Session J18: Jets II

Linearizing a reacting turbulent jet flow: How to do it, and how not to.

In the last years, linearized methods, such as the resolvent analysis, have helped significantly to enhance comprehension of coherent structures in turbulent flows and how they emit acoustic fluctuations/noise. One specific type of such flows and their acoustic emissions is of significant importance for gas turbine combustion chamber engineers: A turbulent reacting flow, i.e. a turbulent flame. Including the reaction chemistry in the linearized framework would allow to investigate the coherent structures in turbulent reacting flows and the resulting noise emissions by using the resolvent analysis. Furthermore, it could constitute a novel method to tackle the seemingly everlasting problem of thermoacoustic instabilities, a phenomenon with potentially catastrophic consequences in gas turbine combustors.

Driven by this motivation, we investigate the question if and how the strongly non-linear system of a turbulent flame can be treated in a linearized framework. To achieve this goal, we analyze the generic configuration of a perfectly premixed turbulent jet flame. Large Eddy Simulations (LES), which use a progress variable and tabulated chemistry, provide the temporal means. These serve as input to the linear analysis. We demonstrate why a straightforward linearization of the LES reaction model cannot be a valid approach. Instead, we linearize the eddy break up model, a standard RANS reaction model, and superimpose it on the LES mean flow. The linearized reacting flow equations are derived and solved in frequency domain for incomming acoustic perturbations. The resulting flame responses are compared to acoustically forced LES, showing very good agreement. Finally, a perspective on the next necessary steps towards a fully holistic approach to investigate thermoacoustic fluctuations and an appication of the resolvent analysis on the configuration are given.   

The shear layer structure of a tabbed jet in crossflow

Passive devices such as tabs have been shown to be an effective method of altering the characteristics of regular and transverse jets, although the specifics of how tabs affect jet flow fields are not well understood, especially for jets in crossflow. Direct numerical simulations of a tabbed jet in crossflow are performed at a jet Reynolds number of 2000 at jet-to-crossflow velocity ratios of R = 2 and 4. The simulations use the unstructured overset method developed by Horne & Mahesh [J. Comput. Phys (2019) 397: 108790] to allow for the study of two tab positions with respect to the crossflow direction. The jet structure is shown to vary significantly with the tab placement and R, with upstream tabs producing secondary vortices with remarkable similarities to those observed in plane shear layers.   

Effects of Initial Shear Layer State on Screech in a Rectangular Jet

Jet screech from a 4:1 rectangular nozzle at under-expanded supersonic conditions is studied using high-fidelity large-eddy simulation data. By creating a small groove on the nozzle surface, we numerically modify the boundary layer and the jet initial shear layer. A total of 5 cases with different groove sizes are studied in comparison with the no-groove baseline case. The tripping method increases the turbulent kinetic energy and boundary layer thickness at the nozzle exit. The modification in the initial shear layer leads to different shear layer thicknesses and shock cell decay rates near the end of the jet potential core. The screech amplitude is varied in the tripped cases, with largest difference measured to be 2.9 dB compared to the baseline value. The dominant coherent structures associated with screech generation are studied with spectral proper orthogonal decomposition. The amplitudes of the guided upstream-traveling wave and the downstream-traveling Kelvin-Helmholtz wave are calculated. The differences in the screech tone amplitude are found to be related to the strength of the Kelvin-Helmholtz wave, due to the modified receptivity of the initial shear layer. The gain of the guided upstream-traveling wave from the K-H wave and shock interaction is estimated.

    

Large-eddy simulation and modal analysis of low-frequency jet physics

Advances in high-performance computing have enabled high-fidelity simulations of turbulent jets. To alleviate the cost of such simulations, the jets are typically truncated in the streamwise direction just tens of nozzle diameters from the nozzle exit. These simulations can therefore only capture flow structures that are supported within their truncated domains. In contrast, coherent structures that oscillate at low frequencies exist over large spatial scales, which can exceed the boundaries of finite domains. They are thus inaccurately captured at best, completely discarded at worst. This is at odds with past experimental observations showing that low-frequency structures in fact possess significant energy. Linear stability analyses have also demonstrated the effects of domain truncation on the calculation of jet instabilities. To accurately simulate low-frequency jet dynamics, in this work we extend an experimentally validated large-eddy simulation of a subsonic turbulent round jet. We elongate the original domain length from 30 to 100 nozzle diameters, and build a long-time database for statistical analysis with spectral proper orthogonal decomposition, focusing on the eduction of low-frequency coherent structures.   

Toward resolvent-based estimation and control of high-speed turbulent jets

Reducing noise produced by high-speed jets is an important goal for civil and military aviation.   Strong links exist between turbulent structures in the jet nozzle, downstream wavepackets, and far-field noise; resolvent analysis offers an effective lens to examine these complex interactions.  This work aims to build on the success of resolvent analysis as a jet-noise model by using recently developed resolvent-based estimation and control tools (Martini et al., J. Fluid Mech. Vol. 937: A19, 2022) to cancel noise-generating wavepackets and reduce far-field noise.  We discuss the implementation of these tools in a large-scale unstructured compressible solver and provide an initial demonstration of their capabilities by estimating and controlling axisymmetric wavepackets in a high-speed jet.  We also explore a data-driven approach for building the resolvent-based estimation & control kernels as well as other considerations such as the placement of sensors/actuators, the selection of control targets, and the extension from the axisymmetric framework to the fully three-dimensional case. 

    

Active Control of Noise in Supersonic Rectangular Jets

High-speed jets generate extreme levels of noise, resulting in hazardous working conditions for crew, and detrimental effects on aircraft structure and environment. Using large-eddy simulations, we explore the noise-reduction potential of Localized Arc Filament Plasma Actuators (LAFPA) based active control, in the context of a Mach 1.5 perfectly expanded jet exiting a 2:1 aspect ratio rectangular nozzle. LAFPA control at Strouhal number (St)=1 and duty cycle of 50%, decreases the peak far-field noise levels by attenuating radiated energy within the spectral bands, St=0.2-0.4 and St=0.65-0.8. The effect of control is most visible on the major axis plane. LAFPA actuation induces adjacent horseshoe vortices staggered in the streamwise direction within the shear layer of the jet. The resulting manipulation of coherent structures reduces the supersonic component of energy within the column mode of the jet by redistributing it into a higher spectral band at St=1. Further, LAFPA promotes azimuthal percolation of energy into the first and second helical modes, thus reducing the overall radiative efficiency of the jet. An energy analysis confirms the effectiveness of LAFPA control by identifying the weaking of noise-causing vortex intrusion mechanisms in the potential core of the jet.   

Input-output modes and their relevance to jet noise reduction of supersonic jet flows from bi-conical nozzles.

Input-output analysis is used to describe the radiated noise of the shock-laden, supersonic exhaust generated by an axisymmetric bi-conical jet nozzle. Large-eddy simulations (LES) are used to predict the unsteady, turbulent jet at a fixed nozzle pressure ratio (NPR) of 4 and three total temperature ratios (TTR) of 1, 3, and 7. The non-modal, input-output analysis is performed about each time-averaged base flow, providing the optimal linear forcing and corresponding response modes, using a singular value decomposition of the input-output operator. Additionally, sub-optimal modes associated with a smaller gains are also computed. The physical relevance of the resulting modes are evaluated through projection onto the unsteady data. Sensitivities of the gains to the TTR are measured and compared with analytical predictions.   

Dynamics of Unsteady Turbulent Plane Jets Relevant to Synthetic Jet Control

Incompressible Direct Numerical Simulation (DNS) of a planar pulsed jet for Strouhal (0.05, 0.3, and 0.6) and Reynolds (3000 and 10000) numbers are used to understand the effect of inlet conditions in the development of vortical structures, their interactions among themselves and with turbulence, that are produced by the pulsations. This study is motivated by our interest in the dynamics of synthetic jets. With low St, the formed turbulent shear layer tends to roll into a vortex pair with higher circulation, this enhances the global flux of momentum and mass. When turbulence intensity on the inlet flow is increased, no pairing is observed. Low-frequency pulsations generate vortex pairing farther to the inlet with a very weak steady-like jet flow. Further details will be discussed.

    

Resolvent Analysis of a Supersonic Rectangular Multi-stream Jet Flow

We perform a resolvent analysis of a supersonic multi-stream jet flow to understand its principal dynamics, with the ultimate goal of acquiring physical insights for flow control design. The nozzle configuration comprises three primary shear layers: (i) an upper shear layer between the flow on the upper single-sided expansion ramp stream and the freestream, (ii) a splitter plate shear layer from the mixing of a core Mach 1.6 stream and a sonic lower stream which evolves above an aft deck-plate and (iii) a lower shear layer downstream of the aft-deck plate where it mixes with the freestream flow. The time-averaged flow field obtained from an LES is used as the base flow to construct a resolvent operator. A range of frequency-spanwise wave number pairs is considered. For cases with large gains, the dominant response modes are observed in the lower shear layer region downstream of the aft deck-plate. In sub-dominant modes, the response is concentrated in the upper shear layer and the splitter-plate shear-layer regions. Furthermore, a discounted resolvent analysis is performed over a finite-time window by introducing a discount parameter in the resolvent operator. Observations from the resolvent analysis will imply active and passive flow control design and parameter selection.   

Computational investigation of surface enhancements in jet impingement boiling

Numerous studies identify jet impingement boiling as an effective method to transfer energy in high heat flux applications. While most experimental studies so far focus on flat surfaces, there is some evidence that surface enhancements improve the Critical Heat Flux (CHF). Local dry-out is a consequence of stagnating flow initiated by flow obstruction. However, the influence of stagnating flow on CHF due to surface augmentation is an unanswered question. Here, we numerically investigate the effect of surface augmentation in the form of grooves and pin fins, particularly to understand the effect of surface configuration and layout on heat transfer, phase change, and turbulent exchange between the two phases. Heat transfer is predicted using the Eulerian multiphase framework in conjunction with the Rensselaer Polytechnic Institute (RPI) boiling model. Numerical results correspond well with reported surface enhancements subjected to single and array of round jets. A 2D flow boiling study over consecutive micro-grooves suggests that groove dimensions can prevent local dry-out due to the manipulation of flow patterns. A strong correlation is found between flow Reynolds number, average turbulent kinetic energy, and Heat Transfer Coefficient (HTC). A 3D parametric study suggests that surface augmentation can improve heat transfer at the stagnation region if local dry-out is avoided.

    

Computational investigation of the influence of jet parameters for single and multi-jet array impingement boiling

Jet impingement boiling is identified as one of the most promising thermal management techniques for high heat flux applications. Unfortunately, only a few numerical studies have been reported in literature and these are limited to single jets. In this work, we have numerically investigated both submerged single round jets and confined multi-jet arrays in the fully developed nucleate boiling regime. Considerable attention is given to the heat transfer mechanisms (i.e. quenching and evaporation) at the heated surface, which prove to be highly dependent on the jet parameters and heating method (i.e. isoflux or conjugation). The boiling phenomenon of jet impingement was compared against experimental case studies of both a submerged single round jet and a confined multi-jet array impinging on heated copper blocks. The Eulerian multiphase framework with the Rensselaer Polytechnic Institute (RPI) boiling model is used to predict heat transfer, phase change, and turbulence interaction between the two phases. Our study suggests that the modelling of the substrate (conduction) is important to accurately predict surface temperatures for a given heat flux. The results of our parametric analyses showed good agreement with the experimental observations of single submerged jets. Our results also indicate that multi-jet arrays are less sensitive to changes in the jet parameters.