Application of machine learning to determine the shear stress and fltration loss properties of nano‑based drilling fuid

Ning, Yee Cai and Ridha, Syahrir and Ilyas, Suhaib Umer and Krishna, Shwetank and Dzulkarnain, Iskandar and Abdurrahman, Muslim (2023) Application of machine learning to determine the shear stress and fltration loss properties of nano‑based drilling fuid. Journal of Petroleum Exploration and Production Technology. pp. 1031-1052.

Text Application of machine learning to determine the shear stress and filtration loss properties of nano-based drilling fluid.pdf - Published Version Download (7MB)

Abstract

A detailed understanding of the drilling fuid rheology and fltration properties is essential to assuring reduced fuid loss during the transport process. As per literature review, silica nanoparticle is an exceptional additive to enhance drilling fuid rheology and fltration properties enhancement. However, a correlation based on nano-SiO2-water-based drilling fuid that can quantify the rheology and fltration properties of nanofuids is not available. Thus, two data-driven machine learning approaches are proposed for prediction, i.e. artifcial-neural-network and least-square-support-vector-machine (LSSVM). Parameters involved for the prediction of shear stress are SiO2 concentration, temperature, and shear rate, whereas SiO2 nanoparticle concentration, temperature, and time are the inputs to simulate fltration volume. A feed-forward multilayer perceptron is constructed and optimised using the Levenberg–Marquardt learning algorithm. The parameters for the LSSVM are optimised using Couple Simulated Annealing. The performance of each model is evaluated based on several statistical parameters. The predicted results achieved R2 (coefcient of determination) value higher than 0.99 and MAE (mean absolute error) and MAPE (mean absolute percentage error) value below 7% for both the models. The developed models are further validated with experimental data that reveals an excellent agreement between predicted and experimental data.

Actions (login required)

View Item