Avaliação probabilística do impacto da conexão de veículos elétricos na estabilidade de tensão de redes de distribuição

Abstract

Global demand for reducing carbon emissions has caused profound changes in several industry sectors, for example: electricity generation, heating/ cooling and transport. In transport sector, the main initiative to reduce the emission of pollutants is the use of Electric Vehicles (EV). The EV have a battery-powered propulsion system that is recharged by connecting the EV to electric grid. In some countries there are already targets for the replacement of all combustion vehicles by EV. However, high EV penetration will have major impacts on the operation of the electricity distribution network, for example: lifetime loss of power transformers, harmonics, long-term voltage variations and voltage instability. Voltage stability is defined as the ability of the system to keep voltages controllable after the occurrence a disturbance, for example: equipment failures and load variations. In conventional distribution networks, the occurrence of voltage stability problems is less frequent because there are few voltage control devices (on-load tap changing transformers, voltage regulators and capacitor banks), and the load growth is small in areas with high load density (urbanized and built). In contrast, modern distribution networks are more prone to voltage instability problems due to the following factors: (i) high penetration of conventional and renewable distributed generation (solar and wind) connected to the grid via inverters; (ii) coordination problems between conventional voltage control devices and distributed generators; (iii) sudden load increasing due to EV charging in urbanized areas. It is important to note that the output power of the renewable distributed generators and the demand of the electric vehicles are stochastic in nature due to the variability of the primary energy source and the charging habits of the consumers, respectively. Therefore, the voltage stability analysis in modern distribution networks is subject to uncertainties. In this way, it is not possible to use deterministic techniques to carry out a voltage stability assessment in distribution networks with massive insertion of EV and renewable distributed generation. This dissertation aims to carry out an assessment of the impact of EV on the voltage stability of distribution networks. This assessment will be carried out using probabilistic techniques to model uncertainties associated with: seasonal variations in load, intermittence of solar photovoltaic distributed generation and demand resulting of EV charging. The proposed probabilistic model is based on the combination of the following techniques: Sequential Monte Carlo simulation (to model uncertainties), power flow via Gauss-Zbus Method (to calculate nodal voltages) and D´ Matrix Method (to calculate the voltage stability margin). In addition, the effects of the following voltage control devices in the voltage stability margin are considered: on-load tap changing transformers and reactive power dispatch (of solar photovoltaic distributed generators and EV charging stations). The proposed methodology was tested in the UKGDS medium voltage system by comparing different modes of EV charging. The results showed that the recharge mode in private parking is less vulnerable to voltage instability problems compared to residential recharge due to the concentration of the load in less points, which can facilitate the operation of the voltage control and minimize the reinforcements in the distribution network. In this context, investing in the private recharge mode seems to be a more promising solution in relation to the voltage stability impacts of the distribution network.