通过强大的机器学习方法,精确地建模原油和盐水的界面张力
Chunyan Liu1, Jing Wang2, Jinshu Wang2
1Department of Petroleum Engineering, Hebei Petroleum University of Technology, Chengde, 067000, Hebei, China. 17783103703@163.com.
这项研究开发了机器学习模型,使用现实的条件预测盐水和原油之间的界面张力 (IFT). 决策树对于估计IFT非常准确,有助于加强石油回收 (EOR) 努力.
科学领域:
- 石油工程是石油工程中的一个.
- 化学工程是化学工程的重要组成部分.
- 数据科学数据科学数据科学
背景情况:
- 介面张力 (IFT) 对于增强石油回收 (EOR) 至关重要,但现有的模型往往使用简化的条件.
- 准确的IFT预测需要考虑实际的原油和盐水中的单个盐类型,反映水库的复杂性.
研究的目的:
- 开发先进的机器学习模型,在现实的储条件下,预测盐水和原油之间的IFT.
- 评估各种机器学习算法的性能,包括决策树,随机森林和神经网络,用于IFT预测.
- 通过敏感性分析确定影响IFT的关键参数及其相对重要性.
主要方法:
- 应用机器学习算法:CNN,AdaBoost,DT,RF,KNN,Ensemble学习,SVM和MLP-ANN. 这些算法包括:
- 利用实验数据,考虑实际的原油特性,个别的盐类型,度,原油API,压力和温度.
- 进行了敏感性分析,使用相关性因子来确定参数对IFT的影响.
主要成果:
- 决策树模型显示出高精度 (R-平方:0.9796,MSE: 5e-4) 和成本效益.
- 而AdaBoost的准确度是最低的 (R平方:0.6696).
- 温度被确定为最有影响力的参数,而盐分子重量的影响最小.
结论:
- 机器学习模型,特别是决策树,可以准确地预测原油/盐水IFT,减少了对广泛实验的需求.
- 开发的模型为优化复杂水库环境中的EOR策略提供了一个实用的工具.
- 了解参数的影响有助于针对性的实验设计和石油回收过程优化.
更多相关视频
08:38Microfluidic Devices for Characterizing Pore-scale Event Processes in Porous Media for Oil Recovery Applications
Published on: January 16, 2018
06:31Studying Surfactant Effects on Hydrate Crystallization at Oil-Water Interfaces Using a Low-Cost Integrated Modular Peltier Device
Published on: March 18, 2020
相关概念视频
Modeling and Similitude
Typical Model Studies
Comparing Intermolecular Forces: Melting Point, Boiling Point, and Miscibility
Temporary attractive forces like dispersion are present in all molecules, whether they are polar or nonpolar. They...
Surface Tension of Fluid
Surface tension varies...
Surface Tension, Capillary Action, and Viscosity
The various IMFs between identical molecules of a substance are examples of cohesive forces. The molecules within a liquid are surrounded by other molecules and are attracted equally in all directions by the cohesive forces within the liquid. However, the molecules on the surface of a liquid are attracted only by about one-half as many molecules. Because of the unbalanced molecular attractions on the surface molecules, liquids contract to form a shape that minimizes the number...
Ideal Solutions
