Harnessing Machine Learning to Predict MoS₂ Solid Lubricant Performance

The article presents a comprehensive study on using machine learning to predict the performance of MoS2 solid lubricant coatings in extreme environments. It delves into the challenges of standard liquid lubricants in harsh conditions and the necessity of solid lubricant coatings like MoS2. The research focuses on optimizing Physical Vapor Deposition (PVD) deposition parameters to achieve specific tribological behaviors. By employing linear and non-linear data analysis methods, the authors identify strong correlations between deposition parameters and material properties, such as density and hardness, and their impact on wear rates. The use of Gradient Boosting Regression Trees (GBRT) models demonstrates excellent predictive capabilities for material properties and tribological behaviors. Notably, the study highlights the significant influence of target conditioning on the initial friction coefficient, offering new insights into optimizing MoS2 coating performance. The work also addresses the challenges and successes of building predictive models with limited experimental data, providing a foundation for further advancements in the field.