Análises de estimação de estados em um sistema de distribuição utilizando o software computacional digsilent powerfactory

Abstract

Obtaining real-time data for a safe and reliable state of the electricity distribution systems is of paramount importance, aiming at proper operation and generation of an accurate diagnosis in relation to probable irregularities that will affect the reliability of the system. A composite system with real-time analysis can monitor, control, operate and coordinate actions, thus forming a smart grid. The use of smart meters allocated over the network is an alternative for acquisition of remote measurements in real time, but even with the current reduction in costs for the acquisition of meters, it is still not feasible to have complete network coverage in terms of measures. Thus, there is a need to merge remote measurements with other pseudo-measures, previously acquired or determined by the Power Flow. The estimation of states is a way of performing a statistical study that complements the Power Flow and the actual measurements obtained by the meters, thus comprising real measures with pseudo measures. Therefore, this work seeks, through three case studies, to perform a validation and analysis of the state estimation - applied to a Low Voltage network - to integrate the estimates obtained by the estimator to the remote measurements in real time. It also seeks to determine a general index of the estimator, which can determine if the meters are well situated, so that, if the opposite, it helps in future analysis, to find possible allocation sites for them.