The objective of this study is to assess the impact of reservoir and completion parameters on the b-factor, which is a critical parameter in decline curve analysis (DCA) for unconventional reservoirs. The b-factor determines the long-term decline behavior and has the highest uncertainty among decline curve parameters. By investigating the impact of various reservoir and completion parameters on the b-factor, this study aims to provide a range of b-factors and guidance on DCA parameters for different fields and scenarios.
Initially, an automatic decline-curve fitting process is applied to all Permian unconventional horizontal wells, resulting in the generation of DCA parameters for each well. The range of b-factors associated with various reservoir benches and fields are reviewed. The changes in the b-factor trend over time are then compared to theoretical expectations.
Subsequently, physical models are developed to assess the sensitivity of the b-factor to specific reservoir and completion parameters. Two types of models are employed: a hybrid model rate transient analysis (RTA) and a 3D numerical simulation model. These models allow for the isolation of individual parameters, such as porosity, saturation, well spacing, and fracture designs, to evaluate their impact on the b-factor.
In theory, the b-factor changes with time and flow regime. Specifically, it transitions from a value of 2 during linear flow to less than 1 during boundary flow. However, in practice, a constant b-factor is often used for the entire lifespan of a well, typically ranging from 0.8 to 1.2, with an average value of around 1. By employing both production analysis and physical models, the study identifies the most impactful parameters that influence the b-factor, providing valuable insights into the range of b-factors for different reservoirs and scenarios.
Decline curve analysis (DCA) serves as the primary method for forecasting existing well production. Developed by J. J. Arps in 1944, it is an empirical approach that involves fitting a line to historical performance data and assuming this trend will persist into the future. The beauty of DCA lies in its simplicity: it only requires information on time and production rate. This versatility allows it to be applied across various systems, making it a proven technique for more than half a century.
