Research Of Short-term Power Price Forecasting Model In Electric Power Market

The power price is the key factor in the electric power market, when market conception came into electric industry, this key factor became more complex,. It is affected by a lot of elements, for example, the weather, load and something else.Price forecast is one of the most important thing in all commodity markets. In the lately years, electric power was exchanged between different power market as special good. As electric power can not be stored effectively, transmission block obstructs the free exchange, power price forecasting is more difficult. Compare with load, power price is more active, and is affected by more elements. Many uncertain elements are very difficult to monitor and quantify, and it's impossible to input these factors to the math model of price forecasting.The principle, current status and development of the electric power market short-term and mid-long-term price forecasting are generalized in this thesis. Varieties of traditional and modern prediction techniques for price forecasting are summarizes, and the differences and features of these methods are also emphasized. Based on the research of artificial neural network(ANN), analogous theory, fuzzy theory and many clustering theories. Owning to the traits of electric power market short-term price forecasting, a analogous theory with ANN and dynamic fuzzy neural network(DFNN) model in price forecasting are proposed, the corresponding program of these price forecasting model is developed and applied to two power market.The results show they are credible and practicable in short-term price forecasting. Compare with the forecasting result of ANN, these models can get a more precise result, especially they forecast the trend of the price change. At the end of the paper, the main problems in the research of short-term price forecasting are analyzed and the further work is pointed out as well.