A comprehensive review and trends in lubrication modelling

Area of Science:

• Tribology
• Materials Science
• Computational Science

Background:

• Lubrication is critical for mechanical systems, impacting economics and the environment.
• Advanced simulations are accelerating the study of surfaces, lubricants, and additives.
• Modeling lubricated contacts requires balancing continuum and atomistic accuracy due to multiscale phenomena.

Purpose of the Study:

• To explore the challenges in modeling multiphysics tribological interactions across spatiotemporal scales.
• To critically examine current modeling tools, their applications, and limitations.
• To investigate the potential of machine learning in lubrication modeling.

Main Methods:

• Review of state-of-the-art computational modeling tools for tribology.
• Analysis of multiscale simulation approaches (continuum vs. atomistic).
• Exploration of machine learning applications in aggregating data and accelerating simulations.

Main Results:

• Lubrication modeling faces a dilemma in balancing computational efficiency with atomistic accuracy.
• Existing tools have limitations across different spatiotemporal domains.
• Machine learning shows promise for handling multiphysics complexities and enhancing simulation workflows.

Conclusions:

• Accurate and efficient modeling of tribological interactions across scales is crucial.
• Machine learning presents transformative potential for the future of lubrication science.
• Integrating diverse modeling techniques is key to understanding and improving tribosystem performance.