Price forecasting using an integrated approach

In this paper, a BP model for short-term price forecasting and a linear regress model for mid/long-term price forecasting are described together with forecasting results. A detailed discussion on the choice of forecast models and forecast variables is reported. A salient feature of the reported methods is that the forecast models can take into account the influence of market power on electricity prices. Specifically, a market power index, namely Must Run Ratio (MRR), is judiciously selected as an input to the price forecast models. An added advantage of presented models is that the number of required input variables can be reduced significantly without perceived loss of accuracy. The suggested methods are being utilized, by the transmission company, to forecast short-term, mid/long-term prices in Zhejiang Electricity Market. The results show that the proposed forecast models meet the basic requirement of Zhejiang electricity market operation.