一个大规模的数据集和基于物理的神经网络,用于预测多组分水溶液和有机溶液中的粘度
Soheil Kavian1, Arian Zarriz1, Matthew J Powell-Palm1,2,3
1J. Mike Walker'66 Department of Mechanical Engineering, Texas A&M University, College Station, Texas 77843, USA.
The Journal of chemical physics
|February 23, 2026
概括
一个新的数据集和物理信息神经网络 (PINN) 模型准确地预测了复杂工业液体的粘度. 这种方法克服了经典模型的局限性,为多组件解决方案提供了更好的洞察力.
科学领域:
- 物理化学 物理化学
- 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:
- 现代工业液体使用复杂的配方,但粘度模型仅限于简单的成分.
- 由于理想化的假设和缺乏数据,现有的模型与多组件系统作斗争.
- 这种差距阻碍了在与应用相关的组成空间中准确预测.
研究的目的:
- 创建一个全面的数据集为多组分溶液粘度.
- 开发一个具有更高准确度和物理洞察力的预测模型.
- 解决复杂混合物中经典粘度相关性的局限性.
主要方法:
- 创建了44316个粘度测量的数据集,用于最多17个组件的溶液.
- 开发了一个基于物理学的神经网络 (PINN) 模型,以经典相关性为指导.
- 在复杂混合物中对粘度的机器学习残留非理想贡献.
主要成果:
- 经典模型 (Katti-Chaudhuri,增强的Adam-Gibbs) 显示有系统的失败,但保留了趋势信息.
- 在保留数据上,PINN模型表现出比经典和仅数据的ANN具有更高的预测能力.
- 模型性能在不同的解决方案复杂度中保持稳定.
结论:
- PINN模型为多组分溶液粘度提供了前所未有的预测能力.
- 在复杂的混合物中,没有被经典模型完全捕捉到的热现象似乎至关重要.
- 开发的模型和软件应用程序可以帮助设计工业液体配方.
相关概念视频
Viscosity of Fluid
1.4K
Viscosity measures the resistance a fluid offers to flow and deformation. It results from internal friction between layers of fluid moving relative to one another. Dynamic viscosity, denoted by the Greek letter mu (μ), quantifies the force needed to move one fluid layer over another. For Newtonian fluids like water and air, the relationship between the shearing stress and the rate of shearing strain is linear, meaning their viscosity remains constant regardless of the applied stress.
1.4K
Newtonian Fluid: Problem Solving
1.0K
Newtonian fluids exhibit a constant viscosity, meaning their shear stress and shear strain rate are directly proportional. This property ensures a predictable and stable response to applied forces, maintaining a linear relationship between force and flow. Examples include water, air, and light oils, consistently demonstrating this proportional behavior regardless of external conditions.
A velocity gradient forms within the fluid when a Newtonian fluid is placed between two parallel plates, with...
A velocity gradient forms within the fluid when a Newtonian fluid is placed between two parallel plates, with...
1.0K
Typical Model Studies
648
Fluid mechanics model studies often utilize scaled-down systems to predict fluid behavior in full-scale environments, such as river flows, dam spillways, and structures interacting with open surfaces. Maintaining Froude number similarity in river models is crucial, as it replicates surface flow features like wave patterns and velocities.
648
Navier–Stokes Equations
2.4K
For incompressible Newtonian fluids, where density remains constant, stresses show a linear relationship with the deformation rate, defined by normal and shear stresses. Normal stresses depend on the pressure exerted on the fluid and the rate of deformation in specific directions, which determines how fluid flows under varying pressures. Shear stresses, on the other hand, act tangentially across fluid layers. They explain how adjacent fluid layers slide relative to one another, connecting...
2.4K
Accelerating Fluids
2.3K
When a fluid is in constant acceleration, the pressure and buoyant force equations are modified. Suppose a beaker is placed in an elevator accelerating upward with a constant acceleration, a. In the beaker, assume there is a thin cylinder of height h with an infinitesimal cross-sectional area, ΔS.
The motion of the liquid within this infinitesimal cylinder is considered to obtain the pressure difference. Three vertical forces act on this liquid:
The motion of the liquid within this infinitesimal cylinder is considered to obtain the pressure difference. Three vertical forces act on this liquid:
2.3K
Surface Tension, Capillary Action, and Viscosity
33.8K
Surface Tension
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...
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...
33.8K


