Method for predicting recoverable reserves of shale gas well based on early yield data

Abstract

The invention provides a method for predicting recoverable reserves of a shale gas well based on early production data. The method comprises the steps: 1, preparing historical yield data of the shale gas well; 2, using a beta function method for flowing stage division and data processing; 3, fitting the yield of the shale gas well in each production period according to a power exponent decline curve, and predicting an upper limit value of the recoverable reserves according to a fitting curve; 4, predicting a lower limit value of the recoverable reserves of the shale gas well in each production period according to a daily gas yield linear extrapolation model; 5, respectively fitting the upper limit value and the lower limit value of the recoverable reserves with a exhaustion decline model, and determining a predicted upper limit value corresponding to the upper limit value of the recoverable reserves and a predicted lower limit value corresponding to the lower limit value of the recoverable reserves; 6, determining an upper and lower limit interval of the recoverable reserves according to the predicted upper limit value and the predicted lower limit value of the recoverable reserves. According to the method, the predicted recoverable reserves are more reliable by using the early yield.

Description

technical field [0001] The invention relates to the technical field of shale gas exploitation, and more specifically, to a method for predicting recoverable reserves of shale gas wells based on early production data. Background technique [0002] Predicting recoverable reserves by using early production data of shale gas wells is a very important task in shale gas development. For shale reservoirs, due to the characteristics of tightness and desorption, the production cycle lasts for decades, and the flow in the formation will be in the unsteady flow stage for a long time, so it is very difficult to predict the recoverable reserves. Especially in the early stage of development, conventional prediction methods, such as flow mass balance, Arps, Fetkovich, Blasingame, Agarwal-Gardner and NPI, etc., do not meet the conditions for the flow to reach a quasi-steady state. Although many prediction models for the recoverable reserves of shale gas wells have been developed, such as t...

AI Technical Summary

Problems solved by technology

However, this method does not quantitatively describe the change law of recoverable reserves in different time periods, and the reliability of the prediction results is not good.

In addition, this method uses the production data of all time periods in the early stage, and the production data of the drainage stage, which has a greater impact on the prediction results, is not excluded, causing the prediction results to deviate from the true value

Images