Modeling Electricity Price Forecast with Grey and Correlation Method in Competitive Markets

This paper proposed a novel grey model (GM) called PSOGM(1,2,g) for short-term electricity price prediction based on Particle Swarm Optimization algorithm (PSO) and correlation method (CM) in competitive power markets. In the presented grey model, the reference sequence (RS) is defined and determined by CM. Furthermore, considering of the influence of grey background, PSO is adopted to optimize the grey background weight parameter. To demonstrate the superiority of the provided model, publicly available data obtained from Nordpool market in North Europe is used for training and testing. Simulation results show that the improved model has higher precision than traditional grey models, and is capable of forecasting short-term price efficiently in competitive power markets.