-Short-Term Load Forecasting using a Gaussian Process Model- Optimal Endogenous Regressor

The presence of an electrical de-regulated market reinforces the need of forecast. Actions like network management, load dispatch and network reconfiguration under quality of service constraints, require reliable load forecasts. This paper establishes a methodological approach based on a Gaussian Process Model in order to choose an efficient endogenous information to be included in the input vector of the forecast. The proposed approach was tested on real-load from three medium-sized supply electrical distribution substations located in the center of Portugal. Considering two types of updating (hourly and daily) for the used load information, the aim will be focused on establishing the optimal regressor vector that can produce the best support load diagrams of prediction. The obtained final results are in accordance with the best values of expected errors for these types of methodologies. The last procedures reveal extremely importance and utility in activities such as planning, operating and controlling electric energy systems.