Triboinformatic modeling of the friction force and friction coefficient in a cam-follower contact using machine learning algorithms


Tribology International, vol.181, 2023 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 181
  • Publication Date: 2023
  • Doi Number: 10.1016/j.triboint.2023.108336
  • Journal Name: Tribology International
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Applied Science & Technology Source, Chimica, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC, Metadex, Civil Engineering Abstracts
  • Keywords: Tribology, Machine learning, Cam mechanism, Friction coefficient, Friction force, ARTIFICIAL NEURAL-NETWORK, PREDICTION, BEHAVIOR, WEAR, COMPOSITES, DESIGN, BRAKES
  • Karadeniz Technical University Affiliated: Yes


© 2023 Elsevier LtdIn this study, the coefficient of friction and friction force in a cam follower mechanism were estimated using modern machine learning (ML) algorithms. Three different ML algorithms were implemented to the experimental tribological data to estimate the change in the friction coefficient and friction force depending on the cam rotation angle: Artificial Neural Network (ANN), Support Vector Machine (SVM), and Gaussian process regression (GPR). We demonstrated through performance analysis that ML-based models can effectively estimate the change in the friction coefficient and friction force. We also comparatively evaluated the performance of ML-based models. The FF-ANN model estimated the friction force with the best performance while the FC-GPR model was more successful in estimating the coefficient of friction. The models show different estimation performances at different preloads and different cam rotation speeds.