Stratified turbulent wakes: Probing past the cusp into the strongly stratified high Reynolds number regime and machine-learning-based analysis of internal wave radiation
Abstract
Submerged naval wakes in stratified waters operate at values of body-based Reynolds number Re = O(108) and internal Froude number Fr = O(10) to O(103). Previous and ongoing ONRfunded research by the PI has been based on parallel implicit spectral multidomain-based Large Eddy Simulations (ILES) of stratified towed-sphere wakes at Re values as high as 105 and 4~105 and at Fr=4, 16 and 64. The analysis of the resulting datasets has established critical insight into the impact of the phenomenon of stably stratified turbulence (SST) in the life-cycle of the stratified wake. This regime emerges at sufficiently high Re, soon after buoyancy has assumed full control of the original actively three-dimensional turbulent near-wake: turbulence is found to continue to overturn intermittently within high-aspect ratio layers, well after the passage of thewake-generating body. The PI~s research has identified scaling laws, of high predictive utility, for the evolution of the fundamental turbulent non-dimensional parameters in the SST regime. As such, a robust quantification of the persistence time of this regime as a function of Re and Fr is now possible. Moreover, SST is found to become a major contributor to the power released bythe wake into internal waves (IWs). However, despite this unprecedented venture into higher Re,only the two higher Re values examined by the PI operate in the SST regime and actually do so in a marginal fashion: the SST regime lasts for a rather limited time, with the associated turbulence supporting a restricted dynamic range, unlike what is expected for navally relevant Re.Capitalizing on a re-structured, modernized, state-of-the-art, high-accuracy element-based flow solver, the proposed work aims to probe beyond the Re cusp and investigate SST in wakes using well-resolved LES at very high Re extending to values as high as Re = 2~106. At such a value of Re, never before accessed in a simulation or the laboratory, the longer-lasting SST regime will beaccompanied, at a considerable fraction of its duration, by non-patchy, wake-core-filling, highly layered turbulence with an internal scale separation of two decades between largest overturning and viscous scales. In parallel with obtaining an enhanced physical understanding of SST, several aspects of the dependence of IW radiation on Re and Fr will be further studied. Criticalto this regard will be the insertion into the flow solver of a strategically distributed network of high-frequency virtual point probes. The virtual-probe-sampled full frequency content of radiated IWs, spanning from near to far-field, will guide the quantification and parameterization of IW-driven momentum and energy fluxes. Finally, machine learning, in one of its first applications to actual 3-D wave fields, will be used to build an automatic tracking/identification tool for far-field 3-D IW packets, creating the foundation for computing conditionally-based statistics of the population of radiated IW events. In terms of naval interests, current fast-turnaround operational models of stratified wakes do not account for SST and wake-driven IW radation. As such, the predictive ability of such tools islikely to suffer at operationally relevant Re. The findings of the proposed research will greatly enhance the foundation needed to build robust and reliable parameterizations of SST and IWradiation into more practically-oriented computational tools such as self-similarity models, lowresolutionLES and unsteady RANS models.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 24, 2019
- Source ID
- N000141912101
Entities
People
- Peter J Diamessis
Organizations
- Cornell University
- Office of Naval Research
- United States Navy