Correlation Between Drilling Parameters and Lithology - The Hidden Geological Information of Drilling Data

Lamik-Thonhauser, Bouchra; Schoen, Juergen H.; Koller, Christian S.; Arnaout, Arghad

doi:10.2118/192916-ms

Cited by 2 publications

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Predicting sonic and density logs from drilling parameters using temporal convolutional networks

et al. 2022

View full text Add to dashboard Cite

Sonic and bulk density logs are crucial inputs for many subsurface tasks including formation identification, completion design, and porosity estimation. Economic and operational concerns restrict the acquisition of these logs, meaning the overburden and sometimes entire wells are completely unlogged. In contrast, parameters that monitor drilling operations, such as weight on bit and torque, are recorded for every borehole. Previous studies have applied supervised machine learning approaches to predict these missing logs from the drilling parameters. While the results are promising, they often do not investigate the importance of different features and the corresponding practical implications. Here, we explored the feasibility of predicting compressional slowness and bulk density logs using various combinations of formation markers, gamma-ray logs, and drilling data recorded at the rig. Our tests utilized a temporal convolutional network to allow the model to learn from sequences of input features. Bayesian-based hyperparameter tuning found the optimum set of parameters for each experiment before producing the final log predictions. Finally, a permutation feature importance analysis revealed which input variables contributed most to the outputs. Although drilling parameters contain some insight into the mechanical rock properties, we found that they cannot produce the high-quality log predictions required for many tasks. Supplementing the drilling parameters with a gamma-ray log and formation data produces good-quality log predictions, with the additional inputs helping to constrain the model outputs.

show abstract

Predicting sonic and density logs from drilling parameters using temporal convolutional networks

et al. 2022

View full text Add to dashboard Cite

show abstract

Correlation Between Drilling Parameters and Geomechanical Properties - The Hidden Geomechanical Information

Lamik-Thonhauser

Schoen²,

Koller³

et al. 2019

Day 3 Wed, November 13, 2019

View full text Add to dashboard Cite

The drilling process is fundamentally controlled by the interaction of the bit with the rock which is characterized by its mechanical properties. The focus of various studies is directed mainly on the optimal design of drill bits and drilling operations related to the (expected) geological situation for a safe and efficient drilling process. The question arises: "Is it possible to extract any rock properties information from drilling data?". In a previous paper the correlation of drilling properties with the lithology of penetrated formation was evaluated. A clear correlation could be demonstrated by applying different statistical methods. Based on these results a simplified model for the rock destruction at the bit is developed and a "formation strength parameter S" is defined. As a cut-force parameter in the equation, weight on bit (WOB) is used primarily for vertical sections. For deviated or horizontal sections the differential pressure Delta-p is a more relevant parameter. A new parameter S is calculated as a function of drilling parameters only (RPM, WOB, Delta-p, Bit-size). This new parameter S is compared with existing parameters for rock characterization (Mechanical Specific Energy) and discussed herein. For four wells this parameter S is plotted as function of the depth representing a first geomechanical model. The main results are: Well A: Geomechanical characterization of different clastic and carbonate sections of a vertical well. Clastic (sandstone, shale) and carbonate (limestone, dolomite) are clearly separated. The carbonate section is detailed and subdivided in dense-hard and fissured-porous parts. Well B: Calculations are made for vertical (based on WOB) and deviated/horizontal section (based on Delta-p) and compared; for the deviated/horizontal section the Delta-p version gives more reliable results. Well C: For a specific section of this well after drilling a profile of Unconfined Compression Strength (UCS) with a standard well-logging algorithm was calculated. A comparison of the "during-drilling" derived geomechanical model in terms of S and the "post-drilling" derived in terms of UCS demonstrates similar tendencies and confirms the character of S as a geomechanical measure. Well D: Data (WOB, ROP, RPM) for this example are logged as part of a test of the new PDS Digital Drilling technology (Powerline Drillstring Technology) in an igneous formation (phyllite). It demonstrates the ability to discriminate different geomechanical sections of one formation.

show abstract

Correlation Between Drilling Parameters and Lithology - The Hidden Geological Information of Drilling Data

Cited by 2 publications

References 8 publications

Predicting sonic and density logs from drilling parameters using temporal convolutional networks

Predicting sonic and density logs from drilling parameters using temporal convolutional networks

Correlation Between Drilling Parameters and Geomechanical Properties - The Hidden Geomechanical Information

Contact Info

Product

Resources

About