Vibration prediction model and vibration characteristics of mining riser used in deep-sea gas hydrate extraction based on deep-learning
Physics of Fluids · 01.01.2025 · 37 (1)
DOI: 10.1063/5.0245296 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physics of Fluids, online erschienen am 07.01.2025. Verfasst von Xiaoqiang Guo und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 62 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Central Guiding Local Science and Technology Development Fund Project, Natural Science Foundation Project of Hebei Province.
Abstract (Originalsprache)
During the deep-sea gas hydrate mining riser operation, the vibration prediction is obtained using a simplified theoretical model, and its accuracy cannot be effectively guaranteed. Deep learning methods can effectively solve this problem. Therefore, a three-dimensional vibration prediction model for the deep-sea gas hydrate mining riser is established using a long short-term memory network based on deep-learning, which can be trained with the help of the vibration data of the mining riser obtained in the field, and realize the advance prediction of the vibration response of the mining riser in the later period. In order to effectively verify the correctness of the model, a similar principle is used to develop an experimental rig for simulating the vibration of the mining riser under the excitation of internal and external flows. The experimental test results are compared with the model prediction results, and the decision coefficient (R2) reaches 99%, which verifies the correctness of the prediction model. Moreover, to further verify that the model can achieve vibration prediction of the deep-see mining riser, the energy method and Hamilton's principle are used to establish a theoretical model of gas–liquid–solid three-phase flow-induced vibration of the deep-sea hydrate mining riser. The results of the predictions in the later period are compared with the results of the theoretical model calculations. It is found that the coefficient of determination (R2) reaches 94.59%, which further verifies the effectiveness of the deep-learning prediction model. On this basis, the vibration responses of the mining riser are predicted under different shear flows, heave motion parameters of platform, lifting flow rate, hydrate abundance, and hydrate particle size. The influences of operational and structural parameters on the vibration response of the mining riser are investigated, and the vibration characteristics of the mining riser are revealed. The study results provide a theoretical foundation and a predictive modeling tool for the safety of hydrate mining risers.
Förderung
- National Natural Science Foundation of China
- Central Guiding Local Science and Technology Development Fund Project
- Natural Science Foundation Project of Hebei Province
- Postdoctoral Science Foundation of China
- Backbone Talent Project
- Open Project of Key Laboratory of Natural Gas hydrates, the Ministry of Natural Resources
- Open Project of Key Laboratory of Gas Hydrate, Guangzhou institute of Energy Conversion, Chinese Academy of Sciences
Zitieren
APA
Guo, X., Li, Y., Li, Q., Song, Y., Xu, J., & Yang, K. (2025). Vibration prediction model and vibration characteristics of mining riser used in deep-sea gas hydrate extraction based on deep-learning. Physics of Fluids, 37(1). https://doi.org/10.1063/5.0245296
Vancouver
Guo X, Li Y, Li Q, Song Y, Xu J, Yang K. Vibration prediction model and vibration characteristics of mining riser used in deep-sea gas hydrate extraction based on deep-learning. Physics of Fluids. 2025;37(1). doi:10.1063/5.0245296
BibTeX
@article{Guo2025,
author = {Guo, Xiaoqiang and Li, Yingwei and Li, Qi and Song, Yuxuan and Xu, Jie and Yang, Kelun},
title = {Vibration prediction model and vibration characteristics of mining riser used in deep-sea gas hydrate extraction based on deep-learning},
journal = {Physics of Fluids},
year = {2025},
volume = {37},
number = {1},
doi = {10.1063/5.0245296}
}
EUVIMED