Research project to use machine learning to detect underground leaks
MIDDLESBROUGH, England, United Kingdom – Researchers at the University of Teesside are working with international partners to implement innovative machine learning techniques to detect leaks during underground carbon dioxide sequestration.
The international research partnership could help reduce the environmental and economic impact of gas transmission pipeline leaks and underground carbon dioxide sequestration using artificial intelligence and machine learning.
Academics from Teesside University’s School of Computing, Engineering and Digital Technologies work with their counterparts from Texas A&M University in Qatar (TAMUQ), Qatar University, Texas A&M (USA), Birch Scientific (USA) and Rock-Oil Consulting of Canada to postulate state-of-the-art machine learning techniques to detect leaks during underground carbon dioxide sequestration in pipelines and well columns.
The project, led by Dr. Aziz Rahman, Associate Professor at Texas A&M University in Qatar, supported by other distinguished international partners including Teesside University, led by Dr. Sina Rezaei Gomari, has been awarded $530,000 (around £430,000) from Qatar. Foundation priority research.
Along with machine learning approach. The research team will also use a new digital twin for leak detection during single-phase (crude oil or gas) and multi-phase (multiple materials) flow during transport and injection of carbon dioxide into the site of underground storage. It involves creating a virtual representation of a gas pipeline that is updated in real time via a network of sensors mounted and installed in real gas pipelines.
Through the use of computational fluid dynamics, in which artificial intelligence simulates the flow of liquids and gases, the team hopes to be able to accurately predict the likelihood and location of leaks in flows. monophasic and multiphasic.
It is hoped that these techniques will make it possible to more accurately predict the location, size, number and direction of smaller and larger chronic leaks and, ultimately, to take preventative action by artificial intelligence without require human intervention.