Perpustakaan Digital ITB

The global demand for natural gas is steadily increasing over time. This increasing demand has to be answered by the increasing gas production. However, challenges emerge due to the difficulty of accurately predicting well performance. The performance of a well holds a pivotal role in the oil and gas industry. Much of the profit a company will gain also depends on the performance of a well being produced. The sooner we know the performance of wells, the sooner the decision and development must be made on the well. In this study, the authors undertake a comparative analysis of four models formulated by Ansah et al (1996) aimed at predicting well performance. Ansah et al's (1996) study seeks to forecast future well performance by correlating viscosity-compressibility and dimensionless reservoir pressure. The models scrutinized in this research encompass zero-order polynomial models, first-order polynomial models, exponential models, and general-order polynomial models. Several assumptions are inherent to this study, particularly concerning the reservoir model, assuming uniform porosity within each reservoir and gas flow adhering to a boundary-dominated condition. The model established by Ansah et al (1996) is juxtaposed against actual data acquired from the CMG software simulator. Furthermore, the authors augment Ansah et al's (1996) model by introducing variables for each model's constant gas rate. The parameters under comparison in this study encompass gas rate, dimensionless reservoir pressure, bottom hole pressure for a constant gas rate, and cumulative production. Based on these parameters, the comparison between Ansah et al's (1996) pre-existing model and the actual data simulated using the CMG software reveals that the general-order polynomial model exhibits the least error.