Avaliação probabilística do perfil de tensão e carregamento do transformador de redes elétricas de baixa tensão da cidade de Manaus mediante a conexão de GDFV e V2G

Abstract

The modernization of the global power sector toward Smart Grids is driven by the massive insertion of Distributed Energy Resources (DERs), with an emphasis on Distributed Photovoltaic Generation (PVDG) and Electric Vehicles (EVs). However, Low-Voltage Distribution Networks (LVDNs), historically passive and unidirectional, suffer from asset overloading and violations of regulatory voltage profiles. In this scenario, this work proposes an integrated probabilistic methodology to quantify the operational impacts of these technologies and to evaluate the mitigating role of bidirectional intelligent control of vehicle batteries in Vehicle-to-Grid (V2G) mode. The simulation \textit{framework} is developed in the Python programming language, utilizing the \textit{py-dss-interface} library, to control the multiphase network simulator \textit{Open Distribution System Simulator} (OpenDSS). The uncertainty of human mobility behavior (EV connection times for charging) and the variability of the photovoltaic solar resource are stochastically modeled, implementing the Monte Carlo Simulation Method through 300 probabilistic sampling iterations for a 24-hour daily cycle. The experimental \textit{setup} utilizes three real low-voltage networks from the city of Manaus, located in the Lírio do Vale, Planalto, and Jorge Teixeira neighborhoods. The results, processed into hourly \textit{boxplots}, reveal that high penetration levels of DERs cause severe overvoltages of up to 1.09 p.u. during peak solar irradiance periods and severe undervoltages of up to 0.88 p.u. in the early morning due to the coincident effect of vehicle charging. The experiments confirm that the infrastructure with a higher density of consumer units (Lírio do Vale) presents a lower hosting capacity, reaching the thermal ceiling of the distribution transformer with peaks of 185 kW under overloading conditions.