Power flow for primary distribution networks considering uncertainty in demand and user connection

Abstract

This paper presents a method for calculating the power flow in distribution networks considering uncertainties in the distribution system. Active and reactive power are used as uncertain variables and probabilistically modeled through probability distribution functions. Uncertainty about the connection of the users with the different feeders is also considered. A Monte Carlo simulation is used to generate the possible load scenarios of the users. The results of the power flow considering uncertainty are the mean values and standard deviations of the variables of interest (voltages in all nodes, active and reactive power flows, etc.), giving the user valuable information about how the network will behave under uncertainty rather than the traditional fixed values at one point in time. The method is tested using real data from a primary feeder system, and results are presented considering uncertainty in demand and also in the connection. To demonstrate the usefulness of the approach, the results are then used in a probabilistic risk analysis to identify potential problems of undervoltage in distribution systems. (C) 2012 Elsevier Ltd. All rights reserved.