The accurate prediction of transport in porous media through realistic modelling is crucial for many engineering problems, including oil and gas development, geothermal energy exploitation, and carbon dioxide sequestration. However, considerable uncertainties are associated with the simulation of the transport phenomenon in porous media, and these have a significant impact on the credibility of the predictions. These uncertainties can originate from insufficient data to characterise the system, incomplete knowledge to understand the system, and improper parameters to represent the system. Machine learning and data-driven methods can help to solve some of these problems, especially those that are difficult to construct using explicit flow models.
This special issue has been organised in association with the 14th Annual InterPore Meeting (Interpore 2022), which will take place on 30 May to 2 June 2022. The aim is to publish state-of-the-art research on the topics of uncertainty quantification and machine learning methods, and on their applications to complex transport problems in the field of petroleum science.
These include, but are not limited to:
- Stochastic modelling algorithms in porous media
- Data-driven models in porous media
- Uncertainty propagation in porous media
- Sensitivity analysis to identify the most influential uncertain variables
- Stochastic inverse modelling to estimate the uncertain properties in porous media
- Surrogate model development for uncertainty quantification in porous media
- Data assimilation to incorporate the effect of the dynamic data to constrain uncertainty
- Data worth analysis under uncertainty in porous media
- Machine learning in uncertainty quantification problems in porous media
- Dimensionality reduction to improve the efficiency of uncertainty quantification
- Optimisation design in porous media under uncertainty
- Multi-fidelity methods of uncertainty quantification in porous media
- Submission deadline: 30 October 2022
All submitted papers must be clearly written in excellent English and contain only original work that a) has not been published by any other journal, and b) is not currently under review by another journal. All papers will be peer-reviewed according to the reviewing policy of Petroleum Science. Please read the Guide for Authors before submitting. All articles should be submitted online; please select “SI: InterPore2022” as the article type.
Lead guest editor:
- Liang Xue
College of Petroleum Engineering, China University of Petroleum - Beijing, China
Interests: Uncertainty quantification in porous media, machine learning in reservoir engineering, history matching and optimisationAssociate Editor of Petroleum Science
- Xiaodong Luo
Norwegian Research Centre (NORCE), Norway
Interests: big data, optimisation, data assimilation, digitisation, reservoir modelling and simulation, machine learning, history matching and 4D seismic
- Bailian Chen
Los Alamos National Laboratory, USA
Interests: Oil & gas production forecasting and optimisation, deep learning and reduce-order modelling, data assimilation and uncertainty quantification
- Qinzhuo Liao
Department of Petroleum Engineering, King Fahd University of Petroleum and Minerals, KSA.
Interests: Uncertainty quantification, history matching, upscaling for flow and transport in porous media