基于容积卡尔曼滤波的发电机动态状态估计  

Dynamic State Estimator for Synchronous Machines Based on Cubature Kalman Filter

作　　者：陈亮 毕天姝 李劲松[2] 薛安成 李强[2] 杨奇逊 

机构地区：[1]新能源电力系统国家重点实验室(华北电力大学),北京市昌平区102206 [2]中国电力科学研究院,北京市海淀区100192

出　　处：《中国电机工程学报》2014年第16期2706-2713,共8页Proceedings of the CSEE

基　　金：国家重点基础研究发展计划项目(973项目)(2012CB 215206);国家自然科学基金项目(51222703);高等学校博士学科点专项科研基金(20120036110009)~~

摘　　要：同步相量测量单元(phasor measurement unit,PMU)能够对动态过程中同步发电机功角进行直接量测,但随机误差`PMU存在量测误差,从而对应用造成不良后果。提出一种机电暂态过程中发电机动态状态估计新方法。基于发电机四阶动态方程建立了发电机动态状态估计模型;给出了过程噪声误差计算方法;由于动态状态估计模型为非线性,进而提出基于容积卡尔曼滤波(cubature Kalman filter,CKF)的动态状态估计方法,利用球面-径向规则生成Cubature点,通过发电机动态方程对Cubature点进行变换,得到发电机状态量预报值。仿真分析利用某实际大区域电网同时实现了基于CKF和无迹卡尔曼滤波(unscented Kalman filter,UKF)的动态状态估计。对这两种方法的估计性能指标进行对比分析,结果表明CKF法状态估计效果和计算效率均优于UKF法状态估计。Phasor measurement unit （PMU） can measure the angle of synchronous machines in dynamic processes. However, the random errors may cause the errors of PMU measurement. The unfavorable consequences may be caused by the PMU applications. A novel dynamic state estimation for synchronous machines in the electromechanical transient process is proposed in this paper. The model of dynamic state estimation for synchronous machines is built based on the fourth order swing equation. The computation method for the process errors is given. The dynamic state estimation method based on the cubature Kalman filter （CKF） is proposed for the reason that the dynamic state estimation model is nonlinear. By the spherical-radial rule, the Cubature points are created, which are transformed by the discrete swing equation to compute the prediction of the states. The dynamic sate estimation based on CKF and unscented Kalman filter （UKF） are used in a practical power system respectively. The simulation shows that the filtering performance and the efficiency of the proposed method are superior to the UKF method.

关 键 词：机电暂态 容积卡尔曼滤波 发电机 相量测量单元 动态状态估计 

分 类 号：TM71[电气工程—电力系统及自动化]