永磁同步电动机高性能无传感器控制技术研究

发布时间：2018-06-06 19:24

本文选题：永磁同步电动机 + 无传感器控制　；参考：《华南理工大学》2014年博士论文

【摘要】：永磁同步电动机由于具有体积小、功率密度和效率高，运行性能好等优点，在理论研究和实际应用中得到广泛重视；目前，永磁同步电动机作为控制系统执行元件的核心已广泛地应用于数控机床、机器人以及航空、航天和航海等领域中。本文结合电动汽车应用的需要，以无位置传感器永磁同步电动机驱动系统为研究对象，有针对性地对电机转子位置和转速的在线估计、电机的控制策略以及死区补偿等问题进行了较为深入的研究。目的是为了降低电动汽车电气驱动系统的成本与复杂性，进一步提高控制系统的可靠性和控制性能。在研究课题中，首先根据坐标变换理论推导了永磁同步电动机在两相静止坐标系下的电机模型，并结合扩展卡尔曼滤波器理论设计了相应的转子速度、位置观测器。虽然扩展卡尔曼滤波算法能实现电机的自启动，但由于转子初始位置角未知，算法在启动中可能会出现的收敛错误和失速问题，本文针对这一问题给出了详细解释并讨论了相关的解决方法。由于噪声协方差矩阵对估计性能有很大的影响，本文通过仿真分析了不同矩阵取值对结果产生的影响，并总结了一套参数试凑方法。由于扩展卡尔曼滤波器的估计精度受电机模型参数变化影响，通过仿真总结了参数变化对估计精度影响的规律。针对这一问题，对自适应渐消扩展卡尔曼滤波器进行了较为深入的研究。引入衰减因子对原扩展卡尔曼滤波器的误差协方差矩阵进行加权，这样能够减小陈旧量测值对估计的影响，强化新的量测数据在滤波中所起的校正作用，，从而能提高跟踪速度和估计精度。考虑到卡尔曼滤波器在高阶时计算量大的问题，引入一种两段式结构将扩展卡尔曼滤波器分解成两个并行的低阶滤波器，达到节省运算量的目的，通过乘法和加法运算量的对比体现出两段扩展卡尔曼滤波器在运算量上所具有的优势，利用滤波器之间的等效性验证了所提出滤波器的稳定性。结合自适应扩展卡尔曼滤波器和双段扩展卡尔曼滤波器各自的特点，提出一种新的自适应双段扩展卡尔曼滤波器，并采用相同的等效性证明验证了其稳定性，这种滤波器是将双段结构应用到自适应扩展卡尔曼滤波器上而得出，同时具有自适应滤波器强跟踪、鲁棒性好和双段滤波器节省运算量的优点。在电流控制中，针对已有的线性比例微分控制策略存在的动态响应速度慢，对控制器参数的依赖度高等问题，采用一种无差拍预测电流控制方法来进行永磁同步电动机的电流控制。由于这类基于模型的控制方法对参数精确度要求较高，设计了扰动观测器来估计未建模的不确定项，针对电压型逆变器中的死区时间和非线性等因素造成的电压损失，通过相应的死区电压观测器在线估计，并将两个观测器的估计值加入到电压指令值中进行补偿。由于速度环PI控制器的非线性饱和特性，提出了一种变结构抗饱和PI速度控制器来提高转速控制性能。针对自适应双段扩展卡尔曼滤波器和无差拍预测电流控制方法，设计了基于Expert3系统的全数字无位置传感器永磁同步电动机控制系统。在此基础上，对各研究内容进行了深入的仿真研究和实验验证。
[Abstract]:Permanent magnet synchronous motor (PMSM) has been widely used in theoretical research and practical application because of its advantages of small size, high power density, high efficiency and good operating performance. At present, permanent magnet synchronous motor (PMSM), as the core of the actuating components of control system, has been widely used in the fields of CNC machine tools, robots, aerospace and navigation. In this paper, based on the needs of the application of electric vehicle, the research object of the permanent magnet synchronous motor drive system without position sensor is the research on the on-line estimation of the rotor position and speed, the control strategy of the motor and the dead zone compensation. The purpose is to reduce the electric drive system of the electric vehicle. The cost and complexity of the system further improve the reliability and control performance of the control system.
In the study, the motor model of the permanent magnet synchronous motor in the two phase stationary coordinate system is derived according to the coordinate transformation theory, and the corresponding rotor speed and position observer are designed with the extended Calman filter theory. Although the extended Calman filter algorithm can realize the self starting of the motor, the initial position angle of the rotor is due to the initial position angle of the rotor. Unknown, the convergence error and stall of the algorithm may appear in the start up of the algorithm. This paper gives a detailed explanation of this problem and discusses the related solutions. Because the noise covariance matrix has a great influence on the performance of the estimation, this paper analyzes the effects of the different moment array values on the results by simulation and summarizes a set of results. Parameter test method.
Because the estimation accuracy of the extended Calman filter is influenced by the change of the parameter of the motor model, the influence of the parameter variation on the estimation accuracy is summarized by simulation. In this problem, the adaptive fading and expanding Calman filter is deeply studied. The error covariance of the attenuating dependent son on the original extended Calman filter is introduced. The difference matrix is weighted, which can reduce the influence of the old measurement value to the estimation, strengthen the correction function of the new measurement data in the filtering, and thus improve the tracking speed and the estimation precision. Considering the problem that the Calman filter has a large computation in the high order, a two segment structure is introduced to decompose the extended Calman filter into the problem. Two parallel low order filters achieve the purpose of saving the computation. By comparing the multiplication and addition operations, the advantages of the two extended Calman filter are demonstrated. The stability of the proposed filter is verified by the equivalence between the filters. The adaptive extended Calman filter and the two segment expansion are combined. A new adaptive double segment extended Calman filter is proposed, and its stability is verified by the same equivalence proof. This filter is used to apply the double segment structure to the adaptive extended Calman filter, and has the strong robustness and two segments of adaptive filter with strong tracking of adaptive filter. The filter saves the advantage of the amount of operation.
In current control, the current linear proportional differential control strategy has a slow dynamic response and a high dependence on the parameters of the controller. A current control method is used to control the current of the permanent magnet synchronous motor. The disturbance observer is designed to estimate the unmodeled uncertainty. The voltage loss caused by dead time and nonlinear factors in the voltage inverter is estimated by the corresponding dead range voltage observer, and the estimated values of the two observers are added to the voltage instruction value to be compensated. Because of the speed loop PI controller, A variable structure anti saturation PI speed controller is proposed to improve the speed control performance.
In this paper, a full digital sensorless permanent magnet synchronous motor control system based on Expert3 system is designed for the adaptive two segment extended Calman filter and the non difference beat prediction current control method. On this basis, the research contents are thoroughly simulated and tested.
【学位授予单位】：华南理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM341

【参考文献】