感应电动机模型参数在线辨识的UKF算法

Unscented Kalman Filter algorithm for on-line identification of parameters of induction motor model

针对高阶非线性动态系统参数估计的非线性特征,介绍了无味卡尔曼滤波(UKF)算法。在给出了UKF的算法描述的基础上,从一般意义上讨论了无味变换(UF)仅用有限的参数来近似随机变量的概率统计特征,避免了传统的通过线性化来估计非线性系统而带来的误差,进而将该算法用于电力系统感应电动机动态负荷模型的参数估计。算例利用某电网同步相量测量(PMU)采集数据,利用所提算法实时跟踪模型参数,结果表明该算法能够实时有效地辨识出感应电动机负荷模型的参数,有望在实际工程中得到应用。

For the nonlinear characteristics of parameter estimation for high-order nonlinear dynamic systems, the Unscented Kalman Filter （UKF） algorithm is introduced. The UKF algorithm description is provided and the probability and statistics features of Unscented Transformation （UT） which uses limited parameters to approximate the random variables are discussed. Also, the error in the traditional estimation by linearization of nonlinear system is avoided. The algorithm is applied to the parameter estimation of induction motor dynamic load model in power systems. Results of case study clearly indicate that the algorithm can quickly and efficiently identify the parameters of this induction motor dynamic load model, and is expected to be implemented in practical engineering operations.