准确的动态状态估计对于电力系统实时运行状态的监测至关重要.针对同步发电机中非高斯噪声导致状态估计器性能下降的实际情况,提出一种基于基于柯西核最大相关熵(Cauchy kernel maximum correntropy,CKMC)的容积卡尔曼滤波(cubature KF,CKF)算法(简称CKMC-CKF算法).首先,建立CKMC目标函数,采用2种加权局部相似度来更新噪声协方差矩阵,从而降低不良数据的权重;其次,利用线性化回归方程统一目标函数中的状态和测量误差,并通过定点迭代法获得最佳估计状态.最后,以IEEE 39节点系统为算例分析验证所提出方法的有效性.与CKF和最大熵CKF相比,CKMC-CKF在非高斯噪声环境下具有更好的估计精度和更强的鲁棒性.
Dynamic State Estimation of Generators Based on CKMC-CKF Under Non-Gaussian Noise
Accurate dynamic state estimation is very important for real-time monitoring of power system operation state.In view of the fact that the performance of the state estimator is degraded by non-Gaussian noise in the synchronous generator,this paper proposes a cubature Kalman filter(CKMC-CKF)algorith based on the maximum correntropy of the Cauchy kernel.Firstly,the maximum correntropy objective function of Cauchy kernel is established,and the noise covariance matrix is updated by two weighted local similarities,so as to reduce the weight of bad data.Secondly,the state and measurement errors in the objective function are unified by linearized regression equation,and the best estimation state is obtained by fixed-point iterative method.Finally,the effectiveness of the proposed method is verified by an example of a standard IEEE 39 bus system.Compared with the CKF and the maximum correntropy CKF,the proposed method has better estimation accuracy and stronger robustness in non-Gaussian noise environment.
generatordynamic state estimationcubature Kalman filtermaximum correntropynon-Gaussian noise
万方数据
维普
