15 - Future directions

Abstract

Future directions for research that can extend our understanding beyond the scope of this book, as well as new areas for exploration within the scope of this book are discussed. Turbulent particle-laden flows are necessarily characterized in a high-dimensional parameter space, and opportunities for using simulations in combination with artificial intelligence and machine-learning tools to map out this parameter space are discussed. Emerging frontiers in the discovery and quantification of flow physics, and their mechanistic description in terms of multiscale particle-fluid interactions that can be enabled by such simulations, are then discussed. Our current knowledge of the multiscale nature of particle–fluid interactions already poses significant challenges to theoretical treatment of turbulent particle-laden flows, and principal directions to overcome these are then discussed. Successful application of theories requires robust and accurate models of unclosed terms that appear in the governing equations. This leads to the identification of limitations of current models, and their applicability to a specific range of scales, which in turn limits their ability to represent dependencies on patterns and structures in particle spatial configuration. The development of new theories to lift these limitations gives rise to additional modeling needs that constitute important future research directions. New directions in the development of computational methods and simulations are briefly discussed, and finally the need for collaborative efforts is emphasized.