The back propagation neural network and the gray model for predicting natural gas consumptionhave been introduced in the present article. Two new methods for predicting natural gas consumptionin China have also been pointed out based on the back propagation neural network and the gray model. The experimental results show that both models can predict natural gasconsumption in China and the experimental data are in agreement with the quantitative analytical conclusions drawn from the calculated data. This proves that two types of new models can be used to predictnatural gas consumption. It results in good economic and social benefits in China.