Abstract. In order to improve the application area and the prediction accuracy of classical GM (1, 1) and Non Linear Grey Bernoulli Model (NGBM (1, 1)), a Fourier Grey model FRMGM (1, 1), and Fourier Non Linear Grey Bernoulli Model (abbreviated FRMNGBM (1, 1)) are proposed in this paper. These proposed models were built by using Fourier series to modify their residual values. To verify the effectiveness of these proposed models, the total coal consumption demand in China during period time from 1980 to 2012 was used to exam the forecast performance. The empirical results demonstrated that the accuracy of both GM (1, 1) and NGBM (1, 1) forecasting models after using Fourier series revised their residual error provided more accuracy than original ones. Furthermore, this paper also indicated that the FRMNGBM (1, 1) is the better model with MAPE=0.003.