☆ 4.7 Article

Expert system for early sign stuck pipe detection: Feature engineering and fuzzy logic approach

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2024)

Factors affecting drilling incidents: Prediction of suck pipe by XGBoost model

Talgat Kizayev et al.

Summary: This research utilizes XGBoost models to analyze the parameters influencing stuck pipe accidents. The results indicate that factors such as measured depth, rate of penetration, gel time, plastic viscosity, mud weight, and yield point have a positive contribution to the model predictions, while factors such as flow rate, rotation per minute, filtrate, and bottom hole assembly length have a negative contribution. The findings of this study are useful in drilling operations.

ENERGY REPORTS (2023)

添加到收藏夹

Article Energy & Fuels

Early sign detection for the stuck pipe scenarios using unsupervised deep learning

Konda Reddy Mopuri et al.

Summary: This paper presents a novel approach for early detection of drilling stuck events using Deep Learning, specifically utilizing Autoencoder to anticipate these events during the drilling process. By building Autoencoders on Recurrent Neural Networks to model normal drilling activity, the approach successfully identifies stuck incidents as anomalous activity with promising results on actual drilling data. Reconstruction analysis on individual drilling parameters is also conducted to explain the trained model's predictions.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2022)

添加到收藏夹

Article Engineering, Petroleum

Early Stuck Pipe Sign Detection with Depth-Domain 3D Convolutional Neural Network Using Actual Drilling Data

Naoki Tsuchihashi et al.

Summary: This paper studies a real-time stuck pipe prediction using the deep-learning approach, successfully detecting early signs of stuck using a 3D-CNN model with depth-domain data clip, providing important information for rational decision-making by drilling engineers.

SPE JOURNAL (2021)

添加到收藏夹

Article Energy & Fuels

Data-driven indicators for the detection and prediction of stuck-pipe events in oil&gas drilling operations

Aida Brankovic et al.

Summary: This work focuses on developing three indicators based on mudlog data to detect stuck-pipe phenomena and reduce the risk of sticking. By combining statistical models and advanced machine learning tools, a real-time updated map of stuck-pipe risk is created to help mitigate the issue.

UPSTREAM OIL AND GAS TECHNOLOGY (2021)

添加到收藏夹

Article Energy & Fuels

Drilling stuck pipe classification and mitigation in the Gulf of Suez oil fields using artificial intelligence

Haytham H. Elmousalami et al.

JOURNAL OF PETROLEUM EXPLORATION AND PRODUCTION TECHNOLOGY (2020)

添加到收藏夹

Expert system for early sign stuck pipe detection: Feature engineering and fuzzy logic approach

相关参考文献

Factors affecting drilling incidents: Prediction of suck pipe by XGBoost model

Early sign detection for the stuck pipe scenarios using unsupervised deep learning

Early Stuck Pipe Sign Detection with Depth-Domain 3D Convolutional Neural Network Using Actual Drilling Data

Data-driven indicators for the detection and prediction of stuck-pipe events in oil&gas drilling operations

Drilling stuck pipe classification and mitigation in the Gulf of Suez oil fields using artificial intelligence

导出引文

分享论文