Using Artificial Intelligence Techniques in Modeling and Predicting the Rheological Behavior of Nano-Based Drilling Fluids

Abstract

Abstract Drilling process is one of the main operations in the extraction of hydrocarbons from petroleum reservoirs. It comes right after the exploration processes. Drilling fluids are necessary for controlling the wells and performing different functions during the drilling operation. They perform many roles in lifting the cuttings from the bottom of the well to the surface and cooling/lubricating the drill pipes and bit. Furthermore, they provide the desired hydrostatic pressure to overbalance pore pressure in addition to produce a thin/impermeable filter cake that can prevent or reduce the possible damage to the formations. It is mandatory to keep monitoring, enhancing, and optimizing the properties of the drilling fluids. Recently, different additives, among which nanoparticles (NPs), have been investigated to improve, and maximize the benefits of the drilling fluids accordingly to meet the new challenges. The rheological behavior of such complex fluids has shown different enhancements up on the utilization of those additives. The rheological properties of the drilling fluids are accurately measured on the surface; however, the behavior of those properties may change with time and under harsh drilling conditions, such as high pressure/high temperature environments. For that, different models are introduced and used to predict and optimize the rheological characteristics of such fluids. Bingham, Herschel-Bulkley, Power Law, Casson and others are commonly used as rheological models to predict the drilling fluid behavior. In the last decade, a new trend of developing new models and correlations using the artificial neural networks (ANN) have been introduced to the petroleum field. Mathematical formulas can be developed using ANN, which then can be used to predict the behavior of certain parameter(s) by knowing other ones. Using ANN have shown to be more reliable and accurate in predicting the rheological properties of the drilling fluids, such as apparent viscosity (AV), plastic viscosity (PV), yield point (YP), maximum shear stress, and change in the mud density at various conditions. This work aims at using ANN technique to develop suitable models that can predict the rheological behavior of nano-based drilling fluids. The effect of NPs-type, -size, -concentration, and drilling fluid formulations will be considered, which may pave the road for new applications and efficient utilizations.