Mitigating Drilling Dysfunction Using a Drilling Advisory System: Results from Recent Field Applications

Abstract

The ExxonMobil Drilling Advisory System (DAS™) is a rig-based drilling surveillance and optimization platform that encourages regular (guided) drill-off tests, carefully monitors drilling performance and provides recommendations for controllable drilling parameters to help improve the overall drilling process. It is well known that drilling with dysfunction(s), such as BHA whirl or bit stick-slip, among others, can damage drilling tools, lead to poor borehole quality and result in significant reductions in Rate of Penetration (ROP). The ultimate objective of DAS is to help drillers achieve consistently better and longer bit runs by identifying and mitigating drilling dysfunctions.In this paper we describe the DAS workflow, provide an overview of many of the key technical components of DAS and summarize results from several recent field applications. The DAS workflow includes a learning (or calibration) mode that encourages the driller to explore different operating parameters by conducting drill-off tests, which in turn allows the system to build an understanding of performance trends. This understanding is then exploited by an application mode which leverages results obtained during the learning (or calibration) phase. The workflow is iteratively applied in real-time as drilling operations and geological conditions change.Key technical components of the DAS application include: high quality adaptive filtering algorithms for computing various drilling system performance measures (e.g., MSE, ROP, and stick-slip severity, among others), data encapsulation/dimensionality reduction techniques (i.e., the notion of Љresponse pointsЉ) and reduced order modeling techniques (simplices, smooth response surfaces, etc.) which are ultimately used to generate recommendations for improved performance.Using examples from recent offshore field applications, we illustrate the use of adaptive filtering and data encapsulation to process noisy real-time data and extract meaningful trends to enable drilling performance optimization. We show how fitting response surfaces to response points can help elucidate, in real-time, the occurrence of classical drilling dysfunctions such as lateral vibrations (as manifest in increased MSE) and stick-slip torsional vibrations. Using results from recent applications of DAS to unconventional pad-drilling operations, we show the impacts of focused drilling performance limiter redesign and real-time DAS parameter changes on improved overall ROP in subsequent wells on a pad.

Real-time rate of penetration prediction for motorized bottom hole assembly using machine learning methods

“…The concept of Mechanical Specific Energy (MSE) was firstly presented by Teale in 1965 who defined it as the amount of work required to remove a unit volume of the rock 13 . Different drilling bits performance can be evaluated with the help of applying MSE concept that leads to a significant improvement in the drilling efficiency by enhancing the ROP 19 , 20 . Moreover, MSE has different useful applications such as diagnose different drilling problems and take appropriate steps to correct them through a continuous monitoring of MSE 20 , 21 .…”