Тип публикации: статья из журнала
Год издания: 2025
Идентификатор DOI: 10.46544/AMS.v30i3.02
Ключевые слова: natural gas, gas hydrate, ice plug, early detection, gas collection system, automation
Аннотация: The transition to sustainable development and the movement towards carbon neutrality pose new challenges for the production of natural gas, which is considered both a "transitional" and a "stable" fuel. In the Arctic region, the formation of hydrates and ice plugs in gas collection and treatment systems is a significant unresolved Показать полностьюproblem. Implementing measures to prevent this can have a substantial impact on gas production operating costs. The aim of this work is to develop a prototype of an automated analytics system that detects pressure anomalies early, potentially signaling the onset of hydrate formation in the gas collection system. To solve this problem, the methods of searching for anomalies and averaging data (sliding window, moving average, threshold value) were used. The article highlights the shortcomings of existing approaches, which are associated with insufficient accuracy of modeling due to the complexity and multifactorial nature of hydrate formation, increased measurement error, imperfection of devices, and communication channels of sensors with the control panel. The study subject is a gas collection system. In the process of implementing the tasks, a decision-making algorithm was developed, and a mathematical model was developed at the first level of this algorithm, which is adaptive, flexible, and easily scalable for working with other types of time series of parameters (flow rate, temperature), as well as various types of gas collection systems. A mathematical description of the developed model was presented. The possibility of scaling and developing the idea was demonstrated, along with the feasibility and possibility of its implementation. The results of experimental verification are also given.
Журнал: Acta Montanistica Slovaca
Выпуск журнала: Т. 30, № 3
Номера страниц: 571-580
ISSN журнала: 13351788
Издатель: Technicka Univerzita v Kosiciach