A novel hybrid prediction model for aggregated loads of buildings by considering the electric vehicles

Abstract

In this paper, a new prediction model for aggregated loads of buildings has been propose. Due to high correlation of prediction performance with related horizons and aggregated more customers, a new strategy is developed to provide a forecasting model based on high accuracy. While, consumption shape of a single users normally has low structure to be exploited and vice versa. So, combining different users develops the relative prediction performance up to special point. Outside this point, no more enhancement in relative performance can be found. This model is useful in optimal power system operation and planning in micro-grids. In this paper, besides the aggregated loads of building, the electric vehicle (EV) impact on network has been considered. While, by considering the load growth prediction and impact of EV adaption on load curves this problem can be an important issue for the power grid operation and management. So, an accurate prediction model is presented in this work which is composed of new feature selection and enhanced support vector machine (ESVM) based forecast engine. Effectiveness of the proposed model is carried out to the part of Budapest city through comparing with other prediction models. Obtained results demonstrate the validity of proposed method.