Distributed Kalman Filter for Large-Scale Power Systems With State Inequality Constraints
Zhijian Cheng, Hongru Ren, Bin Zhang, Renquan Lu
Abstract
This article is concerned with a hybrid distributed dynamic state estimation (DSE) algorithm for large-scale power grids. Based on the mixed phasor measurement unit (PMU) and remote terminal unit measurements model, a modified distributed Kalman filter (KF) is designed. Different from the centralized KF algorithm, the distributed approach is capable of independently estimating local states by local measurements. Moreover, in each local region, the multiple missing measurements problem is considered in the modified distributed KF algorithm design. The internodal transformation theory is employed to deal with the communication problem between the distributed subsystems. Therefore, the proposed method can reduce the communication latency while ensuring the estimation accuracy. Considering the inequality constraints, the particle swarm optimization algorithm and the probability-maximization method are applied to tackle the corresponding constrained estimation issue. The proposed distributed DSE algorithm is tested on an IEEE benchmark 14-bus system to demonstrate its effectiveness and applicability.