StackGBM: Stacked Gradient Boost Machine for Accurate Lost Circulation Prediction

DOI

10.2991/978-94-6463-266-8_25How to use a DOI?

Keywords

Gradient boost machine; Lost circulation prediction; Machine learning; Energy security

Abstract

Lost circulation leads to severe downhole accidents in some cases, and is common in oil or gas drilling. Lost circulation has become a serious threat to energy security and environmental protection and has thus attracted widespread attention. Recently, several studies introduce machine learning algorithms into lost circulation prediction, among which the Gradient Boosted Decision Tree (GBDT) methods take the lead. However, utilizing one single GBDT method can hardly generate optimal results. In this paper, we tackle this issue by the stacking technique. Besides, a lost circulation dataset is collected for further experiments. The proposed Stacked Gradient Boost Machine (StackGBM) adopts the two-stage paradigm to further enhance the original results that are produced by XGBoost, LightGBM and Catboost. In the second stage, a neural network system is employed due to its great prediction capability. Comprehensive experiments show that StackGBM achieves state-of-the-art performance in lost circulation prediction. In addition, we perform ablation studies on the variation of StackGBM architecture. The proposed StackGBM algorithm will benefit the development of drilling engineering in the long-term.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Li Liang
AU  - Deng Hongmei
AU  - Yang Zhuo
AU  - Su Jianhua
AU  - Jiao Yang
AU  - Xie Yaorong
AU  - Wu Chengyou
PY  - 2023
DA  - 2023/10/10
TI  - StackGBM: Stacked Gradient Boost Machine for Accurate Lost Circulation Prediction
BT  - Proceedings of the 2nd International Conference on Intelligent Design and Innovative Technology (ICIDIT 2023)
PB  - Atlantis Press
SP  - 225
EP  - 233
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-266-8_25
DO  - 10.2991/978-94-6463-266-8_25
ID  - Liang2023
ER  -