Wellbore Schematics to Structured Data Using Artificial Intelligence Tools
- Vanessa Ndonhong Kemajou (Halliburton) | Anqi Bao (Halliburton) | Olivier Germain (Halliburton)
- Document ID
- Offshore Technology Conference
- Offshore Technology Conference, 6-9 May, Houston, Texas
- Publication Date
- Document Type
- Conference Paper
- 2019. Offshore Technology Conference
- 1.1 Well Planning, 7.6.6 Artificial Intelligence
- well archives, computer vision, digitization, data science, well schematics
- 293 in the last 30 days
- 301 since 2007
- Show more detail
- View rights & permissions
Wellbore schematics are essential to well planning and operations because they detail well design, completions, and sometimes the production mechanism. There are multiple formats and types of wellbore schematics; however, they generally consist of a well diagram accompanied by tables of annotations listing components and equipment details such as depths and diameters. Paper-based wellbore schematic reports are often distributed as the primary account of technical information concerning old wells after being acquired by oil and gas operators. Any intervention or further operation on those wells would require a thorough and manual interpretation of those reports, which can be lengthy and prone to errors. Therefore, to automatically convert the diagram and annotations into a readable database, a practical technique or tool has to be developed.
Artificial intelligence (AI)-powered image analysis addresses similar problems for other engineering disciplines and industries, and with the latest advances for software and computer hardware capabilities, it is possible to design specialized solutions for the oil and gas industry. Therefore, a methodology was defined and implemented to import the available machine learning technology for automating the interpretation and analysis of wellbore schematics. With this novel tool, scanning the paper-based wellbore schematic results in digital and easily shareable structured data that can be used to regenerate a digital wellbore schematic. This method analyzes the diagram and the annotations on the wellbore schematic file and then combines the analysis results by matching the diagram with the surrounding annotations and engineering constraints.
The methodology was tested on a set of wellbore schematic files, and digital schematics were regenerated. Fundamental components and equipment were detected that matched the original schematics in terms of depths and diameters. The designed tool saves considerable time and effort while providing accuracy and repeatability. These results highlight some of the benefits of applying multidisciplinary ideas for data management to the industry.
The object detection technique in image analytics is new to the oil and gas industry for identifying components in well schematics. Further, this project is comprehensive because it identifies the diagram and related annotations. Challenges and breakthroughs experienced in this research will be addressed.
|File Size||1 MB||Number of Pages||18|
Bao, A. and Gildin, E. 2017. Data-Driven Model Reduction Based on Sparsity-Promoting Methods for Multiphase Flow in Porous Media. Presented at the SPE Latin America and Caribbean Petroleum Engineering Conference, Buenos Aires, Argentina, 17–19 May. SPE-185514-MS. https://doi.org/10.2118/185514-MS.
Bao, A., Gildin, E., and Zalavadia, H. 2018. Development of Proxy Models for Reservoir Simulation by Sparsity Promoting Methods and Machine Learning Techniques. Presented at the ECMOR XVI-16th European Conference on the Mathematics of Oil Recovery. https://doi.org/10.3997/2214-4609.201802180.
Chen, B., Harp, D. R., Lin, Y.. 2018. Geologic CO2 sequestration monitoring design: A machine learning and uncertainty quantification-based approach. Applied Energy, 225: 332–345. https://doi.org/10.1016/j.apenergy.2018.05.044.
Danvk.org. 2015. Extracting text from an image using Ocropus. https://www.danvk.org/2015/01/09/extracting-text-from-an-image-using-ocropus.html (accessed on 06 July 2018).
Encana Natural Gas. 2011. Life of the Well 2011. https://www.encana.com/pdf/communities/usa/LifeOfTheWell2011.pdf (accessed 20 June 2018).
Engineer's Data Model (EDM). 2018. http://www.landmark.solutions/Engineers-Data-Model (accessed 06 July 2018).
Engineer's Desktop™ (EDT™) Software. 2018. https://www.landmark.solutions/Portals/0/LMSDocs/Datasheets/EDT-datasheet.pdf (accessed 12 June 2018).
Girshick, R. 2015. Fast R-CNN. Presented at the 2015 IEEE International Conference on Computer Vision, Santiago, Chile, 7–13 December. https://doi.org/10.1109/ICCV.2015.169.
Girshick, R., Donahue, J., Darrell, T.. 2014. Rich feature hierarchies for accurate object detection and semantic segmentation. Presented at the IEEE Conference on Computer Vision and Pattern Recognition. https://doi.org/10.1109/CVPR.2014.81.
He, K., Zhang, X., Ren, S.. 2016. Deep Residual Learning for Image Recognition. Presented at the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, Nevada, 27–30 June. https://doi.org/10.1109/CVPR.2016.90.
Huang, J., Rathod, V., Sun, C.. 2017. Speed/accuracy trade-offs for modern convolutional object detectors. Presented at the 2017 IEEE Conference on Computer Vision and Pattern Recognition. https://doi.org/10.1109/CVPR.2017.351.
Lin, T. Y., Goyal, P., Girshick, R.. 2017. Focal loss for dense object detection. Presented at the 2017 IEEE International Conference on Computer Vision. https://doi.org/10.1109/ICCV.2017.324.
Nwachukwu, A., Jeong, H., Pyrcz, M.. 2018. Fast evaluation of well placements in heterogeneous reservoir models using machine learning. Journal of Petroleum Science and Engineering, 163: 463–475. https://doi.org/10.1016/j.petrol.2018.01.019.
Odi, U. and Nguyen, T. 2018. Geological Facies Prediction Using Computed Tomography in a Machine Learning and Deep Learning Environment. In Unconventional Resources Technology Conference, Houston, Texas, 23–25 July 2018, 336–346. URTEC-2901881-MS. https://doi.org/10.15530/URTEC-2018-2901881.
Qian, F., Yin, M., Liu, X. Y.. 2018. Unsupervised seismic facies analysis via deep convolutional autoencoders. Geophysics, 83(3): A39–A43. https://doi.org/10.1190/geo2017-0524.1.
Redmon, J., Divvala, S., Girshick, R.. 2016. You Only Look Once: Unified, Real-Time Object Detection. Presented at the IEEE Conference on Computer Vision and Pattern Recognition. https//:doi.org/10.1109/CVPR.2016.91.
Ren, S., He, K., Girshick, R.. 2015. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. Advances in Pattern Analysis and Machine Intelligence, 39(6): 91–99. https://doi.org/10.1109/TPAMI.2016.2577031.
Zalavadia, H. and Gildin, E. 2018. Parametric Model Order Reduction For Adaptive Basis Selection Using Machine Learning Techniques During Well Location Opt. Presented at the ECMOR XVI – 16th European Conference on the Mathematics of Oil Recovery. https://doi.org/10.3997/2214-4609.201802235.