Integration of artificial neural networks and genetic algorithm to predict oil demand in China

An artificial neural network (ANN) using a genetic algorithm (GA) was designed to predict future oil demand in China. The back propagation network was used to characterize the relationship between gross domestic product; population; oil price; the number of motor vehicles; and oil demand in China. The GA was used to determine the number of neurons in the ANN's hidden layer as well as the momentum and learning rates of the back propagation algorithm. Results of the study showed that China's oil demand can be accurately forecast using the ANN. Results showed that the growth rate of oil demand will be approximately 3.1 percent annually, which is lower than rates predicted by the Energy Information Administration and the International Energy Agency (IEA). The lower percentage was attributed to the impacts of improved energy efficiency, the use of renewable energy sources, and unexpected changes in the Chinese economy over the last 3 decades. 22 refs., 6 tabs., 4 figs.