A Data Mining Approach To Support The Development Of Long-term Load Forecasting
Free (open access)
M. R. Maia, K. de Oliveira Gonçalves Veloso, M. T. Okamoto, A. dos Santos Rigueira, G. M. Tavares, Â. M. Cister, M. A. F. Zarur, F. T. de Souza, G. S. Terra, A. G. Evsukoff & N. F. F. Ebecken
Load forecasting is an important subject for power distribution systems and has been studied comparing different points of view. In general, load forecasts should be performed over a broad spectrum of time intervals, which could be classified into short-term, medium-term and long-term forecasts. Several research groups have proposed various techniques for either short-term load forecasting or medium-term load forecasting or long-term load forecasting. This paper presents two approaches for modelling the long-term load forecasting: a neural network (NN) and a non-linear (cause/effect) model. The data used by the models are gross domestic product (GDP), the national minimum salary, the electrical energy price, the estimated national population and the total number of electrical connections. The suitability of the proposed approach is illustrated through a long-term load forecasting application (electricity consumption in Brazil ten years ahead). Keywords: neural networks; long-term; load forecasting; power distribution systems. 1 Introduction Brazil’s electric power system (EPS) consists of two major interconnected systems (SIN) and many small isolated systems. The energy is mainly generated
neural networks; long-term; load forecasting; power distribution systems.