Neural network models for electricity prices and loads short and long-term prediction

Abstract

In the framework of competitive electricity market, prices and load forecasting has become a real challenge for engineers in electric power systems. The increase of production from renewable sources increased number of factors that affect the cost and power consumption. On the other side, smart grids and the development of computers enable the use of artificial intelligence for solving such problems. This paper presents the multiple uses artificial neural networks (ANN) in long-term and short-term forecasting electricity prices and loads. Databases that are used for training ANN contain hours and thirty minutes data from British and Serbian power system. Data are related to the production of different energy sources, import / export of energy, temperature and load diagram form. Trained ANNs are used to predict energy prices and the formation of dynamic trifling for Serbia. Formed ANN models can be used for real time, on-line prediction of load and electricity price.