A New Method for Natural Gas Reserves and Production Prediction

The traditional life-cycle prediction and numerical simulation are the most widely used and mature methods in oil and gas reserves and production prediction. Although the life-cycle prediction can effectively control the law of development for limited resources and give good long-term forecasting, it is not conducive to handle short-term development plan due to its limited short-term prediction precision. On the other hand, numerical simulation method has the advantage of better fitting result and more precise short-term prediction, but it only relies on the historical trend and ignores the generally recognized law of life development. With the combination of traditional life-cycle prediction and numerical simulation methods, this paper introduces composite models based on gray system and neural network, both of which yielded satisfactory results in real applications.