Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
The idea of this paper is to improve the techno-economic efficiency of underground gas storage by automatic determination of its optimal control using novel IT methods. Modeling of storage dynamics with the use of reservoir simulators causes that its control optimization is a black-box problem with a high evaluation time. As a result, the applicability of traditional optimal control methods is limited what makes the optimization of underground gas storage control a complex and complicated problem. An alternative solution is to use artificial intelligence methods, which have a great potential in solving complex engineering problems. Hence, a novel control optimization algorithm that provides a self-improving automatic control of the underground gas storage has been created in this work. It is based on the artificial intelligence combined with control theory and reservoir simulation. The aim of this work is to determine such control of underground gas storage, which maximizes the total amount of energy that can be obtained from gas extracted during the production cycle. The original solution proposed in this paper is to use the coupling of a parameterized decision tree with the optimization tool and the reservoir simulator. Storage control defined on the basis of a decision tree allows its unambiguous physical interpretation, while tree parameterization consisting in replacing the limit values assigned to the tree branches by parameters allows optimization of decision conditions. Due to the high reservoir simulation evaluation time the innovative machine-learning type optimization method intended for problems whose computational cost is large was used to optimize the tree parameters. This method is learned which parameter values have a better chance of improving the quality of the solution on the basis of the results collected during optimization. The combination of the optimization tool and the reservoir simulator has been implemented with the use of the Python programming language. While internal programming in the reservoir simulator allowed direct declaration of the parameterized decision tree in the simulation input file. Hence the created algorithm enables full automation of the underground gas storage optimal control determination. This paper presents the detailed explanation of the developed algorithm and includes its exemplary application which illustrates its effectiveness. The obtained results indicate that the proposed algorithm determines underground gas storage control that maximizes its energetic efficiency and allows to achieve additional financial income without any extra investment because only wells control is changed.
The idea of this paper is to improve the techno-economic efficiency of underground gas storage by automatic determination of its optimal control using novel IT methods. Modeling of storage dynamics with the use of reservoir simulators causes that its control optimization is a black-box problem with a high evaluation time. As a result, the applicability of traditional optimal control methods is limited what makes the optimization of underground gas storage control a complex and complicated problem. An alternative solution is to use artificial intelligence methods, which have a great potential in solving complex engineering problems. Hence, a novel control optimization algorithm that provides a self-improving automatic control of the underground gas storage has been created in this work. It is based on the artificial intelligence combined with control theory and reservoir simulation. The aim of this work is to determine such control of underground gas storage, which maximizes the total amount of energy that can be obtained from gas extracted during the production cycle. The original solution proposed in this paper is to use the coupling of a parameterized decision tree with the optimization tool and the reservoir simulator. Storage control defined on the basis of a decision tree allows its unambiguous physical interpretation, while tree parameterization consisting in replacing the limit values assigned to the tree branches by parameters allows optimization of decision conditions. Due to the high reservoir simulation evaluation time the innovative machine-learning type optimization method intended for problems whose computational cost is large was used to optimize the tree parameters. This method is learned which parameter values have a better chance of improving the quality of the solution on the basis of the results collected during optimization. The combination of the optimization tool and the reservoir simulator has been implemented with the use of the Python programming language. While internal programming in the reservoir simulator allowed direct declaration of the parameterized decision tree in the simulation input file. Hence the created algorithm enables full automation of the underground gas storage optimal control determination. This paper presents the detailed explanation of the developed algorithm and includes its exemplary application which illustrates its effectiveness. The obtained results indicate that the proposed algorithm determines underground gas storage control that maximizes its energetic efficiency and allows to achieve additional financial income without any extra investment because only wells control is changed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.