Development of New Rheological Models for Class G Cement with Nanoclay as an Additive Using Machine Learning Techniques

Abstract

The rheology of the oil well cement plays a pivotal role in the cement placement. Accurate prediction of cement rheological parameters helps to monitor the durability and pumpability of the cement slurry. In this study, an artificial neural network is used to develop different models for the prediction of various rheological parameters such as shear stress, apparent viscosity, plastic viscosity, and yield point of a class G cement slurry with nanoclay as an additive. An extensive experimental study was conducted to generate enough data set for the training of artificial intelligence models. The class G oil well cement slurries were prepared by fixing the water–cement ratio to 0.44 and adding organically modified nanoclays as a strength enhancer. The rheological properties of the oil well cement slurries were investigated at a wide range of temperatures (37 ≤ T ≤ 90 °C) and shear rates (5 ≤ γ ≤ 500 s –1 ). Experimental data generated were used for the training of feed-forward neural networks. The predicted values of the rheological properties from the trained model showed a good agreement when compared with the experimental values. The average absolute percentage error was less than 5% in both training and validation phases of modeling. A trend analysis was carried out to ensure that the proposed models can define the underlying physics. From the validation and the trend analysis, it was found that the new models can be used to predict cement rheological properties within the range of data set on which the models were trained. The proposed models are independent of laboratory-dependent variables and can give quick and real-time values of the rheological parameters.

A Machine Learning Based Accelerated Approach to Infer the Breakdown Pressure of the Tight Rocks

“…These three ML techniques were Decision Trees (DT), Random Forest Regressor (RF), and K-Nearest Neighbor (KNN). Machine learning and artificial intelligence is making great progress in the oil and gas industry, with various researchers investigating advance applications pertaining to complex/heterogenous systems (Konoshonkin et al, 2020;Ma et al, 2018;Mohaghegh et al, 1994;wu zy and liu, 2018;Zhang et al, 2005Zhang et al, , 2012 (Elkatatny et al, 2016;Tariq, 2018;Tariq et al, 2021Tariq et al, , 2020cTariq et al, , 2020dTariq et al, , 2019aTariq et al, , 2019bTariq and Mahmoud, 2019). A wide range of ML algorithms have been employed to develop models/correlations for estimating various parameters related to hydrocarbons development (Ahmadi et al, 2014;Anifowose et al, 2015;da Silva et al, 2005;Janjua et al, 2016;Khan et al, 2019aKhan et al, , 2019bKhan et al, , 2018bKhan et al, , 2018aKhan et al, , 2018cLi et al, 2020;Tariq et al, 2018Tariq et al, , 2016Tohidi-Hosseini et al, 2016).…”

Influence of Various Nanomaterials on the Rheology and Hydration Kinetics of Oil Well Cement

“…Rheology tests are used to describe the workability, pumpability, consistency, flowability, and stability of a cement slurry [34,35], aiding proper cement placement [35]. It is crucial to characterize and optimize the rheological properties of cement slurries in order to improve mud clean-up and provide good zonal isolation [35].…”

“…Rheology tests are used to describe the workability, pumpability, consistency, flowability, and stability of a cement slurry [34,35], aiding proper cement placement [35]. It is crucial to characterize and optimize the rheological properties of cement slurries in order to improve mud clean-up and provide good zonal isolation [35]. The rheology test conducted for this study was performed at atmospheric pressure using an automatic rotational viscometer (model NXNJ) from Tianjin NITHONS Technology CO., Ltd., Tianjin, China.…”

Melt Instability Identification Using Unsupervised Machine Learning Algorithms

“…Tariq et al used artificial neural networks to predict rheological parameters of cement with nanoclay. [16] Saad [17] used supervised and unsupervised machine learning techniques to study binary compounds and to predict their crystal structure and then associate them to a specific class, like photovoltaic, superconductors, and ferromagnetic materials. Contrary to this work, they applied dimensionality reduction techniques whereas here the focus lies on clustering algorithms.…”