Shear Stress and Filtration Loss Properties Assessment of Nano-Silica Water-Based Drilling Fluid Using Machine Learning Approaches

Abstract

Abstract A complete overview of the rheology and filtration properties of drilling fluids is essential to ensure an efficient transport process with minimized fluid loss. Silica nanoparticle is an excellent additive for rheology and filtration properties enhancement. Existing correlations are not available for nano-SiO2-water-based drilling fluid that can extensively quantify the rheology or filtration loss of nanofluids. Thus, two data-driven machine learning approaches are proposed for prediction, i.e., artificial neural network (ANN) and least square support vector machine (LSSVM). Parameters involved in the prediction of shear stress are SiO2 concentration, temperature, and shear rate, whereas SiO2 nanoparticle concentration, temperature, and time are the inputs to simulate filtration volume. A feed-forward multilayer perceptron is constructed and optimized using the Levenberg–Marquardt learning algorithm. The parameters for the LSSVM are optimized using couple simulated annealing (CSA). The performance of each model is evaluated based on several statistical parameters. The predicted results achieved R2 (coefficient of determination) value higher than 0.99 and mean absolute error (MAE) and mean absolute percentage error (MAPE) 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.