知识与财富的链接
噪声统计特性和模型参数的不确定性,会严重影响动态状态估计的精度。针对该问题,文中提出了一种基于H∞容积卡尔曼滤波(HCKF)的动态状态估计新方法。首先,建立发电机动态状态估计模型;其次,依据H∞滤波理论构造模型不确定性约束准则,并在容积卡尔曼滤波(CKF)中依据该准则计算更新估计误差协方差阵,抑制参数不确定性对状态估计精度的影响;最后,通过对IEEE 10机39节点系统和某实际大区域电网系统的算例测试,将所提方法与CKF方法和改进插值扩展卡尔曼滤波(IEKF)方法的估计性能进行对比。算例仿真结果表明,HCKF方法在估计精度和对模型不确定性的鲁棒性方面较CKF和IEKF方法均有所提高,能够有效抑制模型不确定性对发电机动态状态估计的影响。
The uncertainties of noise statistics and model parameters will seriously affect the accuracy of dynamic state estimation. To deal with this issue, a new dynamic state estimation approach is developed based on H-infinity cubature Kalman filter (HCKF). Firstly, the dynamic state estimation model of generator is established. Secondly, a constraint criterion for model uncertainties is developed by utilizing H-infinity filtering theory.
On this basis, the estimation error covariance matrix in the cubature Kalman filter (CKF) can be updated to suppress the adverse effects on the precision of state estimation caused by parameter uncertainties. Finally, the performance of the proposed method is compared with the CKF method and an improved interpolation extended Kalman filter (IEKF) method in IEEE 10-machine 39-node system and a practical large-area power system. Simulation results demonstrate that HCKF method performs better than CKF and IEKF methods in estimation precision and robustness against model uncertainties, which can restrain the influences of model uncertainties on the dynamic state estimation for generators.
