1887

Abstract

Summary

Drilling safety and efficiency is always a primary focus for petroleum companies. The main objective is to minimize non-productive time during drilling while eliminating hazards for personnel and well control events. One of the severe hazards during drilling is overpressure in rock formations. This overpressure is typically caused by the geological conditions of specific formations and can be managed by optimizing mud weight during drilling. We propose achieving this with foundational seismic machine learning models. We employ autoencoder and transformer architectures for fine-tuning foundational models, enabling accurate predictions of mud weight values either for a target formation in a specific well or for the full vertical mud weight profile. Different architectures are used depending on whether seismic traces are handled individually or collectively within a local radius. The results are validated by real-time drilling data.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202539069
2025-03-24
2026-02-14

  • From This Site

    /content/papers/10.3997/2214-4609.202539069
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Alsinan, S., Nivlet, P. and Alghenaim, H. [2022] Refining our Understanding of the Subsurface Geology using Deep Learning Techniques. In: Second International Meeting for Applied Geoscience and Energy. Society of Exploration Geophysicists.
     [Google Scholar]
  2. Callerio, S., Ashok, C. and Oort, E. [2024] Real-Time Casing Point Selection Optimization for MPD Wells Through Improved Kick Tolerance Analysis. In: IADC/SPE International Drilling Conference and Exhibition, USA, 5–7 March 2024. IADC/SPE International Drilling Conference and Exhibition.
     [Google Scholar]
  3. Li, Y., Alkhalifah, T., Huang, J. and Li, Z. [2023] Self-Supervised Pretraining Vision Transformer With Masked Autoencoders for Building Subsurface Model. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 61.
     [Google Scholar]
  4. Liu, P., Li, J., Chen, B., An, X., Yan, G., Lei, Q., Liang, L., Huang, Y., Zhao, H., Wang, G. and Sun, M. [2024] Digital Wellbore Stability Prediction with Machine Learning. In: International Petroleum Technology Conference. International Petroleum Technology Conference, Saudi Arabia, 12–14 February 2024.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202539069
Loading
Deep Learning Framework to Predict Mud Weight for Well Control Prevention
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.202539069
/content/papers/10.3997/2214-4609.202539069
/content/papers/10.3997/2214-4609.202539069
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error