Medium- and long-term demand estimation integrating emerging technologies

Electric demand estimation and analysis are crucial foundations for seeking the best investments, foreseeing the difficulties, and preparing the power grid for the future. Still, it is difficult to anticipate the uncertain and varying factors (environmental, behavioral, cultural, economic, seasonal, and technological) that affect the demand pattern over long periods, especially when there is a lack of reference or historical data. There is also a need for flexible and robust methodologies for load estimation to explore the effect caused by the increasing integration of grid edge technologies on the load curve pattern. A novel approach, compatible with conventional and contemporary modeling techniques such as regression (often referred to as econometric) and artificial neural networks, is proposed to assess the impact of emerging technologies on the load pattern of future power systems. The proposal consists of defining the necessary steps to generate hourly models of the daily demand curve that simulate changes in demand patterns corresponding to different scenarios of penetration of renewable energy sources and emerging technologies. The proposed methodology was applied with regression and artificial neural network techniques to demonstrate its compatibility with parametric and non-parametric modeling techniques.