Skip to Main Content
The daily clearing price curve in electricity market varied with multi-period and strong fluctuation characteristic. When grey GM (1, 1) model is used in forecast, the forecast error exceeded the permitted precision. This is because GM (1, 1) model is invalidated only if the price series did not follow the rule of exponential growth. In this case, grey model with period residual modification is proposed, which inherits the advantages of grey model and makes the forecasting price curve fluctuated. Meanwhile, a series of technology is used, such as smooth processing to original data, improvement of initial condition and period residual modification. Thus the fitting curve is closer to original data and the forecasting precision is improved. Simulation results verified the feasibility of the proposed approach.