An Artificial Intelligence Approach for The Prediction of Mud Rheological Properties of an Oil Based Drilling Fluid

• Author(s): Christiana Akpan Ukem ; Tity Eshiet Jackson ; Unwana Joseph Ekong
• Paper ID: 1708584
• Page: 1253-1263
• Published Date: 21-05-2025
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 8 Issue 11 May-2025

Abstract

This study aims to develop a model using Artificial neural network for quick and easy prediction of plastic viscosity, yield point and apparent viscosity of an oil-based drilling mud. The ANN model was developed using twenty-one datasets obtained from the laboratory. The datasets were fed into the MATLAB R2015a artificial neural fitting toolbox with an architecture of three (3) inputs, one hidden layer of four (4) neurons and three (3) output layer. A feed-forward propagation method with Levenberg-Marquardt training algorithm was used in the prediction of these rheological properties of an oil-based drilling mud. Mean squared error (MSE), average percent relative error (APRE) and coefficient of determination (R2) were used as criteria for evaluating artificial neural network performance. The developed neural network showed a significant match between the predicted and the measured rheological properties with a coefficient of determination (R2) for apparent viscosity, plastic viscosity and yield point as 0.9981, 0.9912 and 0.9932 with the overall R2 of 0.9981. The overall mean squared error (MSE) was 5.0423E-04 with an average absolute percentage error (AAPE) for apparent viscosity, plastic viscosity and yield point as 5.0, 5.25 and 5.03. The ANN trained network was able to predict its own output very closer to the calculated output of the new dataset with the overall coefficient of determination (R2) of 0.9899. The outcome of this research presents a more reliable model and a speedy tool of predicting other rheological properties of a drilling mud.

Keywords

Artificial Intelligence (AI), Artificial Neural Network (ANN), Oil Based Mud (OBM), Rheological Properties

Citations

IRE Journals:
Christiana Akpan Ukem , Tity Eshiet Jackson , Unwana Joseph Ekong "An Artificial Intelligence Approach for The Prediction of Mud Rheological Properties of an Oil Based Drilling Fluid" Iconic Research And Engineering Journals Volume 8 Issue 11 2025 Page 1253-1263

IEEE:
Christiana Akpan Ukem , Tity Eshiet Jackson , Unwana Joseph Ekong "An Artificial Intelligence Approach for The Prediction of Mud Rheological Properties of an Oil Based Drilling Fluid" Iconic Research And Engineering Journals, 8(11)