A Novel Three-Phase Dynamic State Estimation Method for Low-Voltage Distribution Network Based on Data Fusion Technology
TL;DR
Numerical results indicate that the proposed LDN three-phase DSE method consistently and uniformly exhibits sustained improvements in state estimation accuracy and computational efficiency across various tested scenarios.
Abstract
: To bridge the last mile of situational awareness in the end-section networks of smart distribution grids and achieve precise three-phase dynamic state estimation (DSE) for low-voltage distribution network (LDN), by taking into account both phasor measurement units (PMU) in LDN and smart meters (SM) within the advanced metering infrastructure, a novel three-phase dynamic state estimation method for LDN based on data fusion (DF) technology is proposed in this paper. Firstly, this method constructs a three-phase four-wire network model for LDN that incorporates the impact of the neutral wire, as well as a state-space model capable of describing the dynamic characteristics of loads and distributed generation. Subsequently, a multi-source heterogeneous DF framework is designed. During the filtering process for each snapshot, the method establishes parallel three-phase DSE models for LDN by combining predicted values with PMU measurements and by combining predicted values with SM measurements, respectively. Based on the results of parallel filtering, DF technology is employed to integrate the states and obtain the final estimated values for that time section. Finally, simulations of the proposed LDN three-phase DSE method are conducted by using a modified IEEE-13 bus system, the numerical results indicate that the proposed method consistently and uniformly exhibits sustained improvements in state estimation accuracy and computational efficiency across various tested scenarios.