Energy demand forecasts based on improved Gray neural network algorithm

Abstract

Energy demand forecasting and energy consumption structure analysis were important foundation for energy planning and development of new energy, so energy forecasting result was required as close as possible to the actual value. To improve the prediction accuracy, this article combined back propagation (abbreviated bp) neural network and gray phase to build an improved gray neural network prediction method, and using genetic algorithms to optimize it. The experiment proved that this model has high prediction accuracy, and used it to predict the energy demand in Hebei Province, the results proved the validity of the model. Finally, the article also analyzed the energy structure and the development of new energy based on forecast results in Hebei Province.

DOI : http://dx.doi.org/10.11591/telkomnika.v12i2.3731