Real time prediction of drilling fluid rheological properties using Artificial Neural Networks visible mathematical model (white box)

The main objective of this paper is to use the frequent measurements of mud density, Marsh funnel viscosity, and solid percent to predict the rheological properties (plastic viscosity, apparent viscosity, yield point, flow behavior index, and consistency index of the drilling fluid) by developing empirical correlations based on 3000 field data measurements of KCl–polymer mud using artificial neural network. In this paper and for the first time, the solid percent will be included in the prediction of the drilling fluid rheological parameters. The common procedure on the well site is to perform a rheological test twice a day and measure mud density, Marsh funnel viscosity, and solid percent frequently every 15–20 min. The artificial neural network (ANN) black box was converted to white box to extract the mathematical model that can predict the rheological parameters. The average absolute error for all correlations was less than 6%, and the correlation coefficient was greater than 90%. Using the developed correlations in predicting the drilling fluid properties with high accuracy will eliminate the need for tedious laboratory measurements, and real-time properties can be obtained. This technique will help the drilling engineers better monitor the drilling fluid properties and control the drilling operations by avoiding the common problems, such as pipe sticking, loss of circulation, and hole cleaning issues.

Real-Time Prediction of Rheological Parameters of KCl Water-Based Drilling Fluid Using Artificial Neural Networks

Risk assessment on deepwater drilling well control based on dynamic Bayesian network


 Data-driven methods serve as robust tools to turn data into knowledge. Historical data generally has not been used in an effective way in analyzing processes due to lack of a well-organized data, where there is a huge potential of turning terabytes of data into knowledge. With the advances and implementation of data-driven methods data-driven models have become more widely-used in analysis, predictive modeling, control and optimization of several processes. Yet, the industry overall is still skeptical on the use of data-driven methods, since they are data-based solutions rather than traditional physics-based approaches; even though physics and geology are often part of this methodology. This study comprehensively evaluates the status of data-driven methods in oil and gas industry along with the recent advances and applications.
 This study outlines the development of these methods from the fundamentals, theory and applications perspective along with their historical acceptance and use in the industry. Major challenges in the process of implementation of these methods are reviewed for different areas of applications. The major advantages and drawbacks of data-driven methods over the traditional methods are discussed. Limitations and areas of opportunities are outlined. Recent advancements along with the latest applications, the associated results and value of the methods are provided.
 It is observed that the successful use of data-driven methods requires strong understanding of petroleum engineering processes and the physics-based conventional methods together with a good grasp of traditional statistics, data mining, artificial intelligence and machine learning. Data-driven methods start with a data-based approach to identify the issues and their solutions. Even though data-driven methods provide great solutions on some challenging and complex processes, that are new and/or hard to define with existing conventional methods, there is still skepticism in the industry on the use of these methods. This is strongly tied to the delicacy and sensitive nature of the processes and on the usage of the data. Organization and refinement of the data turn out to be important components of an efficient data-driven process.
 Data-driven methods offer great advantages in the industry over that of conventional methods under certain conditions. However, the image of these methods for most of the industry professionals is still fuzzy. This study serves to bridge the gap between successful implementation and more widely use and acceptance of data-driven methods, and the fuzziness and reservations on the understanding of these methods in the industry. Significant components of these methods along with clarification of definitions, theory, application and concerns are also outlined in this study.

Status of Data-Driven Methods and their Applications in Oil and Gas Industry

Dynamic stress distribution around the wellbore influenced by surge/swab pressure

Time series data analysis for automatic flow influx detection during drilling

This paper looks into the effects of solids on the wellbore pressure profile under different conditions An extensive number of experiments were conducted on a 90-ft-long, 45 in 8 in full-scale flow loop to simulate field conditions The flow configurations are analyzed A solid–liquid two-phase flow configuration map is proposed Significant difference is found between the pressure profile with solids and without solids in the wellbore The results of this study show how the pressure profile in the wellbore varies when solids present in the annulus, which may have important applications in drilling operations [DOI: 101115/14030845]

Pressure Profile in Annulus: Solids Play a Significant Role

A unified transient solid-liquid two-phase flow model for cuttings transport- modelling part

Is Well Clean Enough? A Fast Approach to Estimate Hole Cleaning for Directional Drilling

\n This paper presents an approach that combines a two-dimensional (2D) computational fluid dynamics (CFD) and one-dimensional (1D) continuous model for cuttings transport simulation during drilling of oil and gas wells. The 2D CFD simulates the flow profile and the suspended cuttings concentration profile in the cross section of the wellbore and the 1D continuous model simulates the cuttings transportation in the axial direction of the wellbore. Different cuttings sizes are considered in the model by using a new proposed superposition method. Experimental tests conducted on a 203 × 114 × 25 mm3 flow loop are used to validate the model from three different perspectives: the single-phase flow pressure drop, the steady-state cuttings bed height, and the transient pressure changes. Compared to layer models, the new approach is able to catch accurate flow details in the narrow flow region and overcome the shortcoming of traditional models that underpredict bed height under high flow rate conditions. The computational time increases by the order of 104∼105 from the level of millisecond to seconds but is still within the acceptable range for engineering applications, and the model provides close to three-dimensional (3D) accuracy at a much shorter central processing unit (CPU) time compared to 3D CFD models.

Feifei Zhang

Papers

Time series data analysis for automatic flow influx detection during drilling

Pressure Profile in Annulus: Solids Play a Significant Role

A unified transient solid-liquid two-phase flow model for cuttings transport- modelling part

Is Well Clean Enough? A Fast Approach to Estimate Hole Cleaning for Directional Drilling

Modeling of Dynamic Cuttings Transportation during Drilling of Oil and Gas Wells by Combining 2D CFD and 1D Discretization Approach