Load forecasting, the importance of the probability “tails” in the definition of the input vector

• Main Author: Santos, P. J.
• Publication Date: 2013
• Other Authors: Rafael, Silviano, Pires, A. J.
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.26/4884
• topic: Transport and distribution electrical network; Load forecasting; Smart-Grids; Input vector; Regressive methods; Load behaviour

The load forecast is part of the global management of the electrical networks, namely at the transport and distribution levels. This type of methodologies allows to the system operator, to establish and take some important decisions concerning to the mix production and network management, with the minimum of discretionarity. The load forecast in particularly the peak load forecast, represents an important economic improvement in the global electrical systems. Also in certain circumstances, allow reducing the contribution of the non-renewable units, in the daily mixing production. The regressive methodologies specially the artificial neural networks, are normally used in this type of approaches, with satisfactory results. In this paper is proposed a careful analysis in order to define the best-input vector in order to feed the regressive methodology. It was establish careful analyses of the load consumption series. It makes use of a procedural sequence for the pre-processing phase that allows capturing certain predominant relations among certain different sets of available data, providing a more solid basis to decisions regarding the composition of the input vector to ANN. The methodological approach is discussed and a real life case study is used for illustrating the defined steps, the ANN and the quality level of the results.