Drilling Optimization: Utilizing Lifetime Prediction to Improve Drilling Performance and Reduce Downtime

Abstract

Abstract The capability to optimize drilling performance by predicting the life of drilling components is integral to preventing costly downhole tool failures and ensuring success of any drilling operation. Drilling tools are subject to various parameters such as vibration, temperature, revolutions per minute (RPM) and torque. These parameters can greatly fatigue even the most robust tool depending on the where and how the tool is operated. As a result, there is a need to predict time to failure of components operating in a downhole drilling environment. Analyzing operational data, inclusive of the parameters above, prior to or during maintenance actions and before starting drilling jobs, provides unique insight into how to improve the drilling performance of tools and to reduce downtime. Life prediction provides a cutting-edge way to identify precursors to costly failures in the field and enables proactive guidance during maintenance periods for parts which may otherwise have been disregarded strictly on maintenance procedures. Statistical models that relate operating environment to the component life and are derived from failure data of fielded components, introduce a new way to optimize the efficiency of drilling tools. Utilizing lifetime prediction to optimize drilling performance is a groundbreaking methodology developed to determine life of components operating in benign and harsh drilling environments by incorporating statistical aspects such as those caused because of variation in operating stress and maintenance upgrades. Since the algorithm utilizes field data, the need for costly laboratory experiments are also eliminated. Each model developed is unique to the specified part and can be calibrated for the best fit. In this methodology, a Bayesian-based model selection technique is developed that incorporates operating environment variables after each successful drilling run to dynamically select a model that gives the best survival probability for that component. Dynamic model selection ensures maximum utilization of a component, while avoiding failure to improve the overall reliability of the tool while in the field. The paper describes the methodology used to estimate the life of components in drilling systems by employing operational data, drilling dynamics and historical information.