Uncertainty evaluation in load flow analysis: real MV distribution networks case studies

Abstract

This work investigates the measurement uncertainty impact on load flow analysis in medium voltage (MV) distribution networks. The considered algorithm is a simplified load flow algorithm (LFA), developed by authors, which is based on the load power measurements at each secondary substation and one voltage measurement at the slack bus (i.e. the voltage at the MV bus bars of the primary substation). In the viewpoint of a real implementation, to reduce monitoring system costs, the LFA makes use of low voltage (LV) load power measurements for all the substations except those of MV users, where MV transducers are usually already installed. The uncertainties on LFA input quantities (load powers and slack bus voltage) are calculated, considering actual values of loads powers and currents and the accuracy specifications of measurement instruments installed in distribution network. Starting from this, power flows uncertainties are obtained applying a Monte Carlo analysis. The analysis has been carried out for real case studies; i.e. the distribution networks of Favignana and Ustica Islands. The following results refer to the Favignana case.