Obtaining the bottomhole flowing pressure of a producing well from readily available surface pressures had been a significant concern for operators in the petroleum industry, as accurate knowledge of this pressure is crucial for determining the most efficient recovery methods and lifting procedures. Although many existing correlations aim to achieve this, their predictive capabilities are limited due to the inability of current models to account for sand particles in the flow stream and the need to shut in the well for bottomhole pressure predictions, which seems counterproductive. This study introduces a data-driven approach to determine of the flowing bottomhole pressure of a vertical well using surface and well parameters. Existing models and correlations provide insights into the relationship between flowing bottomhole pressures and wellhead pressure, while artificial feedforward neural networks, random forest decision trees and support vector machine algorithms are employed to develop regression models based on available field data. Evaluation metrics such as mean squared error and mean absolute error are used to assess the performance of these machine learning models. The artificial neural network performed best on both training and testing data-sets, predicting the flowing bottomhole pressures with a mean squared error of 7.5% and a mean absolute error as low as 3.9% on the test set. This model offers advantages in estimating flowing bottomhole pressure from real-time surface pressures and well data compared to empirical models that rely on simplifying assumptions.
Flowing Bottom-Hole Pressure, Wells, Reservoir, Hydrocarbon, Well Completions, Production
IRE Journals:
Ikeh Lesor , Osigwe Uche Stanley
"Predicting The Flowing Bottom-Hole Pressure of a Vertical Well Using Surface Pressure and Well Parameters" Iconic Research And Engineering Journals Volume 9 Issue 4 2025 Page 540-552
IEEE:
Ikeh Lesor , Osigwe Uche Stanley
"Predicting The Flowing Bottom-Hole Pressure of a Vertical Well Using Surface Pressure and Well Parameters" Iconic Research And Engineering Journals, 9(4)