We investigate the onset of primary instability in the two-dimensional flow past side-by-side counter-rotating circular cylinders using global stability analysis. The Reynolds number range 50 ≤ Re ≤ 200 is considered. The effect of rotation on the neutral stability boundaries is examined, showing that counter-rotation has a stabilizing effect that is highly sensitive to the gap ratio. Different types of eigenmodes associated with supercritical Hopf bifurcations are identified. As the spacing increases, the leading global modes and the wavemaker evolve from a compact merged-wake structure to a weakly coupled double-wake pattern. The perturbation kinetic energy budget further shows that the instability is primarily sustained by energy extraction from the shear production mechanism. The dominant nonlinear wake inherits the leading linear global mode, as confirmed by both direct numerical simulations and dynamic mode decomposition. The present results provide a comprehensive understanding of the role of rotation and gap spacing in determining wake instability, offering valuable insights for applications in flow control.
Flow-induced vibrations (FIV) are critical phenomena in many engineering applications, affecting the stability and performance of structures subjected to fluid flows. This study numerically investigates the effect of a parallel slit on the FIV of an elastically mounted two-dimensional square cylinder at a Reynolds number Re=170 for slit-width ratios of 0≤S/D≤0.25 over a reduced velocity range of 3≤Ur≤40. The results show that adding a parallel slit suppresses vortex-induced vibration (VIV), enhances galloping, and even leads to the emergence of flutter, depending on S/D. Compared with the square cylinder without a slit, at S/D=0.10, the system transitions from VIV to galloping at slightly lower reduced velocities. When S/D increases to 0.15, a transitional regime is observed, in which flutter-induced lock-in precedes galloping onset. At S/D≥0.20, no galloping regime is observed, and the system transitions from VIV to a flutter-dominated response, where the vortex shedding frequency locks onto the cylinder’s vibration frequency. In the VIV regime, increasing slit width monotonically suppresses VIV by weakening near-wake vortex formation and reducing fluctuating lift, as the slit jet disrupts shear-layer roll-up and shifts vortex formation downstream. Furthermore, across all vibration regimes, the surface pressure becomes less negative and the wake pressure deficit decreases with increasing slit width, indicating weakened wake dynamics. The system response remains wake-mode-dominated in the VIV regime for all slit-width ratios. Beyond the VIV regime, increasing slit width suppresses wake-mode (WM) dominance and enhances structural-mode (SM) control, shifting the system response from galloping-dominated behaviour towards flutter-controlled dynamics.
A coupled computational fluid dynamics-isogeometric analysis (CFD-IGA) framework is developed to simulate hydroelastic fluid-structure interaction (FSI) problems involving complex free surface flows. The fluid flow is solved using the finite volume solver OpenFOAM, while the structural response is computed using the isogeometric analysis solver G+Smo. The coupling between the fluid and structural domains is achieved through the preCICE interface, enabling two-way partitioned FSI simulations. Several benchmark cases are investigated to assess the accuracy and robustness of the proposed framework. The dam-break impact case demonstrates the solver's capability to capture highly transient free-surface evolution and the resulting impulsive structural response. The classical Turek-Hron FSI2 and FSI3 cases are used to validate predictions of vortex-induced vibration and to evaluate numerical stability under strong added-mass effects. Furthermore, a modified Turek-Hron configuration with a free surface is investigated to examine the interaction among vortex shedding, structural motion, and multiphase flow dynamics. In addition, the hydroelastic response of an elastic floating plate under regular-wave excitation is simulated to further verify the proposed framework's predictive capability. The results demonstrate that the proposed CFD-IGA framework can accurately capture complex hydroelastic phenomena involving free surface impact, vortex-induced vibration, and wave-structure interaction. The FSI solver provides a robust and efficient numerical tool for investigating hydroelastic problems encountered in marine and coastal engineering applications.
This study experimentally investigates the energy-harvesting performance of a low mass-damping pendulum-type cylinder undergoing flow-induced oscillations in the wake of a stationary upstream cylinder. The downstream cylinder pivots about its upper end and is subjected to a controllable electromagnetic braking torque, enabling direct measurement of the mechanical power extracted from the oscillatory motion. Experiments were conducted in a water channel and cover a wide parameter space, including variations in reduced velocities (in the range from around 4 to 12), centre-to-centre separations (2, 3, 3.5 and 4 diameters) and different levels of applied braking torque (from minimal to the maximum that allows response to take place).From an energy-harvesting perspective, the smallest separation investigated provides the most favourable performance, with power coefficients reaching values of approximately 0.12 over a broad range of reduced velocities. Increasing the separation distance progressively reduces both the peak power coefficient and the extent of the harvesting region. These findings highlight the potential of wake interactions to broaden the operational bandwidth of pendulum-based flow energy harvesters.
Hydroelastic slamming loads are of primary importance for the design of marine and offshore structures involving slender flexible structures. In this study, a combined experimental and theoretical investigation of the water impact of tubular structures is carried out. First, a dedicated experimental campaign is conducted using a laboratory hydraulic shock machine, which provides controlled impact velocities and high repeatability. Aluminium tubular specimens are tested under horizontal and inclined impact configurations. The transient structural response is measured using strain gauges and accelerometers distributed along the tubes, allowing a detailed characterization of the dominant bending mode, maximum response and coupling effects. Based on these observations, a semi-analytical hydroelastic slamming model is then developed to predict the dynamic response of the structures. The structural behaviour is described using an Euler-Bernoulli beam formulation, while the hydrodynamic loading is evaluated through a strip approach based on Wagner-type impact theory and the Fictitious Body Continuation, accounting for flow separation effects. Fluid-structure coupling is introduced explicitly by expressing the hydrodynamic forces as functions of structural displacement, velocity, and acceleration. A modal decomposition reduces the problem to a system of ordinary differential equations, which is solved numerically. The comparison between model predictions and experimental measurements demonstrates that the semi-analytical approach captures the main features of the hydroelastic response, including oscillation period shifts associated with added mass, the existence of distinct impact response regimes, and the velocity-dependent hydrodynamic damping. The proposed framework provides a computationally efficient tool for the analysis and preliminary design of slender tubular structures subjected to hydrodynamic impact.
A circular cylinder mounted on an elastic support and free to translate along a rectilinear path in a uniform cross-current constitutes a canonical problem for investigating vortex-induced vibrations (VIV), a ubiquitous form of flow-induced vibrations driven by flow-body synchronization. When the direction of motion is normal to the current, previous works have shown that VIV can appear for Reynolds number (Re) values as low as 20 (based on cylinder diameter and inflow velocity), below the critical value of approximately 47, marking the onset of flow unsteadiness for a fixed body. The influence of path orientation on the system behavior has been well documented for Re above the critical value, but it remains unexplored in the subcritical range. This is the focus of the present numerical study, which uncovers the progressive expansion of the subcritical-Re region where VIV occur, along with the amplification of vibrations in this region, as path orientation varies from the streamwise to the normal direction. The study also reveals persistent subcritical-Re features related to vibration onset and force-displacement phasing that are independent of path orientation.
The high-loading, low-damping design significantly increases the risk of flow-induced vibration under harsh operating conditions such as near-stall. It is commonly held that Rotating Instability (RI) induces convective NSV (cNSV) in blades. This paper reports, for the first time, the simultaneous coexistence of RI and another blade vibration phenomenon-flutter-observed in a mistuned 1.5-stage compressor test at one working condition. At near stall condition, unsteady pressure probes detected aerodynamic and vibration signals simultaneously. Analysis indicates that the aerodynamic disturbance is characteristic of RI. The vibration signal, correlated with the blade's first-order torsional mode, exhibits multiple peaks spaced at the shaft frequency and aligns with blade tip timing results showing vibration localization. Since the dominant frequencies of the vibration and the RI differ and exhibit no lock-in, it indicates their concurrent and simultaneous occurrence. Furthermore, the initial exponential growth characteristic of the vibration signal confirms the vibration as flutter. The failure of RI to develop into cNSV is hypothesized to be due to the mistuning effect, which prevents the RI disturbance from locking into a structural mode and forming a coherent feedback loop. The results demonstrate that flutter and RI-traditionally considered a cause of cNSV-can coexist simultaneously. According to the results, the presence of RI is not a sufficient condition for the occurrence of cNSV and the lock-in condition should be carefully examined. Consequently, during development, design, and troubleshooting, careful distinction based on time-domain characteristics of disturbance are essential to guide vibration mitigation strategies.
The underwater trajectory of large-caliber projectiles with different nose shape features is systematically studied. Firstly, the deflection mechanism of long conical-nosed projectile is investigated, then the influences of inclined angle, conical nose shape, and conical section length, etc., on the deflection of projectile trajectory are discussed. The instability of projectile motion attitude is primarily derived from the effect of pitching moments, and the load characteristics and deflection degree are influenced by the above factors. The long conical-nosed and long truncated conical-nosed projectiles exhibit similar trajectories. While the inclined angle is within 20°, the projectile experiences a ricochet phenomenon; while in the range of 40°-60°, the deflection trend is nearly the same, gradually rotating from its initial inclined state to a horizontal and then a vertical orientation, and finally moving underwater in a ‘water-exit’ posture; when it reaches 80°, the projectile rotates to a horizontal position but no longer continue to a vertical state. Regarding to the short conical-nosed projectiles, the deflection patterns are similar to those of long ones, but the rate and degree of deflection are both significantly reduced. For the short truncated conical-nosed projectiles, their trajectories almost exhibit no deflection, and the cavity shapes display a small asymmetry feature.
This paper extends a hybrid model qaleFOAM to assess the nonlinear interaction of waves with an array of point-absorber wave energy converters (WECs). The hybrid model uses an adaptive one-way coupling strategy, combining a two-phase Navier–Stokes (NS) solver interDyMFoam with the fully nonlinear potential theory (FNPT)-based solver QALE-FEM. To minimise the computational cost, an improved wave absorbing technique is adopted to absorb the undesirable reflected waves at the inlet/outlet boundary of the NS domain instead of using the conventional relaxation zone technique. The model is first assessed against available single-buoy benchmark data, and then applied to regular head-wave cases involving CorPower-type buoys arranged in line arrays. The effects of wave steepness, PTO damping and non-dimensional array spacing are examined. For the two-buoy cases, increasing wave steepness increases the mean surge and pitch drifts and enhances the asymmetry of the instantaneous absorbed power, while reducing the normalised first-harmonic motion amplitudes. Increasing PTO damping reduces the heave response but increases restraint forces and power fluctuations, and may promote overtopping under the studied conditions. For the three-buoy cases, the q-factor and the absorbed power of each buoy vary non-monotonically with non-dimensional spacing d/λ, with the minimum q-factor observed at d/λ≈0.5. Shorter waves produce stronger shielding effects on the middle and rear buoys, affecting both hydrodynamic loads and tether-force fluctuations. These results demonstrate that, within the tested configurations, nonlinear wave effects, PTO damping and array spacing can influence motion response, power absorption and load distribution among buoys.
Piezoelectric wave energy converters (WECs) offer the dual advantage of capturing ocean wave energy while improving device protection, durability, and cost efficiency. Thin flexible plates coated with piezoelectric layers on both faces can serve as effective plate-type WECs, yet vertical piezoelectric plate configurations have received little attention in the literature. Motivated by this gap, the present study develops a mathematical model to analyze the wave-current interaction of obliquely incident waves with two bottom-fixed vertical piezoelectric plates in finite-depth water under both regular and random wave conditions. The coupling of structural flexibility and piezoelectric effects introduces boundary conditions with higher-order derivatives and complex coefficients. To address this, a semi-analytical approach is formulated by recasting the problem into coupled integral equations using Green’s function and mixed Fourier transform techniques. These are solved via a multi-term Galerkin expansion to evaluate reflection and transmission coefficients along with the wave power absorption efficiency. Further, graphical analyses of plate deflection, bending moments, shear forces and horizontal forces are presented. Results show that the dual-plate configuration significantly enhances wave power absorption across a broad frequency range compared with a single-plate model, highlighting its potential as an efficient and stable piezoelectric WEC system. The coexistence of waves and ocean currents reveals that the present WEC model captures energy more effectively with increasing current speed. Moreover, under random wave conditions, the system exhibits comparable or even higher efficiency outside the resonance region than that observed for monochromatic waves, highlighting its practical applicability in realistic ocean environments.
Flow–induced vibration remains a major reliability risk in heat exchangers and nuclear steam generators, where tube arrays can experience Fluidelastic Instability (FEI) once a critical cross–flow velocity is exceeded. It has been demonstrated that Rotated Square (RS) tube arrays are susceptible to vibration in the streamwise direction (SFEI), yet the roles of tube spacing and loose support conditions are still not well understood. This paper presents an experimental study to systematically analyze the effects of pitch ratio and support clearance on the FEI of a rotated–square tube array subjected to air cross–flow. Pitch ratios (P/D) between 1.25 and 1.70 were investigated, combined with three tube–to–support radial clearances of 0.3, 0.5, and 0.7 mm, to characterize the nonlinear dynamic response of the loosely supported configuration. The results are compared and validated using data from previous studies and demonstrate that the instability threshold increases for increasing values of P/D and tube–to–support clearances within the range examined. In addition, the role of a stiffness–controlled mechanism that requires coupling between the tubes is identified as crucial for SFEI to occur in rotated square arrays.
Flapping-wing turbines (FWTs) have shown competitive power extraction efficiencies relative to conventional rotary turbines, as demonstrated in literature with high-fidelity simulations and limited experiments on physical prototypes. It has become accepted wisdom in academic circles that FWTs achieve optimal performance when they generate a well-timed leading-edge vortex (LEV) that sheds near the end of each half cycle of motion. However, some prior studies have shown optimal efficiency under attached-flow conditions, calling the benefit of LEVs into question and prompting the present investigation. We hypothesized that, for any given FWT motion, attached flow (if achievable) will yield greater FWT efficiency than any case with an LEV, no matter how well-timed. To test this, a modulated discrete vortex method based on a Leading Edge Suction Parameter (LESP), developed previously in literature, was used to predict aerodynamic forces on the foil. The LESP-Modulated Discrete-Vortex Method (LDVM) uses a time-stepping scheme, in which leading edge separation is modulated based on when a threshold value of leading edge suction pressure parameter, LESPcrit, is exceeded. This parameter serves as a proxy for Reynolds number in the pitching and heaving foil cases, i.e., with increasing LESPcrit and Re tending to promote flow attachment. Results from the LDVM predictions corroborated our hypothesis, indicating that, if maintained, attached-flow conditions achieve the highest cycle-averaged efficiency of 49%, outperforming cases with LEVs present. The LEV is thus not to be understood as an efficiency-enhancing mechanism, but rather as an unavoidable compromise under low-Re conditions.
We show how structures made of ice and free to oscillate in a direction perpendicular to the direction of flow oscillate due to the changes in their size as they melt when they are placed in flow. We show this both for a cylinder and for a square prism. In these experiments, we do not change the flow velocity, or the spring constant. Oscillations are observed only due to the change in the size of the structure as a result of melting. In the case of a cylinder, as the structure melts, its diameter decreases and therefore the reduced velocity increases at a constant dimensional flow velocity. As the reduced velocity is increased, the structure starts oscillating, it sheds 2S, 2T and 2P vortices in its wake, and then stops oscillating, at a higher reduced velocity, following the same pattern that has been extensively observed in the VIV response of a 1 DOF cylinder. The amplitude of oscillations for the square prism increases linearly as it melts such that its cross-section turns into a D-section. Then the amplitudes decrease rapidly and oscillations stop. No galloping is observed in the initially-square structure. As the prism melts, it sheds 2S, 2C, and 2T/T+C vortices in the wake before reverting back to 2S shedding.
To achieve hydrodynamic cargo transport, we propose a hydrodynamic design principle for a swept-back microswimmer composed of one puller (pulling squirmer) and two pushers (pushing squirmers). The two pushers are symmetrically arranged around the central puller to form swept-back wing configuration. The swimmer generates large negative pressure regions and two distinct recirculation zones in its wake. We perform two-dimensional (2D) and three-dimensional (3D) lattice Boltzmann simulations at low Reynolds numbers for swimmers with different spread angles and cargos of varying sizes. The results demonstrate that cargos placed at different positions around the swimmer could be effectively trapped by hydrodynamic pressure forces, and the trapping performance is highly sensitive to the spread angle. After being captured, cargos travel either stably or in an oscillatory manner behind the swimmer, depending on their size and initial position. Remarkably, the swimming speed of loaded swimmers is generally higher than that of unloaded ones, and the underlying mechanism is analyzed in detail. To transport large or multiple cargo units, we develop extended swimmer configurations by adjusting the inclination angle of the pushers or integrating additional pusher units. The recirculation zones of those elongated swimmers can trap at least two cargos simultaneously.
