Annual energy consumption forecasting based on PSOCA-GRNN model (Q1722367)

Summary: Accurate energy consumption forecasting can provide reliable guidance for energy planners and policy makers, which can also recognize the economic and industrial development trends of a country. In this paper, a hybrid PSOCA-GRNN model was proposed for the annual energy consumption forecasting. The generalized regression neural network (GRNN) model was employed to forecast the annual energy consumption due to its good ability of dealing with the nonlinear problems. Meanwhile, the spread parameter of GRNN model was automatically determined by PSOCA algorithm (the combination of particle swarm optimization algorithm and cultural algorithm). Taking China's annual energy consumption as the empirical example, the effectiveness of this proposed PSOCA-GRNN model was proved. The calculation result shows that this proposed hybrid model outperforms the single GRNN model, GRNN model optimized by PSO (PSO-GRNN), discrete grey model (DGM (1, 1)), and ordinary least squares linear regression (OLS\(\_\)LR) model.