计及零注入约束的电力系统动态状态估计

摘要

已有的动态状态估计大多数都没考虑系统的约束条件,状态估计结果无法满足潮流方程。为此,提出计及零注入约束的电力系统动态状态估计模型,并利用投影无迹卡尔曼滤波(projected unscented Kalman filter, PUKF)算法求解。采用无迹变换技术处理非线性方程,避免了线性化误差；采用投影估计法将状态量估计值投影到约束表面,使估计结果严格满足约束条件。以 IEEE-30节点系统为算例进行仿真分析,并将计及零注入约束的 PUKF算法与不考虑约束的扩展卡尔曼滤波(extended Kalman filter,EKF)和无迹卡尔曼滤波(unscented Kalman filter,UKF)算法进行比较,结果表明基于 PUKF的计及零注入约束的动态状态估计较不考虑约束的动态状态估计,具有更好的状态估计、量测滤波、约束满足和不良数据检测性能。%Most of the existing dynamic state estimation methods do not consider constraint conditions for the system and the estimation results can not satisfy flow equations. Therefore,this paper proposes a dynamic state estimation model for power system with zero injection constraints and uses projected unscented Kalman filter (PUKF)algorithm for solutions. Unscent-ed transformation technology is used for processing nonlinear equations and is able to avoid linear errors. Projected estima-tion method is used for projecting estimated values of state vectors on the constraint surface and ensure estimated results strictly satisfy constraint conditions. IEEE-30 node system is used for simulation analysis,PUKF algorithm considering zero injection constraints,extended Kalman filter (EKF)algorithm and unscented Kalman filter (UKF)algorithm are compared as well. Results indicate that compared with dynamic state estimation without regard to constraints,PUKF-based dynamic state estimation considering zero injection constraints has better performance of state estimation,measurement filter,con-straint satisfaction and bad data detection.