Roar Nybø

Statistical Method for Detection of Poor Hole Cleaning and Stuck Pipe

The drilling of an oil or gas well is an expensive undertaking. Hence, it is not surprising that mistakes and accidents during drilling incur a high cost. Accidents could result in the loss of expensive equipment and subsequent delays setting back the operation for days or weeks and thus running up large bills on rig-time and personnel hours. Some types of accidents also pose a risk to the personnel or the environment. In this dissertation we study alarm systems which could give the driller an early warning of upcoming problems, and thus provide time to avoid these accidents. We explore alarm systems which combine advanced physical models of the well and drilling process with artificial intelligence and time series analysis. Finally, we determine the advantages as well as the challenges of this approach. It is our hope that this dissertation is accessible to both practitioners in machine learning and control engineering, as well as to petroleum engineers with a passing familiarity with machine learning. Hence this dissertation starts with a quick introduction to drilling problems and some terms from time series analysis and machine learning. We then briefly describe the theory of observer-based fault detection and isolation. Theories of supervisory control systems are also introduced, as these concern both the choice of algorithms and how AI-based alarm systems integrate with the rest of the operation. From chapter 6 and onward, the challenges to fault detection in drilling are discussed. We focus on clarifying what restrictions the available training data put on our choice of machine learning methods. In chapter 8 and 9, we propose ways to combine machine learning and observer-based fault detection. Experimental results are presented in chapter 10, before we end with concluding remarks in chapter 11. Our main conclusion, reflected in our experimental results, is that physical models and artificial intelligence can be combined to produce hybrid alarm systems that are better than what we could have achieved using these approaches separately. When using artificial intelligence we treat fault detection in drilling as a machine learning problem. In the course of our work we find that this problem domain differs in important respects from textbook examples of machine learning problems. Determining the distinctive characteristics of this problem domain is crucial in designing the alarm system. Drawing on examples from different fields we determine these characteristics and propose novel alarm system architectures that build on recent developments in machine learning.

Spotting A False Alarm. Integrating Experience And Real-Time Analysis With Artificial Intelligence

The drilling of an oil or gas well is an expensive undertaking. Hence, it is not surprising that mistakes and accidents during drilling incur a high cost. Accidents could result in the loss of expensive equipment and subsequent delays setting back the operation for days or weeks and thus running up large bills on rig-time and personnel hours. Some types of accidents also pose a risk to the personnel or the environment. In this dissertation we study alarm systems which could give the driller an early warning of upcoming problems, and thus provide time to avoid these accidents. We explore alarm systems which combine advanced physical models of the well and drilling process with artificial intelligence and time series analysis. Finally, we determine the advantages as well as the challenges of this approach. It is our hope that this dissertation is accessible to both practitioners in machine learning and control engineering, as well as to petroleum engineers with a passing familiarity with machine learning. Hence this dissertation starts with a quick introduction to drilling problems and some terms from time series analysis and machine learning. We then briefly describe the theory of observer-based fault detection and isolation. Theories of supervisory control systems are also introduced, as these concern both the choice of algorithms and how AI-based alarm systems integrate with the rest of the operation. From chapter 6 and onward, the challenges to fault detection in drilling are discussed. We focus on clarifying what restrictions the available training data put on our choice of machine learning methods. In chapter 8 and 9, we propose ways to combine machine learning and observer-based fault detection. Experimental results are presented in chapter 10, before we end with concluding remarks in chapter 11. Our main conclusion, reflected in our experimental results, is that physical models and artificial intelligence can be combined to produce hybrid alarm systems that are better than what we could have achieved using these approaches separately. When using artificial intelligence we treat fault detection in drilling as a machine learning problem. In the course of our work we find that this problem domain differs in important respects from textbook examples of machine learning problems. Determining the distinctive characteristics of this problem domain is crucial in designing the alarm system. Drawing on examples from different fields we determine these characteristics and propose novel alarm system architectures that build on recent developments in machine learning.

Fault detection and other time series opportunities in the petroleum industry

Data-centric methods like soft computing and machine learning have gained greater interest and acceptance in the oil and gas industry in recent years. We give an overview of the opportunities and challenges facing applied time series prediction in this domain, with a focus on fault prediction. In particular, we argue that the physical processes and hierarchies of information flow in the industry strongly determine the choice of soft computing or machine learning methods.

Regularized Nonlinear Moving-Horizon Observer With Robustness to Delayed and Lost Data

Moving-horizon estimation provides a general method for state estimation with strong theoretical convergence properties under the critical assumption that global solutions are found to the associated nonlinear programming problem at each sampling instant. A particular benefit of the approach is the use of a moving window of data that is used to update the estimate at each sampling instant. This provides robustness to temporary data deficiencies such as lack of excitation and measurement noise, and the inherent robustness can be further enhanced by introducing regularization mechanisms. In this paper, we study moving-horizon estimation in cases when output measurements are lost or delayed, which is a common situation when digitally coded data are received over low-quality communication channels or random access networks. Modifications to a basic moving-horizon state estimation algorithm and conditions for exponential convergence of the estimation errors are given, and the method is illustrated by using a simulation example and experimental data from an offshore oil drilling operation.

A Moving Horizon Observer for Estimation of Bottomhole Pressure During Drilling

Abstract To ensure safe and stable drilling operation, bottom hole pressure (BHP) should be kept within some region. However measurement of the BHP is sometimes not available or reliable, especially when the circulation is low, e.g., during pipe connection procedures. This paper presents the application of a moving horizon estimation (MHE) method for online estimation of the BHP during petroleum drilling. In the proposed MHE formulation the states are estimated by a forward simulation with a pre-estimating observer. Moreover, it considers the constraints of states/outputs in the MHE problem. Application of the observer to a real data set from a North Sea oil well illustrates potential benefits.

Real-time optimization of rate of penetration during drilling operation

The increase of drilling safety and the reduction of drilling operation costs, especially the improvement of drilling efficiency, are two important considerations. In general the rate of penetration (ROP) optimization means that the drilling parameters such as weight on bit (WOB) and rotary speed (RPM) are adjusted to drill the present formation most efficiently. In this paper, the Bourgoyne and Young ROP model had been selected to study the effects of several parameters during drilling operation. We present an advanced method for the ROP calculation and its optimization. A moving-horizon multiple regression method is proposed, which reduces the estimation error of the existing ROP models by continuously calibrating the model coefficients based on real-time data. Furthermore, a model predictive control (MPC) strategy is applied to achieve the ROP optimization to satisfy drilling requirements. The performance of the methodology is demonstrated by using realworld data from a North Sea well.

Improving Management and Control of Drilling Operations with Artificial Intelligence

In oil and gas industries, drilling is a complex and critical operation which require constant and accurate real-time monitoring. To this aim, real-time models are required to provide an overview of the drilling operations when direct and reliable measurements are not available. Given the harsh operating environment, sensor reliability and calibration are critical issues and bad data quality is a typical problem which affects the accuracy of the model. As a result, the driller may be misled about the down-hole situation or receive conflicting claims about operating conditions. This paper presents two approaches based on the use of artificial intelligence to improve monitoring of drilling processes in terms of reduced uncertainty and increased confidence. The first exploits the aggregation of the opinion of different experts within a so-called ensemble approach; the second is based on a so-called grey-box approach which combines a physical model and artificial intelligence. The two approaches are applied to the problem of predicting the bottom-hole pressure during a managed pressure drilling operation to demonstrate the improved accuracy and robustness.

Improved And Robust Drilling Simulators Using Past Real-Time Measurements And Artificial Intelligence

The drilling of an oil or gas well is an expensive undertaking. Hence, it is not surprising that mistakes and accidents during drilling incur a high cost. Accidents could result in the loss of expensive equipment and subsequent delays setting back the operation for days or weeks and thus running up large bills on rig-time and personnel hours. Some types of accidents also pose a risk to the personnel or the environment. In this dissertation we study alarm systems which could give the driller an early warning of upcoming problems, and thus provide time to avoid these accidents. We explore alarm systems which combine advanced physical models of the well and drilling process with artificial intelligence and time series analysis. Finally, we determine the advantages as well as the challenges of this approach. It is our hope that this dissertation is accessible to both practitioners in machine learning and control engineering, as well as to petroleum engineers with a passing familiarity with machine learning. Hence this dissertation starts with a quick introduction to drilling problems and some terms from time series analysis and machine learning. We then briefly describe the theory of observer-based fault detection and isolation. Theories of supervisory control systems are also introduced, as these concern both the choice of algorithms and how AI-based alarm systems integrate with the rest of the operation. From chapter 6 and onward, the challenges to fault detection in drilling are discussed. We focus on clarifying what restrictions the available training data put on our choice of machine learning methods. In chapter 8 and 9, we propose ways to combine machine learning and observer-based fault detection. Experimental results are presented in chapter 10, before we end with concluding remarks in chapter 11. Our main conclusion, reflected in our experimental results, is that physical models and artificial intelligence can be combined to produce hybrid alarm systems that are better than what we could have achieved using these approaches separately. When using artificial intelligence we treat fault detection in drilling as a machine learning problem. In the course of our work we find that this problem domain differs in important respects from textbook examples of machine learning problems. Determining the distinctive characteristics of this problem domain is crucial in designing the alarm system. Drawing on examples from different fields we determine these characteristics and propose novel alarm system architectures that build on recent developments in machine learning.