作者基本信息】 大连理工大学 ， 机械设计及理论， 2004， 硕士

【摘要】 直接转矩控制(DTC)技术是上世纪八十年代中期发展起来的新技术。它是继矢量控制技术之后，且与之并行发展的一种新型的高性能的交流调速传动的控制技术。本文详细研究了异步电动机的数学模型，并根据传统直接转矩控制存在的缺点以及对电动机调速性能的要求，改变对定子磁链的调节方式，使定子磁链轨迹近似为一圆形，对转矩采用三值调节器进行调节，以实现转矩的快速调节。为了验证所设计的异步电动机直接转矩控制系统的可行性，通过MATLAB构建了直接转矩控制仿真系统，对直接转矩控制方法的特点及其存在的问题进行了深入的理论与仿真分析研究。 在直接转矩控制中定子电阻是一个十分重要的参数，因此检测异步电机的定子电阻值是非常重要。用定子电流向量的幅值可以校正电机运行和定子电阻值，除定子电阻外的任何电机参数变化不会改变定子电流的幅值；因此采用定子电流幅值偏差和定子电流幅值偏差变化率对定子电阻偏差进行辩识是可行的。由于电流幅值误差及其变化率与电阻之间的关系复杂，难以用精确的数学模型表达，本文提出了模糊神经网络对定子电阻进行在线检测的方案，该网络在离线训练之后，可用于对定子电阻的检测。仿真结果表明了模糊神经网络观测器可实现对定子电阻的精确检测，从而提高直接转矩控制系统的低速性能。 无速度传感器转速估计是现代交流传动研究的一个重要方向，本文在探讨了三种主要的转速估计方法之后，着重分析了基于模型参考自适应系统(MRAS)的速度辨识方法，并通过仿真证明了该方法的可行性。更多还原

【Abstract】 Direct Torque Control (DTC) of asynchronous motors, which has been developed in the recent decades, is a powerful control method for motor drives. Following the vector control technique, the quick development of DTC makes itself a new high-powered Alternating Current (AC) driving technique paralleling to the vector control. In this paper, a study of asynchronous motor model is made; according to the faults of traditional direct torque control and the requirement of an asynchronous motor’s control strategy, the author changes the method of regulating stator flux to make its track approximate to a circle as possible as could. To get quickly torque control, a three-value regulator is adopted. To demonstrate the validity of the asynchronous motor’s direct torque control system, a simulation system of DTC is designed through MATLAB; Moreover, the characteristics of the DTC and its existing problems are analyzed in detail.Stator resistance is a significant parameter in the system of DTC; Therefore, it is important to measure the stator resistance of asynchronous motors. Stator current can revise stator resistance, and any motor parameter can’t change stator current expect stator resistance; So stator current error variation and stator current error variation rate are used to tune stator resistance. This paper presents a scheme of estimating the stator resistance on-line with fuzzy-neural network due to their complicated relations. After trained off-line, the network was used to measure the resistance. Simulation results show that this fuzzy-neural network estimator can precisely measure the value of resistance and improve the low-speed performances of DTC system efficiently.The speed estimation without speed sensor is an important trend of modern AC drive. This thesis has discussed several primary algorithm of speed estimation and emphasized on the algorithm of speed estimate based on model reference adaptive system(MRAS). The results of simulation prove its feasibility.更多还原