Dimension optimization for underground natural gas storage pipeline network coupling injection and production conditions. (1st October 2022)

• Record Type: Journal Article
• Authors: Zhou, Jun; Zhao, Yunxiang; Fu, Tiantian; Zhou, Xuan; Liang, Guangchuan
• Is Part Of: Energy. Volume 256(2022)
• Publication Date: 2022-10-01
• Subjects: Underground natural gas storage -- Investment -- Pipeline diameter optimization -- Coupled injection-production conditions -- Hybrid genetic algorithm; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2022.124651
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: With the increasing proportion of natural gas consumption in the energy market, in order to meet the demand for seasonal peak regulation and emergency gas supply, it is urgent to research and develop the underground natural gas storage (UNGS). Different from the conventional oil and gas fields, the UNGS pipeline network needs to consider the boundary constraints under both injection and production conditions. Therefore, considering the characteristics of injection and production technology, this paper aims to constructs a Multiple Condition Hybrid model (MCH model) for optimizing the design parameters of UNGS pipeline network. This paper proposes a Hybrid Genetic Algorithm (HGA) for solving the MCH model of pipeline network design. In the solution of Case 1, HGA has a 10%–13% lower investment cost than GA while shortening the GA iterations by 50%–70%. Case 2 is revealed that the MCH model can be optimized to obtain lower pipeline network costs under the boundary of injection and production conditions. Finally, HGA is used to optimize the design parameters of the MCH model for the field example Case 3, and the pipeline network parameters are obtained that are about 17% lower than the field costs. Highlights: Established a MCH model for dimension optimization of UNGS pipeline network. MCH model coupling injection and production conditions. Proposed a Hybrid Genetic Algorithm based on MFDM and GA. Verified the MCH model and HGA by three cases.