Predicting performance of SiO2 nanoparticles on filtration volume using reliable approaches: application in water-based drilling fluids

Abstract

ABSTRACT Designing the drilling fluid is crucial in order to make a thin and impermeable filter cake on the wellbore that minimizes the drilling fluid filtration volume. Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Interference System (ANFIS), and Least Square Support Vector Machine (LSSVM) modeling are used to predict the drilling fluid filtration volume as a function of drilling fluid properties, SiO2 nanoparticles, and KCl salt concentration. The performance of the proposed models is evaluated using statistical parameters including Average Absolute Relative Error (AARE), Mean Square Error (MSE), and Correlation Coefficient (R2). The results show that the models could predict the filtration volume well and are in agreement with the actual measured values. The proposed ANN, ANFIS, and LSSVM models could predict the filtration volume with R2 values of 0.9926, 0.9900, and 0.9940, respectively. This indicates that the presented models are accurate enough for predicting the nanoparticles performance on filtration volume of water-based drilling fluid.