ANÁLISE PROBABILÍSTICA DE VARIAÇÕES DE TENSÃO DE CURTA DURAÇÃO EM REDES DE DISTRIBUIÇÃO

Abstract

In the whole world, the restructuring of the electrical sector promoted by the deregulation of the electric utilities has brought economical benefits for the society; furthermore there has been an increase of the pertinent interest in the subject of power quality. The competitive environment created by the deregulation forces the electric companies of generation and transmission to redefine their objectives in order to reduce their operation costs and to maximize their profits to guarantee their survival. Some of the measures adopted to reach these objectives, such as the postponement of the maintenance and investments for the expansion in the electric network, have as direct consequence the deterioration on the levels of reliability and quality of the supply. In distribution systems, this situation is more critical due to the monopolist nature of the electrical utilities. In other words, the consumers are subjected to the lucrative interests of these companies and, for that reasons, in several countries, the regulatory entities have been establishing goals for the reliability and quality of the supply aiming to punish the distribution utilities that violate these goals. In this new environment the interest of the electric utilities, consumers and regulatory entities for the quality of energy have been increasing due to the economical impact experienced by them. One of the problems of the quality of energy are the voltage sags effects, which has increased in recent years due to the crescent use of new equipments, which are more sensitive to voltage variations. The aim of this work is to develop a probabilistic methodology to analyse voltage sags in distribution networks. With the proposed methodology, power quality índices associated with voltage sags can be estimated in distribution systems, considering the system s topology and failure data. Due to the stochastic nature of the frequency and duration of fault, the proposed methodology is based on probabilistic models that permit to represent the system´s operation under faults. The proposed methodology is based on the combination of the Monte Carlo Method with the Admittance Summation Method. The Monte Carlo Method with sequential simulation is used to sample the failure scenarios and the Admittance Summation Method to evaluate fault currents. These fault currents are evaluated using a phase coordinate model of the distribution system. This model has been used due to the distribution networks unbalance caused by loads and untransposed overhead lines. Furthermore, voltage sags duration has been estimated using the current×time curves of the protection devices (fuses, breakers, reclosers). The proposed methodology has been tested in a feeder belonging to the Maranhão Electric Utility(CEMAR) and the results indicate that the proposed methodology is a good tool for studies of sags in distribution systems because it permits the evaluation of the system s performance against voltage sags, also estimating the compatibility between the system and the customer equipments. The results are shown through probability distributions of the SARFI%V, SARFICBEMA and SARFIITIC índices; cumulative and individual probability distributions, scattering diagrams and coordination charts of the voltage sags.