Development of New Models to Determine the Rheological Parameters of Water-based Drilling Fluid Using Artificial Intelligence

It is well known that drilling fluid is a key parameter for optimizing drilling operations, cleaning the hole, and managing the rig hydraulics and margins of surge and swab pressures. Although the experimental works present valid and reliable results, they are expensive and time consuming. On the other hand, continuous and regular determination of the rheological mud properties can perform its essential functions during well construction. More uncertainties in planning the drilling fluid properties meant that more challenges may be exposed during drilling operations. This study presents two predictive techniques, multiple regression analysis (MRA) and artificial neural networks (ANNs), to determine the rheological properties of water-based drilling fluid based on other simple measurable properties. While mud density (MW), marsh funnel (MF), and solid% are key input parameters in this study, the output functions or models are plastic viscosity (PV), yield point (YP), apparent viscosity (AV), and gel strength. The prediction methods were demonstrated by means of a field case in eastern Iraq, using datasets from daily drilling reports of two wells in addition to the laboratory measurements. To test the performance ability of the developed models, two error-based metrics (determination coefficient R2 and root mean square error RMSE) have been used in this study. The current results of this study support the evidence that MW, MF, and solid% are consistent indexes for the prediction of rheological properties. Both mud density and solid content have a relative-significant effect on increasing PV, YP, AV, and gel strength. However, a scattering around each fit curve is observed which proved that one rheological property alone is not sufficient to estimate other properties. The results also reveal that both MRA and ANN are conservative in estimating the fluid rheological properties, but ANN is more precise than MRA. Eight empirical mathematical models with high performance capacity have been developed in this study to determine the rheological fluid properties based on simple and quick equipment as mud balance and marsh funnel. This study presents cost-effective models to determine the rheological fluid properties for future well planning in Iraqi oil fields.

Investigating Nigeria Local Clay for its Marsh Funnel Viscosity and Gel Strength: A Study of Irhodo Bentonite

Due to the daily increasing demand for crude oil fuel for its various capacities of energy production and utilizations, a twin respond of exploration for more hydrocarbon reserves and drilling activities was accompanied. As a result, rate and cost of importing drilling clay (Wyoming Bentonite), a major raw material in drilling mud becomes so high that hundreds of millions of dollar is incurred on company budget (Arinkoola, et al., 2020). Thus, the Federal Government of the federal republic of Nigeria, on sensing the benefits of local content development, then clamour for its use as drilling mud. This therefore becomes the bed-rock which this paper is belt on to investigate local clay for some its properties, like viscosity-gel strength. This paper also reviews the formulation of an equivalent one barrel of a laboratory drilling mud using Irhodo bentonite. This paper report two methods and devices used to determine viscosities; the marsh funnel viscosity method using Marsh Funnel and Fann Viscosity-Gel method using the Rheometer. The result of the experiment for the determination of viscosity using marsh funnel apparatus, for both local and bentonite drilling mud, and show 27.12 and 37.17sec/qt (seconds per quart). And when additives, CMC and guar-gum were added, 27.23sec/qt, 29.47sec/qt and 23.19sec/qt and 29.47sec/qt respectively obtained.

Automated Mud Skid Holds Promise of Safety, Efficiency Improvements With Real-Time Fluid Monitoring

As industry buzzwords go, “automation” has spent its time in oilfield vernacular climbing the ranks of widely used terms. It now resides as one of the go-to designations for signs of advancement in any number of disciplines. Its use has been tied most frequently with drilling operations as contractors look to keep employees out of harm’s way via a robotic take-over of most motion-intensive jobs on the rig’s drill floor—basically anything that grips, clamps, or spins. More recently, the term has moved away from the drill floor and into other well construction operations allowing for things such as remote, real-time measurements without the need for boots on the ground. For areas like west Texas and the Permian Basin shales, having the option for remote readouts and a component of automation that can allow for corrective actions should the need arise can go a long way in terms of safety and efficiency gains as well as better manpower application. Unsurprisingly, the area has become a solid testing ground for new, expanding efforts in automation. With dreams of new drilling-fluid-monitoring automation, Eric van Oort, a professor at The University of Texas at Austin and former Shell research scientist, and select students came up with a new way to automatically measure mud parameters such as viscosity without the use of a traditional viscometer. “The fact that we still use manual measurements, some of them now 90 years old, is quite puzzling in this day and age,” van Oort said. “The Marsh funnel, for instance, was introduced in the 1930s, and other mud tests go back to the 1950s and 1960s. These API measurements have served us well, but the question is, can you do something more now with modern measurement techniques and sensors? So, I started working on new ways of measuring the viscosity and density, and then later fluid loss and even solids and salinity in muds. That proved to be all very successful and promising.” Construction of a mud skid to house the equipment and sensors needed to conduct these tests in real time was the next step in the evolution of van Oort’s concept. That initial skid was a cannibalized and reworked version of a unit that was employed on Shell’s Rig 1, which the supermajor built for its in-house rig-automation research based in Pennsylvania. This early mud skid, considered the prototype of van Oort’s design, was abandoned before it was properly tested. “We generated quite a bit of IP [intellectual property], my students and I at UT,” he said. “The Shell skid hadn’t seen a significant amount of service, and it had some nice components that we could reuse. We took that skid apart and reconfigured it and put it out in the field with Pioneer Natural Resources for a set of field trials in the Permian. Those went well.” The field trial results were shared in a paper presented at the 2019 Unconventional Resources Technology Conference (URTEC 2019-964). The paper concluded that the pipe viscometer employed by the skid allows for the characterization of additional rheology parameters, which cannot be obtained with Couette-type viscometers, such as the critical Reynolds number, characterizing the transition from laminar into turbulent flow, and the friction factor in the turbulent flow regime.

Newly Developed Correlations to Predict the Rheological Parameters of High-Bentonite Drilling Fluid Using Neural Networks

High-bentonite mud (HBM) is a water-based drilling fluid characterized by its remarkable improvement in cutting removal and hole cleaning efficiency. Periodic monitoring of the rheological properties of HBM is mandatory for optimizing the drilling operation. The objective of this study is to develop new sets of correlations using artificial neural network (ANN) to predict the rheological parameters of HBM while drilling using the frequent measurements, every 15 to 20 min, of mud density (MD) and Marsh funnel viscosity (FV). The ANN models were developed using 200 field data points. The dataset was divided into 70:30 ratios for training and testing the ANN models respectively. The optimized ANN models showed a significant match between the predicted and the measured rheological properties with a high correlation coefficient (R) higher than 0.90 and a maximum average absolute percentage error (AAPE) of 6%. New empirical correlations were extracted from the ANN models to estimate plastic viscosity (PV), yield point (YP), and apparent viscosity (AV) directly without running the models for easier and practical application. The results obtained from AV empirical correlation outperformed the previously published correlations in terms of R and AAPE.

Real-Time Prediction of the Rheological Properties of Water-Based Drill-In Fluid Using Artificial Neural Networks

The rheological properties of drilling fluids are the key parameter for optimizing drilling operation and reducing total drilling cost by avoiding common problems such as hole cleaning, pipe sticking, loss of circulation, and well control. The conventional method of measuring the rheological properties are time-consuming and require a high effort for equipment cleaning, so they are only measured twice a day. There is a need to develop an automated system to measure the rheological properties in real-time based on the frequent measurements of mud density, Marsh funnel time, and solid percent. The main objective of this paper is to apply a modified self-adaptive differential evolution technique to determine the optimum combination of an artificial neural network’s variables to precisely predict the rheological properties of water-based drill-in fluid using the frequent measuring of mud density, Marsh funnel time, and solid percent. The second objective is whitening the black box of an artificial neural network by developing five new empirical correlations to determine the rheological properties without the need for the artificial neural network models. Actual field measurements (900 data points) were used to train, test, and validate the artificial neural network models and the developed empirical correlations. The optimization process illustrated that the best training function was Bayesian regularization backpropagation (trainbr), and the best transferring function was Elliot symmetric sigmoid (elliotsig). The optimum number of neurons was 30 for the plastic viscosity and the flow consistency index, while it was 29 for apparent viscosity, yield point, and the flow behavior index. The developed artificial neural network models and empirical correlations predicted the rheological properties with high accuracy. The correlation coefficient (R) was more than 90%, and the average absolute percentage error was less than 8.6%. The new technique for rheological properties estimation is an example of the new development which will help the new generation to discover and extract oil and gas with less cost and with safer operations.