An improved grey model for short-term electricity price forecasting in competitive power markets with punishment function

In this paper, an improved GM (1,2) model for short-term price forecasting in competitive power markets with particle swarm optimization algorithm (PSO) and punishment function method (PFM) is proposed. Considering each historical data has different impact extent to forecasting value, thus the punishment function is constructed with adjustable factor; Furthermore, considering the influence of grey background-value, the PSO algorithm is adopted to optimize the punishment function factor and the grey background value weight parameter. Thus the improved forecasting model is founded. The historical data from the Nordpool power market is used for computing, and the numerical results demonstrate the validity of the improved GM(1,2) model.