Prediction of Downhole Water-Based Mud Density Using Artificial Neural Network That's the Title

Student: Nsongurua Akpan Udoh (Project, 2025)
Department of Petroleum Engineering
University of Uyo, Akwa Ibom State

Abstract

With the evolution of technology, data is a major sector of engagement in the oil and gas industry. Various studies are done on the subject of artificial intelligence and Artificial Neural Networks are commonly employed. This study seeks to apply Artificial Neural Network to the prediction of downhole density of water-based mud. Artificial Neural Network (ANN) from MATLAB was used to predict the density and the total data of 117 gotten were input into the neural network and trained to get a better model. The three statistical parameters used for selecting the neural network with the best predictions were correlation coefficient (R²), the mean square error (MSE) and the root mean square error (RMSE). In comparison with existing models, the ANN model developed in this study performed better. The sensitivity analysis performed shows that the initial mud density has the greatest impact on the final mud density downhole. Furthermore, this method eliminates the need for surface measurement equipment, while at the same time, representing more accurately the downhole mud density at any given pressure and temperature.

Keywords
neural density artificial network downhole better model square error performed
Filters
Institutions