A drill string is used to drill a hole in a hydrocarbon reservoir formation by following a well plan. The lowest part of a modern drill string for directional drilling is the bottom hole assembly (BHA), which mainly comprises a drill bit, a rotary steerable system, and one or more measurement while drilling (MWD) tools. A rotary shouldered threaded connection is made up with a pin element and a box element and is used to screw together the tools and subs used on a BHA. Rotary shouldered connections are the most fatigue critical components on the BHA when experiencing a large number of revolutions in a curved well section. Once initiated, a fatigue crack could propagate and ultimately result in a twist-off event, which could result in an expensive fishing operation and a tremendous amount of non-productive time. Efficient fatigue life prediction of rotary shouldered connections is a key enabler for real time fatigue damage tracking during drilling operations.

Rotary shouldered connections could come in various designs and sizes to suit different BHA configurations. Many factors could affect the fatigue strength of a rotary shouldered threaded connection, including the inner and outer diameters of the connection, the thread form, the use of stress relief features, the material properties, the surface conditions, and the service loads applied. Full-scale rotating bending fatigue testing is the direct way of characterizing the fatigue strength of a rotary shouldered connection, but it is expensive and is affordably limited to a few sample configurations and load conditions. Nonlinear finite element analysis (FEA) could provide cost-effective solutions. However, it takes hours to reach at a solution, which prevents its use as a computing engine for real-time fatigue damage monitoring applications.        

In this paper, a hybrid computational framework that combines the state-of-the art machine learning algorithms and nonlinear finite element analysis for efficient and accurate fatigue life prediction of rotary shouldered threaded connections is presented. Specifically, a large set of simulation data from nonlinear FEA, along with a small set of experimental data from full-scale fatigue tests, constitutes the dataset required for training and testing of a fast-loop predictive model that could cover most commonly used rotary shouldered connections. Feature engineering was first performed to explore the compressed feature space to be used to represent the data. An ensemble deep learning algorithm was then developed to learn the underlying pattern, and hyperparameter tuning techniques were employed to select the learning model that provides the best mapping, between the features and the fatigue strength of the connections. The resulting fatigue life predictions were found to agree favorably well with the experimental results from full-scale bending fatigue tests. This newly developed hybrid modeling framework paves a new way to real-time predicting the remaining useful life of rotary shouldered threaded connections for prognostic health management of the drilling equipment.  


A Hybrid Supervised Deep Learning and Nonlinear Finite Element Framework for Efficient Fatigue Life Predictions of Rotary Shouldered Threaded Connections

Division / Sponsors: 
Aerospace, Bioengineering, Applied Mechanics, Design Engineering Division, Advanced Energy Systems, Manufacturing Engineering Division, Noise Control and Acoustics, Nondestructive Evaluation, Diagnosis, and Prognosis (NDPD), Nuclear Engineering Division, Process Industries Division, Fluids Engineering, Heat Transfer, Materials, Microelectromechanical Systems, Safety Engineering and Risk Analysis, Computers & Information in Engineering


Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2020/volume/84478

IMECE2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.