Between Dec 19, 2024 and Jan 2, 2025, datasets can be submitted to DRUM but will not be processed until after the break. Staff will not be available to answer email during this period, and will not be able to provide DOIs until after Jan 2. If you are in need of a DOI during this period, consider Dryad or OpenICPSR. Submission responses to the UDC may also be delayed during this time.

Browsing by Subject "Turbulent flow reconstruction"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Detecting, identifying, and reconstructing fluid flows
    (2020-11) Wang, Mengying
    The estimation of fluid flow from limited sensor measurements is central for many applications in physics and engineering. Sensing instruments allow us to detect the spatiotemporal signatures generated by the surrounding flows, such that the information about what generated those flows can be learned. Moreover, we can model the flow dynamics if the flow type is classified by analyzing those signatures. The flow reconstruction allows us to estimate the flow field by fusing information from model predictions and measurements. Subsequently, control law can be designed based on the estimation for various applications, such as drag reduction and transient energy suppression. In this thesis, we focus on the sensing and estimation of two typical fluid flows: (1) vortex wakes, and (2) wall-bounded turbulence. The first part of this thesis investigates the detection and identification of vortex wakes. Vortex wakes commonly arise in marine locomotion and can exhibit markedly different dynamics. We propose a vortex wake detection protocol, such that different wakes can be classified using hydrodynamic signals measured at a single location on the nearby surface of a fish-like body. The model parameter of the wakes can be further identified given a short period of the vortex trajectories. In the second part of this thesis, we propose a multi-sensor fusion algorithm to reconstruct the turbulent flow by fusing a linear model, non-time-resolved field measurements, and judiciously placed point sensor measurements. Estimation of turbulent flows is more challenging, due to the breadth of spatial and temporal scales underlying dynamics. Field measurements are constrained to sub-Nyquist sampling frequencies, and point measurements provide spatially limited information. The multi-sensor fusion approach is demonstrated using direct numerical simulation data from the Johns Hopkins Turbulence Database. The results show that model-based multi-sensor fusion is a promising approach to reconstruct turbulence with improved spatiotemporal resolution.