Abstract

Advanced simulation techniques which have been applied successfully for some decades by the aerospace and automotive industries, are now being used to design more robust drilling tools, improve data quality, optimize bottom hole assemblies, minimize drilling severity, construct better quality boreholes, simulate the effects of drilling dysfunction and prevent non-productive time. Extensive combinations of various bottom hole assembly designs, wellbore trajectories and drilling controls are now being simulated without the risk or costs associated with the physical experimentation that is traditionally required to enable design and performance improvements. Simulation has the potential to help drill more productive wellbores with less tortuosity and reduced workover costs. Simulation also provides a virtual reality with helpful visualizations of failure mechanisms that result from drilling inefficiency.

Drilling engineers often find that vibration frequencies observed using physical downhole sensors are different to frequencies predicted by finite element structural mode shape and critical speeds analyses that hypothetically permit a drill string to extend beyond its borehole diameter when in resonance and ignore non-linear damping and frequency shifts associated with wall contacts. Non-linear finite element and multi-body dynamics simulation techniques more realistically constrain the drill string to remain within the borehole and can predict actual physical amplitude and frequency responses with higher fidelity.

Another significant benefit of simulation is how virtual sensors enable the measurement of dynamic forces and motions anywhere between bit and top-drive - in particular within vulnerable drill string components where physical sensors are impractical. Whenever unpredicted dynamic forces associated with drilling dysfunction are detected by physical sensors, simulation can propagate those dynamics along the drill string and provide useful insights at critical locations where physical sensors cannot be located.

This paper presents various simulation use cases together with a field example of how downhole physical measurements combined with a calibrated simulation running in a timely manner could have anticipated and potentially prevented the fishing of a drilling motor that twisted off. The timeliness of drilling decisions and how the characteristics of surface and downhole physical measurements affect simulation calibration and the influence of physical data on simulation fidelity are also discussed.

Definitions

There is some confusion about the meaning of various terms related to drilling automation stemming from a diversity of published definitions. Sometimes the term automation itself is used to embrace only the process of parametric analysis while excluding any simulation validation and control processes. The terms digitization and digitalization are also sometimes interchanged and incorrectly used synonymously.

The value derived from drilling simulation is predicated upon the assumption that the entire coupled, non-linear and dynamic hydro-thermo-geomechanical process can be modeled and that drilling severity in response to various combinations of drilling controls can be predicted with adequate fidelity. This implies that benign as well as severe dynamic drilling behaviors should each be able to be simulated in a predictable and well-behaved manner. It should be noted, however, that when a drill string component fails to operate in a normal manner, it requires a comparison of a physical measurement to its digital twin to discern that there is a drilling dysfunction which could not be predicted alone by either the physical data or by simulation. The difference between the simulated digital twin and its physical counterpart can further be exploited with the dynamic virtual characteristics of a dysfunction propagated along the drill string and assessed for their significance at critical drill string locations even where physical sensors cannot be placed.

In order to assure a more consistent understanding from the terminology used in this manuscript, the following definitions (Google, Collins et al., Theys, 1999 & Theys, 2011) are adopted:

Accuracy: Closeness between a measurement and the true value of what is being measured 
Automation: Controlling a process while reducing human intervention to a minimum 
Calibration: Adjusting sensor settings to improve accuracy or simulation inputs to improve fidelity 
Digitalization: Integration of digital technologies into everyday life using digitization 
Digitization: Conversion into a digital form that can be processed by a computer 
Dysfunction: Fault in a machine exhibiting different behavior from what would be considered normal 
Fidelity: Measures of how well the simulation results correspond to reality 
Precision: Closeness between repeated measurements acquired in the same manner 
Resolution: Smallest detectable difference between two separate and close measurements 
Severity: Condition of being harsh, intense or damaging 
Spread: Measure of the magnitude of error due to alignment and environmental effects 
Uncertainty: Numerical composite sum of all errors in a measurement 
Validity: Relative indication of the appropriateness of the simulation for a specific purpose 
Accuracy: Closeness between a measurement and the true value of what is being measured 
Automation: Controlling a process while reducing human intervention to a minimum 
Calibration: Adjusting sensor settings to improve accuracy or simulation inputs to improve fidelity 
Digitalization: Integration of digital technologies into everyday life using digitization 
Digitization: Conversion into a digital form that can be processed by a computer 
Dysfunction: Fault in a machine exhibiting different behavior from what would be considered normal 
Fidelity: Measures of how well the simulation results correspond to reality 
Precision: Closeness between repeated measurements acquired in the same manner 
Resolution: Smallest detectable difference between two separate and close measurements 
Severity: Condition of being harsh, intense or damaging 
Spread: Measure of the magnitude of error due to alignment and environmental effects 
Uncertainty: Numerical composite sum of all errors in a measurement 
Validity: Relative indication of the appropriateness of the simulation for a specific purpose 
You can access this article if you purchase or spend a download.