A novel grey model to forecast short-term electricity price for Nordpool using particle swarm optimization and correlation hours method

Short-term electricity price forecasting in competitive power markets is essential both for producers and consumers in planning their operations and maximizing their benefits. This paper proposed a new grey model, called PGM(1,2), based on Particle Swarm Optimization algorithm (PSO) and correlation hours method (CHM) in order to forecast short-term price in the Nordpool market. The main sequence is composed of prediction period price data and the reference sequence is composed of hour-before period price data. Considering of the influence of grey background, the PSO is adopted to optimize the grey background weight parameters, thus the PGM (1,2) forecasting model is founded. Comparison of forecasting performance of the PGM (1,2) with that of the traditional GM (1,1) and GM (1,2) is presented. Simulation results demonstrate the validity of the PGM (1,2) model.