Invited review - Molecular dynamics simulations in reservoir analysis of offshore petroleum reserves: A systematic review of theory and applications
Masoud Seyyedattar, Sohrab Zendehboudi, Stephen Butt
Abstract
Masoud Seyyedattar, Sohrab Zendehboudi, Stephen Butt
Abstract
Production from offshore reservoirs has contributed significantly to supply required hydrocarbons in the past several decades. The world's ever-growing energy demand has made the industry extend exploration and production activities to much deeper waters and harsher offshore environments. However, development of offshore reservoirs is a challenging task, and the role of efficient and reliable reservoir characterization tools in successful offshore developments cannot be overstated. Reservoir analysis in terms of determination of the properties of rock and fluids and their interactions constitutes a major part of reservoir characterization. Traditionally, laboratory tests and measurements are considered the most reliable characterization approach. However, in many cases, reservoir rock and fluid samples might be unavailable or insufficient for all the required tests and measurements. Empirical correlations and theoretical relationships can be employed in the absence of or to augment experimental data. However, characteristic predictions obtained from this deterministic strategy can be highly uncertain due to sensitivity to the property ranges, oversimplifying assumptions, and local variations of properties in heterogeneous reservoirs. Various forms of connectionist techniques have recently been used as an alternative method for property predictions with higher accuracy and reliability. Yet, their applications have been mostly limited to research studies rather than industrial projects. Recently, Molecular Dynamics (MD) simulations have gained considerable attention in studying reservoir rock and fluid properties, interactions, and associated phenomena at the atomic level. In MD, the properties of interest are extracted from the analysis of time evolution of atomic positions and velocities by numerical solution of Newtonian equations for the motion of all the atoms in the system. This technique facilitates performing “computer experiments” that may otherwise be impossible, extremely costly, or very dangerous to conduct. In this paper, the MD simulation technique and its applications to investigate rock and fluid properties are reviewed. The MD theoretical concepts and procedures are addressed, particularly in reservoir analysis. The main advantages and disadvantages of MD are listed. This review manuscript will provide useful guidelines to characterize the reservoir rock and fluid and their behaviours in various petroleum reserves, which considerably help to conduct better design and optimal operation of production plants.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Production from offshore reservoirs has contributed significantly to supply required hydrocarbons in the past several decades. The world's ever-growing energy demand has made the industry extend exploration and production activities to much deeper waters and harsher offshore environments. However, development of offshore reservoirs is a challenging task, and the role of efficient and reliable reservoir characterization tools in successful offshore developments cannot be overstated. Reservoir analysis in terms of determination of the properties of rock and fluids and their interactions constitutes a major part of reservoir characterization. Traditionally, laboratory tests and measurements are considered the most reliable characterization approach. However, in many cases, reservoir rock and fluid samples might be unavailable or insufficient for all the required tests and measurements. Empirical correlations and theoretical relationships can be employed in the absence of or to augment experimental data. However, characteristic predictions obtained from this deterministic strategy can be highly uncertain due to sensitivity to the property ranges, oversimplifying assumptions, and local variations of properties in heterogeneous reservoirs. Various forms of connectionist techniques have recently been used as an alternative method for property predictions with higher accuracy and reliability. Yet, their applications have been mostly limited to research studies rather than industrial projects. Recently, Molecular Dynamics (MD) simulations have gained considerable attention in studying reservoir rock and fluid properties, interactions, and associated phenomena at the atomic level. In MD, the properties of interest are extracted from the analysis of time evolution of atomic positions and velocities by numerical solution of Newtonian equations for the motion of all the atoms in the system. This technique facilitates performing “computer experiments” that may otherwise be impossible, extremely costly, or very dangerous to conduct. In this paper, the MD simulation technique and its applications to investigate rock and fluid properties are reviewed. The MD theoretical concepts and procedures are addressed, particularly in reservoir analysis. The main advantages and disadvantages of MD are listed. This review manuscript will provide useful guidelines to characterize the reservoir rock and fluid and their behaviours in various petroleum reserves, which considerably help to conduct better design and optimal operation of production plants.
Key concepts: Submarine pipeline, Characterization (materials science), Petroleum engineering, Reliability (semiconductor), Reservoir modeling, Computer science, Biochemical engineering, Production (economics)