Affiliation:
1. Petroleum Development Oman
Abstract
Abstract
Wellbore stability problems, such as stuck pipe and tight spots, are one of the most critical risks that affect drilling operations. Over several years, Oil and Gas Operators in the Middle East have been facing problems associated with stuck pipe and tight spot events, which have a major impact on drilling efficiency, well cost, and the carbon footprint of drilling operations. On average, the operator loses around 200 days per year in terms of Non-Productive Time(NPT) on stuck events and associated fishing operations.
Wellbore stability problems are hard to predict due to the varying conditions of drilling operations: different lithologies, drilling parameters, pressures, equipment, shifting crews, and multiple well designs. All these factors make the occurrence of stuck events quite hard to mitigate when relying on human intervention only.
In Petroleum Development Oman (PDO), we lost, between 2013 and 2017, a total of 1,568 days due to stuck events. This equated to a cost of US$15.5 million/year on average, which included the cost of tools lost in hole. In 2018 PDO's Management formed an interdisciplinary taskforce with the objective of analysing stuck events to find solutions which would reduce the cost of stuck events by 50% per year.
At the same time PDO decided to develop an Artificial Intelligence (AI) driven tool that leverages the whole breadth and depth of data (reports, sensor data, well engineering data, lithology data, etc.) available to predict and prevent wellbore stability problems. The tool, known as the "Stuck Pipe & Tight-Spot Event Prediction" (STEP) tool, informs well engineers and rig crews about possible risks, both during the well planning and well execution phase, suggesting possible mitigation measures to avoid getting stuck.
The taskforce's hard and diligent work, along with the use of the STEP tool, resulted in a significant reduction in stuck events, associated time, cost, HSE exposure and production deferment.
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Comparing Drilling Anomaly Prediction by Purely Data-Driven and Hybrid Analysis Methods - Case Study of Utah FORGE Geothermal Wells;Day 3 Thu, March 07, 2024;2024-02-27
2. Enhancing Stuck Pipe Risk Detection in Exploration Wells Using Machine Learning Based Tools: A Gulf of Mexico Case Study;Day 2 Wed, March 06, 2024;2024-02-27
3. Digital Twin System Utilizing Hybrid Approach of Physics Model and Machine Learning for Pipe Stuck Early Identification and Warning Based on Real-Time Data;Day 4 Fri, March 01, 2024;2024-02-22
4. Application of Digital Twin for Real-Time Early Identification and Warning of Pipe Stuck Using Hybrid Approach of Physics Model and Machine Learning;Day 4 Thu, October 05, 2023;2023-10-02
5. Leveraging Targeted Machine Learning for Early Warning and Prevention of Stuck Pipe, Tight Holes, Pack Offs, Hole Cleaning Issues and Other Potential Drilling Hazards;Day 1 Mon, May 01, 2023;2023-04-24