作　　者： ; ; ; ; (赵先娴）;

机构地区： 华南理工大学电力学院

出　　处： 《电工技术学报》 2014年第9期110-118,共9页

摘　　要： 高性能内置式永磁同步电机的无传感器控制器需要精确的定子电阻值,将定子电阻作为变量加入到扩展卡尔曼滤波器中进行辨识将使运算矩阵阶数偏大,增加系统的运算负担。为解决这一问题,针对内置式永磁同步电机设计一种非线性两段扩展卡尔曼滤波观测器。该算法在原扩展卡尔曼滤波器的基础上通过坐标变换将其分解成为两个并行的低阶卡尔曼滤波器。所提出的算法是原扩展卡尔曼滤波器的分解运算,两者在数学上完全等价。理论分析和实验结果表明,与原有传统卡尔曼滤波器相比,该算法减少了一定的运算量并保持良好的辨识效果。 The precise stator resistance is very important when designing high performance sensorless controller for Interior permanent magnet synchronous motor（IPMSM）.Using extended Kalman filter（EKF） to estimate rotor position,speed and stator resistance will be calculated with high order operational matrix and make the real system be heavy computation.To solve this problem,a novel Kalman control strategy by using nonlinear two-stage extended Kalman filter（NTSEKF） is proposed.Based on coordinate transformation,the main idea of NTSEKF is to decouple the EKF into two parallel low order filters.The NTSEKF is decomposition operation of EKF,and both of them are mathematically equivalent.Theoretical analysis and experimental results show that the novel strategy can reduce the number of arithmetic operations compared to the conventional extended Kalman filter and keep good performance of identification.

关 键 词： 内置式永磁同步电机 无传感器控制 扩展卡尔曼滤波器 非线性两段扩展卡尔曼滤波

领　　域： [电气工程]