Bibliometric Analysis of Neural Network in Improving Computational Fluid Dynamics Education

Abstract

The integration of neural networks into computational fluid dynamics (CFD) has led to significant advancements in fluid dynamics simulations, enhancing accuracy and effectiveness. This convergence presents a promising avenue for studying fluid systems and has the potential to revolutionize our understanding of their complex behaviours. To foster collaboration and drive research in this field, a comprehensive bibliometric review was conducted, examining the intersection of fluid dynamics education and neural networks. From a pool of 999 articles gathered from the Google Scholar database, 779 articles published between 2018 and 2023 were selected using the Publishing or Perish (PoP) software. These articles were analysed and categorized using VOSviewer, revealing three distinct clusters representing the primary research topics in computational fluid dynamics and neural networks. The findings of this study provide valuable insights into the current state of research in this emerging field, offering a comprehensive overview of key themes and their relationships. These insights can guide researchers and practitioners in identifying gaps, exploring collaborations, and advancing interdisciplinary approaches. Moving forward, further exploration of the identified clusters and associated keywords can deepen understanding of specific research topics, while investigations into methodologies and techniques employed in the articles can contribute to the development of advanced computational tools and frameworks for fluid dynamics simulations. Collaborative efforts and interdisciplinary approaches hold great potential for advancing our knowledge of fluid systems and driving progress in this exciting field.