Artificial Intelligence in Fluid Mechanics: A Review of Machine Learning Using Sparse Data

Strategic Insights -

Researchers from Prince Mohammad Bin Fahd University have published a study on the application of machine learning, specifically physics-informed neural networks (PINNs), to address the challenge of working with sparse data in fluid mechanics. Fluid mechanics often involves complex systems characterized by a large number of physical parameters, making it difficult to obtain complete spatio-temporal datasets. The study reviews various applications in fluid mechanics where sparse data is a common problem and evaluates the effectiveness of PINNs in enhancing flow prediction accuracy.

Key Takeaways:

  • The study focuses on the integration of machine learning (ML) and physics-informed neural networks (PINNs) to address the challenge of working with sparse data in fluid mechanics.
  • The research concludes that the use of PINNs can enhance flow prediction accuracy by directly incorporating governing physical equations into neural network training.
  • An overview of diverse PINNs methods, their applications, and outcomes is discussed, demonstrating their flexibility and effectiveness in addressing challenges related to sparse data.
  • The study highlights the potential of PINNs in combining data-driven approaches with established physical theories, indicating that the future of fluid mechanics lies in this synergy.
  • Mouhammad El Hassan, a researcher from Prince Mohammad Bin Fahd University, cited as stating, "The difficulty of obtaining complete spatio-temporal datasets is a common issue with conventional approaches, such as computational fluid dynamics (CFDs) and various experimental methods."
  • Ali Mjalled, Philippe Miron, Martin Monnigmann, and Nikolay Bukharin are also contributors to this research.

Statistics:

  • 10(9):226 is the issue number and page number where the study is published in the journal Fluids.
  • Fluids is a journal published by MDPI AG, with a free version of the article available at https://doi.org/10.3390/fluids10090226.
  • Performance metrics of PINNs, such as accuracy and training time, are assessed in various applications in fluid mechanics.
  • 31952 is the postal code of Al Khobar, Saudi Arabia, where Prince Mohammad Bin Fahd University is located.

Sources:

  • NewsRx. Findings from Prince Mohammad Bin Fahd University in Machine Learning Reported [Machine Learning in Fluid Dynamics-Physics-Informed Neural Networks (PINNs) Using Sparse Data: A Review]. Journal of Engineering. October 13, 2025; p 768.
  • Fluids, Machine Learning in Fluid Dynamics-Physics-Informed Neural Networks (PINNs) Using Sparse Data: A Review. Fluids, 2025,10(9):226.
  • (Fluids - http://www.mdpi.com/journal/fluids).