Artificial Neural Networks Improve Performance of Sliding Bearings with Optimized Dimple Textures

Strategic Insights -

Researchers from Indira Gandhi Delhi Technical University for Women have discovered that surface texture can enhance the performance of sliding bearings, with optimized dimple textures leading to remarkable improvements in load-carrying capacity and coefficient of friction. The study, published in the Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, employed Artificial Neural Networks and Genetic Algorithms to optimize the parameters of rectangular dimples on the surface of sector shape pad thrust bearings.

Key Takeaways:

  • The researchers used Reynolds equation incorporating mass-conservation algorithm as a mathematical model to predict the performance parameters of the bearings.
  • Artificial Neural Network model was applied to predict the performance parameters, including load carrying capacity and coefficient of friction.
  • Genetic Algorithms were employed to optimize the dimple parameters, with fitness evaluations based on the prediction model.
  • The findings illustrate a remarkable enhancement in load-carrying capacity (up to 35%) alongside a substantial reduction in the coefficient of friction (by up to 27%).
  • The study highlighted the potential of surface texture optimization using Artificial Neural Networks and Genetic Algorithms to improve the performance of sliding bearings.
  • Additional authors on the research include Dhanishta Sirohi and Manuj Aggarwal from Indira Gandhi Delhi Technical University for Women.

Statistics:

  • 35% enhancement in load-carrying capacity
  • 27% reduction in coefficient of friction
  • Researchers employed Artificial Neural Network model to predict performance parameters
  • Genetic Algorithms were used to optimize dimple parameters
  • 2025 publication date
  • Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology is a peer-reviewed journal
  • Sage Publications Ltd is the publisher of the journal

Sources:

  • VerticalNews, "New research on Artificial Neural Networks is the subject of a report."
  • Optimization of Rectangular Dimple Textured Thrust Pad Bearing Using Artificial Neural Network and Genetic Algorithm. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, 2025.
  • Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology can be contacted at: Sage Publications Ltd, 1 Olivers Yard, 55 City Road, London EC1Y 1SP, England.