For the most recent research topic, please look at Theses.
The skin friction drag in turbulent flow on a wall is significantly larger than the laminar flow at the same Reynolds number. This is a major cause of energy loss in the high-speed transportation such as high-speed trains, aircrafts, ships. In this study, we attempt to reduce such friction drag by using active control or passive control techniques. We also attempt to enhance heat transfer while keeping the friction drag at the same level.
Also in the flow around a bluff body, such as circular and square cylinders, the drag (mainly pressure drag) can cause energy loss. Flow around a bluff body often separates to alternately emit vortices in the wake, which will cause vibration and noise. In this study, we attempt to reduce these by using active or passive control technique. As one of the passive control techniques, we also attempt to dynamically optimize the shape of bluff bodies by using numerical simulations.
Recent development of numerical and experimental techniques has enabled us to easily acquire time-space data of complicated flow including turbulence. However, millions to tens of billions of degrees of freedom of those data make modeling and control difficult. In this research, we are working on a feedback control method of turbulent flow based on linear theory such as Resolvent analysis. We are also working on turbulence modeling and development of turbulent inflow generator using machine learning. Our final goal to construct a feature extraction method by combining linear theory and machine learning.
Fukami et al. Phys. Rev. Fluids (2019). |