Related Experiment Video
Updated: Sep 29, 2025
07:23
Experimental Study of the Relationship Between Particle Size and Methane Sorption Capacity in Shale
Published on: August 2, 2018
7.6K
Novel Self-Adaptive Shale Gas Production Proxy Model and Its Practical Application
Lu Qiao1,2, Huijun Wang1,2, Shuangfang Lu1,2
1Key Laboratory of Deep Oil and Gas, China University of Petroleum, Qingdao 266580, China.
ACS Omega
|March 21, 2022
Summary
This study introduces adaptive proxy models to speed up petroleum production optimization. The new method uses a self-adaptive algorithm to improve prediction accuracy, significantly reducing computational costs.
Area of Science:
- Petroleum Engineering
- Computational Science
Background:
- Production optimization in the petroleum industry is computationally intensive.
- Numerical reservoir simulators are critical but costly for evaluating production functions.
Purpose of the Study:
- To develop adaptive proxy models for efficient production optimization.
- To reduce the computational burden associated with reservoir simulation.
Main Methods:
- Utilized a self-adaptive difference evolution algorithm (SaDE) to optimize least-squares support vector machine (LSSVM) hyperparameters.
- Employed self-adaptive response surface experimental design (SaRSE) for model training.
- Implemented cross-validation for recursive training and evaluation.
Main Results:
- The developed adaptive proxy model demonstrated superior performance compared to traditional regression methods.
- The model maintained high prediction accuracy even with updated experimental data.
- Achieved a fast and accurate approximation of the reservoir simulation model.
Conclusions:
- The proposed adaptive proxy modeling approach significantly enhances production optimization efficiency.
- The method offers a computationally efficient and accurate alternative to traditional reservoir simulation.
- This approach is effective for optimizing block gas reservoir models.
