Accurate prediction of production levels and estimated ultimate recovery (EUR) is crucial for the development of shale gas reservoirs. Empirical decline methods are widely used in the oil and gas industry due to their simplicity and effectiveness. However, these methods often fail to provide accurate predictions for wells experiencing fluctuating production or transient flow. To address this pressing issue, three empirical methods, namely Arps' decline method, stretched exponential production-decline method, and Duong method are compared based on their principles and characteristics.
The findings demonstrate that the traditional Duong method exhibits the highest reliability. However, to overcome the impact of production fluctuations associated with wells exhibiting boundary-dominated flow (BDF), enhancements are proposed for Duong's method. To illustrate, a new approach is suggested by combining Arps' method with the decline exponent of Duong method. This new approach aims to achieve reasonable production and EUR forecasts for wells experiencing a rapid and unstable decline in TF production.
To evaluate the accuracy and applicability of the new approach, field cases from the Haynesville Shale , Marcellus Shale , and Marcellus-Upper Shale are utilized. The comparative analysis between the new approach and the three aforementioned methods reveals a successful reduction in the error percentage during the transient flow period by 6 to 9% and the BDF period by 9 to 14%. These findings highlight the improved accuracy and applicability of the proposed approach, demonstrating its potential for predicting production and EUR in shale gas reservoirs characterized by fluctuating production and transient flow conditions.