Optimization of running-in surface morphology parameters based on the AutoML model

Area of Science:

• Tribology and Materials Science
• Surface Engineering
• Machine Learning Applications

Background:

• Running-in significantly alters surface morphology, impacting friction and wear.
• Predicting post-running-in surface characteristics is vital for optimizing component performance.
• Existing models often lack the precision to capture complex surface evolution.

Purpose of the Study:

• To develop and evaluate machine learning models for predicting running-in surface morphology.
• To identify optimal surface parameter adjustments for specific functional requirements.
• To establish a predictive link between surface design and tribological performance.

Main Methods:

• Application of five common machine learning algorithms to model running-in surface morphology.
• Utilizing surface morphology parameters as input features for model training.
• Evaluating model performance, with a focus on Support Vector Machines (SVM).

Main Results:

• Support Vector Machine (SVM) demonstrated superior performance in predicting running-in surface morphology.
• Identified specific relationships between surface parameters (Sq, Sdq, Sk, Sdc, Sdr) and functional outcomes.
• Provided parameter recommendations for enhanced oil storage, reduced friction, and improved support.

Conclusions:

• Machine learning, especially SVM, offers a robust approach for modeling and predicting running-in surface evolution.
• The study provides practical guidelines for surface design optimization based on desired tribological properties.
• This research bridges the gap in quality monitoring for component lifecycles by linking surface design to performance.