A novel voltage sag state estimation method based on complex network analysis

Abstract

Voltage sag state estimation focuses on estimating the number of voltage sags at unmonitored sites. Existing methods require power quality monitors having to be allocated optimally, which are generally unpractical. To this end, an alternative estimation method is proposed for the scenario where the monitoring scheme is not optimal. Since there is a general rule in a rather long span that some network sites always record more sags than others even when short-circuit fault distribution varies, how power network structure impacts voltage sag is discussed firstly. The strength of interaction among sites, proved to be related to the number of sags, is quantified based on complex network theory analysis. The stronger the interaction between any two sites, the more possible sag events propagate from one site to another. Hence, according to the interaction field between monitored and unmonitored sites, the estimation model is built as the under-determined system of equations. Proposed method has been validated by different IEEE test systems, where the influences of fault distribution and monitoring scheme on estimation results are discussed. The results show that proposed method has better performance especially when the observability of monitoring scheme cannot be ensured. Moreover, the drawback of computation for traditional methods is overcome also, since there is no need to build estimation equations repeatedly.