Stuck-Pipe Prediction by Use of Automated Real-Time Modeling and Data Analysis

Abstract

Summary A real-time method is presented to predict impending stuck pipe with sufficient warning to prevent it. The new method uses automated analysis of real-time modeling coupled with real-time data analysis. It can be applied to all well types for any well operation, including drilling, casing running, completion activities, and re-entries. The method uses leading indicators of stuck pipe that were identified by use of historical data sets of 36 stuck-pipe incidents in the Eagle Ford, Utica, and Permian and in the Gulf of Mexico. Two case histories show the utility of the new method in shale and carbonate horizontal wells for both drilling and off-bottom operations. The new method combines two types of analysis: use of hydraulics and torque-and-drag (T&D) software to determine deviation of real-time data from the real-time model, and trend analysis (i.e., rate of change) of real-time data. Parameters used are pump pressure, flow rate, torque, rotary speed, hookload and drag, and weight on bit (WOB), along with static inputs such as bottomhole-assembly (BHA) and drillstring configuration and directional surveys. Additional parameters, such as downhole equivalent circulating density (ECD), are used when available and improve the results. But the method is designed to monitor all well types and provide a stuck-pipe-risk log even by use of only basic instrumentation. A novel algorithm predicts the probability of stuck pipe, which is presented in a real-time log. Results demonstrate that there is no single leading indicator in all stuck-pipe incidents. Our early-detection method, called the stuck-pipe-risk (SPR) log, relies on multiple indicators to strengthen the likelihood of detecting impending stuck pipe while avoiding false alerts. A key element to automating the process is the use of filtering for rig activity. The first indicator is deviation of actual data from model predictions. A second indicator is trend analysis (specifically, rate-of-change calculations), which provides valuable insight into rapidly deteriorating wellbore conditions when deviation from model predictions does not respond quickly enough over a short depth or time interval. Results are presented that show the SPR-detection method successfully detected impending stuck pipe on four historical shale wells an average 38 minutes before sticking and on one historical carbonate well more than 2 hours before the event. No false alerts were recorded in these wells. These results were viewed as meeting the initial goal of providing relevant alerts with sufficient time to prevent the pipe from becoming stuck.

Digital transformation: a review on artificial intelligence techniques in drilling and production applications

“…Salminen et al [ 46 ] also study the prediction pipe sticking in real time. They use numerical and statistical analysis and basically compute the changes, deviations, and rates of change between the data previously observed and the data being received.…”

“…Results show that AI techniques can identify and distinguish mechanical from differential stuck pipe [ 36 , 42 , 43 , 46 ].…”

“…More recent studies show that oil companies are working to adopt a prevention problem culture in drilling by monitoring operations in real time, as shown by Gandelman et al [ 44 ], Salminen et al [ 46 ], and Marques [ 47 ].…”

“…61 sticking in 32 wells Moazzeni et al [ 35 ] Prediction of stuck pipes and loss of circulation Pore pressure, hydrostatic pressure, mud weight, mud fluid flow Neural networks Neural network–based tools were used for prediction of stuck pipe and loss of circulation in 32 wells drilled in the Iranian onshore field of Marun Nybøet et al [ 34 ] Alarm problems in kick occurrences Pore pressure, hydrostatic pressure, mud weight, mud fluid flow Real-time data acquisition, historical data analysis Many variables generate complexity and reduce kick alarm system performance. Authors use AI techniques to obtain better alarm prediction efficiency Salminen et al [ 46 ] Pipe sticking prediction in real time Drilling mud properties and drilling parameters Real-time data analysis, historical data analysis, numerical and statistical analysis Comparison between historical data and data being received allows real-time pipe sticking prediction; no false alert has occurred Veeningen et al [ 39 ] High inclination wells with problems of instability, cleanliness and, drilling string vibration Depth of sensors and well, pressure at different points in the column, pressure in the annular and logging Real-time data acquisition from sensors in the well, digital transmission Installation of acquisition equipment and transmission of data help prevention of problems …”

A review of technological advances and open challenges for oil and gas drilling systems engineering

“…Their model is able to capture the dominant characteristics of the cuttings transport process while incorporating process disturbances and uncertainties in real time field measurements. Salminen et al 217 developed a real-time method for predicting impending stuck pipe with sufficient warning for its prevention. Their method implements a hydraulic model, real-time and historical data and some data analytics techniques for predicting the risk of a stuck pipe.…”