A hybrid annual power load forecasting model based on generalized regression neural network with fruit fly optimization algorithm

Accurate annual power load forecasting can provide reliable guidance for power grid operation and power construction planning, which is also important for the sustainable development of electric power industry. The annual power load forecasting is a non-linear problem because the load curve shows a non-linear characteristic. Generalized regression neural network (GRNN) has been proven to be effective in dealing with the non-linear problems, but it is very regretfully finds that the GRNN have rarely been applied to the annual power load forecasting. Therefore, the GRNN was used for annual power load forecasting in this paper. However, how to determine the appropriate spread parameter in using the GRNN for power load forecasting is a key point. In this paper, a hybrid annual power load forecasting model combining fruit fly optimization algorithm (FOA) and generalized regression neural network was proposed to solve this problem, where the FOA was used to automatically select the appropriate spread parameter value for the GRNN power load forecasting model. The effectiveness of this proposed hybrid model was proved by two experiment simulations, which both show that the proposed hybrid model outperforms the GRNN model with default parameter, GRNN model with particle swarm optimization (PSOGRNN), least squares support vector machine with simulated annealing algorithm (SALSSVM), and the ordinary least squares linear regression (OLS_LR) forecasting models in the annual power load forecasting.