Metodologias para modelagem de cargas de consumidores de baixa tensão considerando a integração de resposta da demanda, geração distribuída e veículos elétricos

Abstract

Accurate load modeling is a crucial task in distribution systems expansion planning. Traditionally, the load peak, which is viewed as the worst-case scenario, has been used to quantify new investment requirements. However, under the influence of the Distributed Generation, Demand Response, and Electric Vehicles, the load is becoming an Active Demand. In these conditions, the characteristics of the worst-case scenario may change as a result of the intermittent behavior of the renewable generation, the uncertainty in the consumers’ response to price signals as well as the uncertainties in electric vehicles charging. This thesis proposes new models for active low voltage consumers and electric vehicles on distribution systems expansion planning studies. In these models, the load uncertainty is considered by establishing different patterns for the behavior of LV consumers in the presence of DR programs. The load consumption is segmented according to the different uses of energy to stimulate behavioral adjustments based on the preferences and gains of different types of consumers. Electrical Vehicles charging is modeled considering different charging strategies in order to characterize different types of consumers. A case study based on the modified IEEE 33 Bus test system with real data collected from a Brazilian distribution company is performed in order to analyze the impact of new Load Profiles (LPs) in scenarios with high penetration of renewable DG. Optimal 5-year expansion plans for AD quantiles were obtained using the metaheuristic EPSO (Evolutionary Particle Swarm Optimization) combined with nonlinear programming. Different incentive policies for AD are also analyzed to determine their impact on DS expansion planning. The experiments carried out reveal that considerable monetary savings in the DS can be achieved (up to 37%) as compared with the alternative with no AD by exploiting the flexibility associated with the active behavior of consumers and electric vehicles, by responding to price signals and by permitting adequate levels for the integration of DG into distribution grids. In addition, the results demonstrate that extreme scenarios of DR and/or DG penetration can result in investment expenditures greater than in the alternative with no AD, allowing to identify the best policy for DR and the optimal DG penetration level that result in the lowest investment cost.