Experiments in Fluids最新文献

We present measurements of temperature fields and flow structures in a liquid metal Rayleigh–Bénard convection at a low Prandtl number, which were carried out for the first time using embedded fiber Bragg grating sensors (FBG) in combination with ultrasonic Doppler velocimetry (UDV). The FBG sensors enable minimally invasive, spatially resolved temperature measurements in optically opaque and electrically conductive liquids, thereby overcoming significant limitations of conventional thermocouples and optical techniques. This approach was applied in a cuboid Rayleigh–Bénard cell with an aspect ratio (Gamma = 5) filled with GaInSn. In this paper, we present measurements at two Rayleigh numbers, (textrm{Ra} = 6.8 times 10^4) and (2.1 times 10^5). At the lower Rayleigh number, a coherent three-roll structure is observed with low-frequency modulation of thermal fluctuations. At the higher Rayleigh number, a cellular convection regime emerges, featuring checkerboard-like temperature patterns in the mid-plane and periodic plume emissions. Spectral analysis reveals a dominant oscillation frequency near (f = 0.029) Hz, while autocorrelation and extremum tracking highlight strong temporal coherence near the center and more volatile plume behavior near the sidewalls. The results of the temperature measurements and the UDV velocity measurements are consistent, thus confirming the capability of FBG sensors as a robust tool for investigating the spatio-temporal dynamics in convective systems.

This review aims to provide an overview of laboratory models of nearshore morphodynamics, focusing specifically on the role of wave action. In this context, shoreline evolution is driven by the direction of sediment flux induced by wave dynamics, corresponding to erosion or accretion processes. These processes, which naturally modify the shape of coastlines, are influenced by factors such as sediment availability, wave climate, and soil strength, among others. Starting from unconsolidated sandy materials and the equilibrium-based concepts used in natural beach classification, the reanalysis of laboratory experiments shows that they can reproduce natural conditions and serve as conceptual models for understanding nearshore morphodynamics. However, the temporal evolution of the shoreline and the source of available sand are difficult to capture through equilibrium-type physical processes and therefore requires a specific focus on localized zones of the nearshore, depending on the prevailing hydrodynamics and sediment strength, i.e., consolidation. Accordingly, the local dynamics near the shoreline are closely linked to the behavior of the swash zone and to cliff erosion processes, which constitute a central focus of this review from a laboratory experiment perspective.

In urban environments, pollutant ingress from outdoor sources poses a significant challenge to indoor air quality. Cross-ventilation, while essential for passive cooling and natural airflow, can also facilitate the entry of outdoor contaminants into indoor spaces. To investigate the dynamics of outdoor-to-indoor pollutant transport, the present study employs an idealized configuration, namely, a hollow cube representing a scaled-down model building with window openings in the upstream and downstream faces, subjected to an upstream passive scalar source within an atmospheric boundary layer. The experiments are conducted in two distinct facilities: a water tunnel using Rhodamine dye as the scalar, and a wind tunnel using propane gas, all performed at a specified flow Reynolds number of ( text{Re} = U_{{{text{Ref}}}} H/nu approx 50,000 ) for a fixed boundary layer-to-cube height ratio of about 3; here, ( U_{{{text{Ref}}}} ) is the streamwise velocity at cube’s height (H) measured without the cube. The scalar, released from a ground-level upstream source, is predominantly transported by a streamwise advective flux, while relatively weaker wall-normal advective and turbulent fluxes contribute to vertical dispersion and local mixing. A fraction of the oncoming scalar enters the cube intermittently, through the upstream window. Inside, a central jet-like flow carries the scalar parcels primarily by streamwise advective flux, while also interacting with the upper and lower recirculation regions, enabling scalar exchange across these zones through wall-normal advective and turbulent fluxes. While the time-averaged concentration field inside the cube is nearly uniform, suggesting effective mixing, instantaneous concentration traces exhibit strong intermittency, with sporadic peak events, highlighting the risk of transient peak exposures. The indoor concentration exponentially decays over time once the source is turned off, with a slower decay in the upper recirculation region, implying relatively prolonged exposure near the ceiling region. Both experimental setups produce closely matching values and consistent trends in the spatio-temporal dynamics of scalar concentration, and also highlight their complementary nature, with each offering distinct advantages. The present findings will deepen our understanding of pollutant ingress and mixing in buildings in cross-ventilated flows and also offer valuable insights to future modeling of pollutant exposure in urban indoor spaces.

The relationship between in-cylinder flow structures and subsequent turbulent flame growth in a production spark-ignition engine was investigated using combined phase-locked particle image velocimetry (PIV) and high-speed flame chemiluminescence (CL) imaging via endoscopic access. Measurements were taken in a 4-cylinder production engine operating at 2000 rpm and 75 Nm torque with a stoichiometric fuel–air mixture. For two-dimensional flow field, a double-frame sCMOS camera was used to acquire 143 cycles of particle image pairs at 50°CA bTDC (15.5°CA before ignition timing). Simultaneously, time-resolved flame propagation images were captured every second combustion cycle using a high-speed CMOS camera (Vision Research Phantom v7.3) at a frame rate of 11 kHz. Broadband chemiluminescence imaging began at 50°CA bTDC and continued until 6°CA bTDC. The acquired images were analyzed in conjunction with pressure-derived heat release rates and mass fraction burned (MFB) to elucidate the relationship between engine performance and the physical characteristics of flame propagation. Analysis of instantaneous velocity fields from individual cycles unveiled substantial cyclic variations in flow structure. Combined flow–flame imaging demonstrated that the direction and magnitude of the flow near the spark region significantly influenced spark plasma orientation and early flame kernel development. Correlation map analysis indicated a strong positive correlation between local velocity magnitude and average flame speed, particularly in the vicinity of the spark plug. The high-correlation region is associated with the largest differences in the mean flow fields between slow and fast cycles.

This paper investigates the acoustic and velocity fields due to a circular rod and an aerofoil placed in the wake of, and perpendicular to, a rod. Simultaneous measurements were conducted using a microphone array and time-resolved particle image velocimetry (TR-PIV). The interaction was characterized through acoustic spectra and the coherence between microphone signals and the three velocity components. Coherent structures were identified with spectral proper orthogonal decomposition (SPOD) using a norm based either on turbulence kinetic energy (SPOD-u) or on pressure (SPOD-p). An advantage of SPOD-p is that it identifies velocity modes associated with a large acoustic energy. Peaks of energy were observed at (St approx 0.2) and 0.4–Strouhal numbers based on rod diameter and free-stream velocity. At (St approx 0.2), the dominant feature is von Kármán vortex shedding from the rod. At (St approx 0.4), a wave-train structure in the rod wake impinging on the aerofoil leading edge is captured by the rank-1 SPOD-p mode, with coherence levels reaching 60% for the (u_2) component (upwash/downwash relative to the aerofoil). This structure also appears at (St approx 0.2), but as the rank-2 SPOD-p mode. A mode-switching occurs around (St approx 0.3): below this value, the rank-1 mode corresponds to von Kármán shedding (cylinder branch), while above it, the rank-1 mode tracks the interaction of the aerofoil with the rod wake (aerofoil branch). Both branches were also identified via beamforming using low-rank cross-spectral matrices derived from SPOD-p modes.

In this study, inverted flexible and rigid splitter plates are applied to control the vortex-induced vibration of a circular cylinder. The vortex-induced vibration (VIV) characteristics and vortex dynamics are experimentally investigated with a Reynolds number ranging from 1970 to 10,590. To examine the effect of the streamwise length, five different lengths are selected. It is indicated that the VIV of the circular cylinder is suppressed by the inverted rigid splitter plate and the suppression effect is improved with an increase in the streamwise length of the plate, however, the vibration response behaviors remain the same for all streamwise lengths. Compared to the rigid one, the inverted flexible splitter plate causes various vibration responses with the change of its streamwise length. The diverse vibration responses are related to the kinematic characteristics of the inverted flexible splitter plate and the vortex dynamics. Five vortex shedding modes, including “Kármán vortex”, “Bi-LEV + Bi-WV” (Bi-LEV: bilateral leading-edge vortex; Bi-WV: bilateral wake vortex), “2Bi-LEV + Bi-WV” (2Bi-LEV: two pairs of bilateral leading-edge vortex), “Uni-LEV + Uni-WV” (Uni-LEV: unilateral leading-edge vortex; Uni-WV: unilateral wake vortex), and “K-H instability” (Kelvin–Helmholtz instability), are found based on the number of vortices in one vibration cycle. The correlation between the VIV and the vortex shedding mode is revealed. The “Kármán vortex” and “K-H instability” modes are corresponding to a better effect of suppressing VIV. The control effect of “LEV + WV” modes is affected by the kinematics of the plate, as it is found that the vibration amplitude of the circular cylinder and the inverted flexible splitter plate is positively correlated.

To investigate the typical unsteady structures and gas–solid coupling mechanisms in particle-laden transverse jets in high-enthalpy supersonic flows, a reliable experimental approach was developed and systematically validated. The experimental system consists of a high-enthalpy supersonic flow facility and a custom-designed particle injection device. It generates a Mach 2.63 (~ 1406 m/s) flow with a total temperature of ~ 1633 K, along with continuous and stable injection of SiO₂ particles ranging from 0.1 to 150 μm in diameter. A multimodal diagnostic strategy was employed to provide insights into the flow field. High-speed focused shadowgraphy captured the aggregation and sweeping–ejection processes of large-scale vortices acting on particle clusters. Planar laser scattering (PLS) revealed three representative particle distribution patterns—fluctuation, trailing, and roller type—as well as their quasi-periodic cascading evolution. Two-dimensional, two-component particle image velocimetry (2D–2C PIV) provided quantitative measurements of the particle velocity field. The results showed peak velocities up to ~ 1300 m/s and a pronounced non-monotonic stratification linked to particle inertia effects. These results demonstrate that the developed experimental platform and diagnostic methodology can reliably capture the complex unsteady features of particle-laden transverse jets in high-enthalpy supersonic flows. This provides a robust experimental basis and essential data for advancing the understanding of particle–turbulence mixing and transport mechanisms in high-speed flows.

Hybrid pulse-burst Doppler global velocimetry/pulse-burst, cross-correlation Doppler global velocimetry was demonstrated for the first time in the NASA Langley 4-foot Supersonic Unitary Plan Wind Tunnel, allowing simultaneous acquisition of mean and instantaneous high-repetition-rate velocity fields. The technique was used to make planar velocity measurements across an oblique shockwave generated by a large splitter plate set at a − 2° angle of attack in a Mach 2.4 flow. Sequences of more than 100 consecutive instantaneous velocity fields were obtained at a rate of 100 kHz. Velocity fields from both the mean and instantaneous measurements from the hybrid technique indicated errors less than 1 percent of anticipated velocities. Likewise, assessment of measurement precision yielded velocity measurements of approximately 1.6 percent of the local velocities. Additional demonstrations in different flowfields including a subsonic axisymmetric jet and a supersonic Mars reentry vehicle model further illustrate the utility of the new hybrid method. Evaluation of these other test cases indicated the potential for retroactive extraction of unsteady velocity information from previously acquired mean data.

Graphic abstract

Volumetric three-component particle tracking velocimetry flow measurements were taken to characterize the vortical structures past isolated and sheltered low-aspect-ratio finite wall-mounted circular cylinders (FWMCCs) immersed in a turbulent boundary layer. In addition to an isolated FWMCC case, sheltering effects are investigated for tandem arrangements at streamwise spacings of 2d and 6d, where d is the FWMCCs’ diameter. For each spacing, two height ratios (h_1/h_2 = 1) and 0.5 are considered, where (h_1) and (h_2) are the heights of the upstream and downstream FWMCCs, respectively. Measurements were taken for FWMCCs occupying 20% or less of the incoming boundary layer thickness. For the isolated FWMCC, the wake is dominated by two types of coherent vortical structures: an arch vortex with an inverted U shape and a streamwise counter-rotating vortex pair. The wake of sheltered FWMCCs is also dominated by a coherent U-shaped arch vortex and streamwise vortices; however, sheltering weakens these structures and affects their characteristics. For fully sheltered downstream FWMCCs ((h_1/h_2 = 1)), the wake includes a single streamwise counter-rotating vortex pair originating at the upstream FWMCC. When the free end of the downstream FWMCC is unsheltered ((h_1/h_2 = 0.5)), a streamwise counter-rotating vortex pair is induced at the downstream free end; this vortex pair merges, farther downstream, with the streamwise vortex pair originating at the upstream FWMCC. A third streamwise counter-rotating vortex pair is observed past the downstream FWMCC when it is sheltered by a FWMCC with (h_1/h_2 = 0.5) positioned 2d upstream, further highlighting the dependence of the vortical structure organization on both the height ratio and streamwise spacing.