无迹卡尔曼滤波及其平方根形式在电力系统动态状态估计中的应用  

Application of UKF and SRUKF to Power System Dynamic State Estimation

作　　者：卫志农[1] 孙国强[1] 庞博[1] 

机构地区：[1]河海大学能源与电气学院,江苏省南京市210098

出　　处：《中国电机工程学报》2011年第16期74-80,共7页Proceedings of the CSEE

基　　金：国家自然科学基金项目(50877024)~~

摘　　要：针对扩展卡尔曼滤波(extended Kalman filter,EKF)的不足,将不需要对非线性系统函数进行线性化的无迹卡尔曼滤波(unscented Kalman filter,UKF)方法引入电力系统动态状态估计,采用生成Sigma点数量最少的比例最小偏度单形采样策略进行无迹变换。以IEEE 14系统为算例,仿真结果表明引入UKF后,估计结果的精度有所提高,但算法的效率较低,且数值稳定性较差。进一步引入平方根形式的UKF(square root UKF,SRUKF)模型,IEEE 14及IEEE 30测试系统的仿真结果证明:在不需要大量牺牲计算时间的同时,算法的数值稳定性得到了改善。表明SRUKF的引入对动态状态估计方法的改进是有效的。Aiming at the shortcomings of the extended Kalman filter （EKF）, the unscented Kalman filter （UKF）, which avoids the linearization of the nonlinear system function, is introduced into power system dynamic state estimation. The sampling strategy called scale-corrected minimal skew simplex sampling is adapted so the least Sigma points are generated in the unscented transform process. For the IEEE 14-bus system, the estimation accuracy is improved, while the efficiency is lower and the numerical stability is poorer than EKF. Then, the square root UKF （SRUKF） is introduced. Simulations are carded out for the IEEE 14-bus system and IEEE 30-bus system, which show that the calculating time is saved and the numerical stability is improved. The introduction of SRUKF model is effective for improving dynamic state estimation approach.

关 键 词：电力系统 动态状态估计 扩展卡尔曼滤波 无迹 卡尔曼滤波 平方根形式的无迹卡尔曼滤波 

分 类 号：TM74[电气工程—电力系统及自动化]