Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm
Using surrogate model to assist parameter optimization of oil and gas reservoir development can greatly reduce the call times of numerical simulator and accelerate the optimization process. However, for serial simulators or parallel simulators with low speedup ratio, the conventional method is still...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-01-01
|
| Series: | Geofluids |
| Online Access: | http://dx.doi.org/10.1155/2022/2056323 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832565261386907648 |
|---|---|
| author | Le Zhang Xin Huang Jiayuan He Xueqi Cen Yongge Liu |
| author_facet | Le Zhang Xin Huang Jiayuan He Xueqi Cen Yongge Liu |
| author_sort | Le Zhang |
| collection | DOAJ |
| description | Using surrogate model to assist parameter optimization of oil and gas reservoir development can greatly reduce the call times of numerical simulator and accelerate the optimization process. However, for serial simulators or parallel simulators with low speedup ratio, the conventional method is still time-consuming. Firstly, an improved surrogate model assisted particle swarm optimization (PSO) algorithm was proposed in this paper. Then, the performance of the algorithm was analyzed using the Rastrigin function. Finally, the key operation parameters of a gas hydrate reservoir by depressurization−to−hot−water−flooding method were optimized with the new method. The results show that the new method only affects the update of the global optimal particle without interfering with the calculation process of the local optimal particles at the early stage of optimization. It realizes the rapid addition of the particle samples through the good parallel features of the PSO algorithm, and therefore, improve the precision of surrogate model in a short time. At the late stage of optimization, it is transformed into a local surrogate model to achieve rapid convergence, when the training time of the surrogate model exceeds the calculation time of the simulator. Both the optimization of Rastrigin function and operation parameters of gas hydrate development reveal that the new algorithm greatly reduces the number of iterations under the same accuracy and thus successfully accelerates the optimization process. |
| format | Article |
| id | doaj-art-cd0a8b25271247cfbfd6c468521bcc5b |
| institution | Kabale University |
| issn | 1468-8123 |
| language | English |
| publishDate | 2022-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Geofluids |
| spelling | doaj-art-cd0a8b25271247cfbfd6c468521bcc5b2025-02-03T01:08:58ZengWileyGeofluids1468-81232022-01-01202210.1155/2022/2056323Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm AlgorithmLe Zhang0Xin Huang1Jiayuan He2Xueqi Cen3Yongge Liu4State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective DevelopmentState Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective DevelopmentState Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective DevelopmentState Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective DevelopmentChina University of Petroleum (East China)Using surrogate model to assist parameter optimization of oil and gas reservoir development can greatly reduce the call times of numerical simulator and accelerate the optimization process. However, for serial simulators or parallel simulators with low speedup ratio, the conventional method is still time-consuming. Firstly, an improved surrogate model assisted particle swarm optimization (PSO) algorithm was proposed in this paper. Then, the performance of the algorithm was analyzed using the Rastrigin function. Finally, the key operation parameters of a gas hydrate reservoir by depressurization−to−hot−water−flooding method were optimized with the new method. The results show that the new method only affects the update of the global optimal particle without interfering with the calculation process of the local optimal particles at the early stage of optimization. It realizes the rapid addition of the particle samples through the good parallel features of the PSO algorithm, and therefore, improve the precision of surrogate model in a short time. At the late stage of optimization, it is transformed into a local surrogate model to achieve rapid convergence, when the training time of the surrogate model exceeds the calculation time of the simulator. Both the optimization of Rastrigin function and operation parameters of gas hydrate development reveal that the new algorithm greatly reduces the number of iterations under the same accuracy and thus successfully accelerates the optimization process.http://dx.doi.org/10.1155/2022/2056323 |
| spellingShingle | Le Zhang Xin Huang Jiayuan He Xueqi Cen Yongge Liu Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm Geofluids |
| title | Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm |
| title_full | Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm |
| title_fullStr | Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm |
| title_full_unstemmed | Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm |
| title_short | Parameter Optimization Study of Gas Hydrate Reservoir Development Based on a Surrogate Model Assisted Particle Swarm Algorithm |
| title_sort | parameter optimization study of gas hydrate reservoir development based on a surrogate model assisted particle swarm algorithm |
| url | http://dx.doi.org/10.1155/2022/2056323 |
| work_keys_str_mv | AT lezhang parameteroptimizationstudyofgashydratereservoirdevelopmentbasedonasurrogatemodelassistedparticleswarmalgorithm AT xinhuang parameteroptimizationstudyofgashydratereservoirdevelopmentbasedonasurrogatemodelassistedparticleswarmalgorithm AT jiayuanhe parameteroptimizationstudyofgashydratereservoirdevelopmentbasedonasurrogatemodelassistedparticleswarmalgorithm AT xueqicen parameteroptimizationstudyofgashydratereservoirdevelopmentbasedonasurrogatemodelassistedparticleswarmalgorithm AT yonggeliu parameteroptimizationstudyofgashydratereservoirdevelopmentbasedonasurrogatemodelassistedparticleswarmalgorithm |