To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Management by Exception Surveillance Data for Well Management for Maximizing Oil Production |
|
Author |
Azlinda Abdul Malik, Mohd Hilmi Hasan, Ahmad Nazeer Azhar, and Anang Hudaya Muhamad Amin |
| Citation |
Vol. 25 No. 11 pp. 190-199 |
|
Abstract |
Current economic realities have pushed oil and gas company to produce “more with less”. The ever-increasing amount of data available to surveillance engineers has caused engineers to spend more time gathering, analyzing them manually which is a daunting exercise and inefficient. The time-consuming process of analyzing surveillance data of a well with unexpected well shut-in time is able to cause a huge loss in a production operation. By leveraging on data-driven surveillance by adopting the principle of management by exception (MBE), the project tries to minimize the manual interaction between data and engineers. The project will be focusing on analyzing the surveillance data from 26 wells of 3 different fields located in North Sea and the data that will be used in the study is the production data in terms of oil production rate (bbl/day), and the time (year) for each well with cross-checking the water cut and gas oil ratio. The study will also focus on the surveillance data in the absence of reservoir characteristics. The objectives of the study are to study the trends of production data available and to design an anomaly detection model (with a certain algorithm) that is able to identify any deviations in production trend. Anomaly detection will be used to gain insight of when the well be offline if the problem keeps on persisting without immediate action from engineer. In order to meet the objectives, data preparation process will be done before developing the anomaly detection model (with a certain algorithm). The trends and analysis of the data will be done in both Microsoft Excel and Python while the anomaly detection model (with algorithm) will be developed by using Python. |
|
Keywords |
Anomaly detection; well surveillance; oil production prediction |
|
URL |
