Fluid-Structure-Sound Interactions and Control

This paper highlights the differences between Computational Fluid Dynamics (CFD) and Computational Aeroacoustics (CAA). The two major differences are the design of the computational algorithm and the distinct differences between boundary conditions for fluid problem used in CFD and fluid and wave boundary conditions used in CAA problems. In regard to computational algorithm, CFD methods are primarily designed to compute fluid flows. Whereas CAA methods are designed to compute fluid flows and the waves supported by the flow. Boundary conditions are specific to each problem. Here three selected problems are considered to show the richness of CAA boundary conditions. First problem involves outgoing waves. Radiation boundary conditions are required to avoid waves reflecting back into the computational domain. Second problem has incoming and outgoing waves at the domain boundary. Boundary conditions that generate the incoming wave and absorbs all outgoing waves are necessary. Third problem involves an incoming mean flow, an incoming wave and a spectrum of outgoing waves at the computational boundary. For this type of problems, boundary conditions imposed at the inflow boundary are required to generate the mean flow, the incoming wave and at the same time be capable of absorbing all outgoing waves.

This paper provides a description of the important noise sources for low-Mach number ducted propellers. It presents an analysis of various noise source strengths using generic acoustic models from the literature. The paper shows that for low-Mach number applications, unsteady noise sources are of major importance. The duct expansion ratio is shown to reduce tonal noise from steady loading because as the expansion ratio increases, the rotor thrust is reduced. Haystacking noise, created by multiple cuts of the same eddy by subsequent rotor blades, is analysed using an analytical approach. The time and length scales within the ducted propeller are shown to significantly affect the intensity of the noise created by this interaction. The relationship between rotor thrust, duct geometry, turbulence distortion and noise is found to be an important research gap that needs to be addressed by future research programs.

This work aims to investigate experimentally the dependence on the friction Reynolds number Reτ of drag reduction (DR) in a turbulent boundary layer (TBL) using plasma-generated streamwise vortices (PGSV). The Reτ examined is from 520 to 750. The developed actuator array produces three pairs of streamwise counter-rotating vortices in a fully developed TBL. The measured maximum spatially averaged drag reduction DRmax reaches 70% over the FE area (210 mm × 240 mm) at Reτ = 520. It has been found for the first time from the empirical scaling analysis of obtained experimental data that the DR scales with the strength of PGSV. which may have a profound impact upon engineering applications.

A surface-mounted flat plate is an important fundamental shape in fluid dynamics, due to the interaction of the plate with the boundary layer on the surface (or ground plane), and the vortex structures produced in the wake. The plate acts as a bluff body when oriented normal to the flow. When angled with respect to the flow, a low-aspect-ratio flat plate may be used as a vortex generator. In the present study, the flow over a surface-mounted flat plate is examined experimentally in a low-speed wind tunnel using a seven-hole pressure probe and particle image velocimetry, with the aim of better understanding the behaviour of the flow in the near wake. This study investigates the wake region for four low-aspect-ratio rectangular flat plates with aspect (height-to-width) ratios of AR = 0.2, 0.35, 0.5, and 1. In each test, the plate was oriented normal to the flow, with a boundary layer thickness to plate width ratio of δ/D = 0.6 and a Reynolds number of Re = 7.5 × 104. For the three lowest aspect ratios, there were two streamwise vortex pairs within the wake, leading to upwards flow at the centerline and one region of downwash on either side. Additionally, in the vertical symmetry plane, these three aspect ratios had a saddle point near the downstream edge of the recirculation zone that defines the flow field and leads to regions of upward and downward directed flow at the edge of the mean recirculation zone. In contrast, for the plate with AR = 1, there is only downwash on the wake centerline, a single pair of streamwise vorticity concentrations, and the saddle point is absent in the vertical symmetry plane.

Motivated by the substantial drag reduction of Plasma-generated streamwise vortices (Cheng et al. in Flat plate drag reduction using plasma-generated streamwise vortices. J Fluid Mech 918, 1), this study explores the feasibility of using one array of miniature vortex generators (MVGs) to reduce the skin-friction in a turbulence boundary layers (TBL). Various MVG arrays, amount to a total of 486, are examined, including different MVG shapes, height, induced angle and transverse spacing. The skin friction is measured using a high-resolution floating element (FE) force balance. It has been found for the first time experimentally that one array of simple triangle shape MVGs may achieve a drag reduction (DR) up to 4.3% at a momentum-thickness-based Reynolds number of $${Re}_{\theta }=$$ Re θ = 14,800. This DR exhibits a strong dependence on the MVG height, induced angle and transverse spacing. With the ease to implement such a simple device in practice, this finding opens opportunities for a wide arrange of engineering applications.

This study investigates the efficacy of hydrophobic coatings in controlling flow separation and turbulence in high Reynolds number flows (Re = 2.2 × 105). Results show that sectoral coating by a hydrophobic fluoropolymer effectively controls the turbulent wake behind a cylinder by delaying flow separation on one side (Lebedev in Phys Fluids 33, [1]). This delay causes the trajectory of large-scale vortex structures to shift. The properties of the coatings remained consistent even after many hours of continuous facility operation.

This work aims to develop a technology to reduce skin friction in the turbulent boundary layer (TBL) of a high-speed train, which is both effective and efficient even at high friction Reynolds numbers (Reτ) that occur in engineering applications say a 600 km/hr super high-speed maglev train. To this end, the high-resolution floating-element force balance developed by Cheng et al. in (J Fluid Mech 918:A24 [1]), Cheng et al. in (Meas Sci Technol 32, 2]) has been substantially modified and improved in order to measure accurately the space-averaged drag reduction (DR) over the control area. Arrays of steady blowing microjets through spanwise slits are deployed, which have been demonstrated to be able to yield a significant DR; meanwhile, its associated surface roughness incurs little additional friction drag. Experiments were carried out with a Reτ range from 1000 to 18,000. Results obtained so far indicate that the maximum spatially averaged DR over the control area may shoot beyond 70%, though the maximum net energy saving occurs when the DR is only 20–30%. This net energy saving grows with increasing free-stream velocity, reaching 15% at 66 m/s. Work is under way to perform experiments to gain the in-depth understanding of the control mechanisms.

The flow separation from a longitudinal slender axisymmetric body is experimentally investigated. The experimental model is a DARPA SUBOFF model with a semi-sphere after-body which is associated with flow separation. The control system employs six tangentially blowing microjets as actuators and a loadcell as the sensor to monitoring the drag force. In conventional open-loop control, the DR is found to be proportional to the total momentum coefficient Cμ of the control jets. At Cμ,ol = 3.23 × 10–2, flow separation is completely suppressed with a pressure recovery of 71.2%, resulting in a considerable DR of 18.23%. However, the control-consumed energy exceeds the energy saved from drag reduction. To improve the control performance, an artificial intelligence (AI) control system is deployed and an unsupervised learning is conducted under an ant colony algorithm with a cost function J = CD. The AI control achieves a DR of 19.8% at Cμ,opt = 0.79 × 10–2, associated with a lower pressure recovery of 94.5%. The power saved from drag reduction is estimated to be 4.9 times of the control input power. The physical mechanism behind the finding is discussed.

The current study aimed to develop a robust and scalable model for numerical investigations of turbulent boundary layer mutually interacting with a compliant panel. The fluid domain is modelled by a hybrid turbulent finite-volume model which is coupled with a structural finite element model (FEM). Some high-performance computing aspects are explained and considered for numerical efficiency of a moderately-refined model. The recently developed Fluent’s native structure model is utilized to enhance coherency of the model, especially when it is to be partitioned for parallel computing. Low Courant number, reasonable y + value and a brief validation (against a Von Karman profile) enhance confidence in the validity of the model and enable temporal and spectral analysis. Illustrative results are presented for short period of real time (0.035 s for both cases) using a fixed time step of 10–4 s. Evaluating displacements of a rubber panel alongside transient readings indicate a wave pattern traveling in the flow direction. Downstream high-frequency components of turbulent fluctuations are shown to gain larger magnitude over a relatively stiff deforming wall.

This work studies experimentally the aerodynamic drag reduction (DR) of a high-speed maglev train (HSMT) model based on artificial intelligence (AI) control, following our successful campaign on the DR of Ahmed bodies. A highly streamlined 3-car HSMT model is used, and the Reynolds number Re is 4.0 × 105 based on the square root of the model cross-section. The aerodynamic drag of the model is measured using two force balances. More than 90 steady jets are deployed on the tail car, in which the dependence of DR on their blowing angles and blowing ratios is documented for each jet. The individual jets produce a maximum DR of 7% and a maximum net power saving of 4%. Seven spatially distributed jets are selected and an AI control system (Zhang et al. in Artificial intelligence control of a low-drag Ahmed body using distributed jet arrays. J Fluid Mech 963 [6]) is deployed to find the best strategies to combine the seven jets in terms of their blowing ratios. Both DR and control power input are incorporated in the cost function. The AI control discovers forcing that creates a DR of 10%. Furthermore, the net power saving reaches about 5% given a DR of 6%.

This paper presents a systematic experimental investigation on a passive flow control of a NACA 0012 airfoil using memetic coverts which are installed on the suction surface. The objective of the present study is to determine the major role of memetic coverts in the aerodynamic performance of the airfoil at post-stall angles of attack. The memetic coverts width w, thickness δ, and its installation position xin are varied from 0.025c to 0.1c, 0.0002c to 0.0005c, and 0.0 to 0.05c, respectively, where xin is measured from the leading edge of the airfoil, and c is the chord length of the airfoil. In the optimum control case, the maximum lift CL,max, and lift-to-drag ratio CL/CD are enhanced as high as 30.18% (α = 18°) and 309% (α = 14°), respectively, relative to the baseline case. Furthermore, the stall angle was delayed by 6°, i.e., from 12 to 18°. By examining the flow around the airfoil, we discover the physics by which the memetic coverts improve the aerodynamic performance of the airfoil.

This paper presents a direct numerical simulation (DNS) of a turbulent channel flow manipulated using steady and unsteady blowing through one array of longitudinal slits, with a view to obtaining skin-friction drag reduction (DR) and understanding how the terms in the budget equation are altered. Investigation is performed at a Reynolds number of 198 based on the channel half height and friction velocity. The DNS code is validated with and without control. It has been found that the steady blowing mode may achieve a spatially averaged DR of 79.6% over the actuation area, exceeding that (56.7%) under unsteady blowing, given the same time-averaged blowing velocity. The two modes further exhibit a difference in affecting the terms in the budget equation. This work also complement experimental findings (Cheng et al. in J Fluid Mech 920, A50, 2021 [1]).

The aerodynamic performance and structural safety of high-aspect-ratio flexible wings can be dramatically unstable once the flutter occurs. In this study, an experimental investigation is performed to understand the mechanism of fluid–structure interactions on a flexible cantilever wing with a high aspect-ratio of 7.5, and develop a passive control method based on bionic leading-edge protuberances. The pitch and plunge deformations of the wing are measured by strain gauges and fiber Bragg grating (FBG) sensors, respectively. With the protuberance, the lift-to-drag ratio is improved in the no limit cycle oscillation (LCO) regime. And the flutter regimes are greatly concentrated. Even in the stall flutter regime, the amplitudes of pitch and plunge are dramatically reduced. Particle image velocimetry measurement shows that the flow separation at high angles of attack is suspended by the protuberances induced streamwise vortices.

In this study effect of using Dynamic Vibration Absorbers (DVA) to attenuate undesired vibration of radar antenna is studied. The vibratory responses of the assembled and disassembled radar antennas are measured. Using the Experimental Modal Analysis (EMA), the lumped model of the antenna and the whole system is obtained. Based on the aerodynamic relations, the frequency of the Fluid–Structure Interaction-based excitation on the antenna is calculated at different wind speeds. The effects of changing the Antenna Tuning Unit’s (ATU) geometry on the vibratory behavior and critical wind speed of the radar antenna are shown. Finally, the DVA is designed for the antenna, and its application to decrease the FSI-based vibration of the antenna is investigated.

We introduce a new definition of vortex formation length for uniform steady flow past a circular cylinder using Proper Orthogonal Decomposition (POD) of the pressure field. Previous definitions each identify a single characteristic length for a given value of Reynolds number. We use the leading modes of the pressure POD to define upper and lower bounds on the vortex formation length. Identifying a range for the vortex formation length is consistent with the hysteresis observed in the critical spacing of two tandem cylinders.

With the gradual maturity of intelligent driving technology, driving in platoon is easy to achieve, which might benefit for energy saving. Hence, it is urgent to discover the mechanism of aerodynamic in platoon condition. Aerodynamic drag and mechanism in two-vehicle platooning in different spacing are studied by numerical method with the DrivAer model. As spacing grows, three configuration models share the same law, but the leading and trailing car appear the opposite drag change compared with that in the isolated condition. Pressure changes of some crucial parts and wake field are analyzed in detail. Some new styles of vehicle shape might be needed for driving in platoon.

We consider bending oscillations of a pipe conveying fluid and lying on an elastic foundation with a non-uniform elasticity coefficient. Despite the absence of local absolute instability, there exists a global growing mode, whose structure is associated with the presence of the turning points (Kulikovskii in J Appl Mech Math, 1993). In the present study, we perform numerical modeling of the development of an initial disturbance of the pipe. In a linear formulation, we demonstrate how a disturbance is transformed into a growing eigenmode. In the nonlinear formulation, we show that the growth of the disturbance is limited, while the oscillations become quasi-chaotic, but do not leave the region bounded by the turning points, which are governed by the linearized problem.

This work aims to conduct a parametric study on the flow induced vibration of an isolated elastic cylinder in axial flow. The cylinder with both ends fixed is free to vibrate in the lateral directions. Large eddy simulation and a two-way coupling CFD-CSM scheme are used to capture the turbulent flow and fluid–structure interaction, respectively. It has been found that the root-mean-square vibration amplitude Arms* of the cylinder exhibits a considerable dependence on a number of parameters, including dimensionless flow velocity $$\overline{U}$$ U ¯ (= 0.65–6.98), turbulence intensity Tu (= 0.7–6.0%), integral length scale Lw* (= 0.2–1.28) of the incident flow and cylinder length-to-diameter ratio L* (= 20–43). It has been found from empirical scaling analysis that Arms* = f1( $$\overline{U}$$ U ¯ , Tu, Lw*, L*) may be reduced to Arms/L = f2( $$\overline{U}_{eff}$$ U ¯ eff ), where f1 and f2 are different functions and the scaling factor $$\overline{U}_{eff}$$ U ¯ eff is interpreted as the effective Reynolds number. Several interesting inferences can be obtained from the scaling law.

Double-suction centrifugal pump (DSCP) exhibits distinctive characteristic of high head in large flow conditions, therefore, DSCP has been widely applied in drainage, irrigation, and other transportation projects. However, the vibration inside the pump due to unsteady flow structure and rotor–stator interaction is raising more and more attention, especially under the command of low vibration and noise environment. In this paper, a DSCP with terrible vibration problem is redesigned through changing the impeller diameter, blade number, and impeller arrangement to enhance the internal flow stability and reduce the impeller-tongue interfere intensity. The results show that, the redesign pump model #2 not only has better hydraulic performance, but also has excellent vibration behavior. Compare the vibration acceleration in vertical direction at outlet flange, and the frequency domain of two models are extracted to make further comparison. The primary blade passing frequency (BPF) in two models is 298 Hz, 347.7 Hz, respectively. The dominant frequencies in frequency domain are both positioned at BPF, and the amplitude in model #2 is much smaller than that in model #1. This phenomenon implies that the vibration in DSCP is induced mostly by the impeller rotation, and the vibration intensity in model #2 is largely reduced after redesign.

This work presents efforts towards accelerating gust load analysis of ultra-high aspect ratio strut-braced wing systems via model order reduction. Reduced Order Models (ROMs) are constructed using the eigenspectrum of the linear aeroelastic system, projecting a Taylor series expansion of the Full Order Model (FOM) onto a reduced basis of eigenvectors. Nonlinearities are retained using source-transformation automatic differentiation to form nonlinear ROM derivative codes. Nonlinear ROMs are formed for a pitch-plunge aerofoil with cubic structural nonlinearity coupled with compressible indicial aerodynamics. Seven modes of the 14 state FOM are required to capture the exact FOM dynamics due to gust-excitation. ROMs are formed also for a strut-braced wing system. The linear dynamics in response to gust excitation of the 986-state strut-braced wing FOM are accurately represented using a basis of 30 modes. The linear ROM performs gust analysis 50 times faster than the linear FOM. The nonlinear ROM has stability issues, where certain modes cause the dynamics to explode. Further investigation of the strut-braced ROM system is required.

The axial-flow-induced vibration (AFIV) of fuel rods in the nuclear power plant is closely related to nuclear safety. In this article, a numerical study of two elastic cylinders arranged side by side in axial flow is performed using the two-way fluid–structure coupling simulation. Large eddy simulation (LES) is employed to predict the turbulent flow. The root-mean-square (rms) vibration amplitude of the cylinder and the critical velocity of buckling instability are found to be in good agreement with experimental data. The vibration of two cylinders in axial flow are simulated at different dimensionless velocities u* ranging from 1.20 to 6.20, turbulence intensity Tu and space ratio P/D. Results show that at low Tu (0.7%), when u* reaches 4.56, both cylinders start to bend outward while vibrating. At higher Tu (2.9 and 5.0%), the two cylinders recover from the outward bending positions. Although Tu significantly affects the amplitude of cylinders, it does not change the vibration frequency of the cylinder and the critical velocity at which buckling instability occurs. As the gap is sufficiently small, the vibration amplitude does not show considerable variation laterally but enhances significantly along the centreline direction.

The wake of a surface-mounted cube has been investigated by means of a large-eddy simulation, with the objective of describing the main characteristics of the dynamic flow under different boundary layers. A laminar boundary layer (LBL) with δ/D = 0.2 and a turbulent boundary layer (TBL) with δ/D = 0.8 have been considered. The spectral proper orthogonal decomposition was used to extract the main antisymmetric and symmetric modes of the flow. A low-order reconstruction using only the main antisymmetric mode pair revealed a closer location of the turbulent horseshoe vortex to the cube and the occurrence of reattachment on the cube’s side faces, which interfered with the vortex formation process for the TBL. Shed structures were composed mainly of streamwise strands, with a well-defined half-loop forming only at the beginning of the shedding process for the LBL. The main symmetric mode pair corresponded to a drift mode. It was found to control the length of the recirculation region in the near wake and the flow over the free end. The drift mode coefficients had a strong correlation with the normal force coefficient of the cube, especially for the LBL.

This work investigates the acoustic radiation of a rotor in interaction with a beam at low Reynolds and Mach numbers. Specifically, non-compressible Implicit Large Eddy Simulations in combination with a solid Ffowcs–Williams–Hawkings formulation are performed for a two-bladed rotor in interaction with circular beams of different sizes. The interest in using numerical simulations partly lies in the possibility to locate and quantify the acoustic sources from different parts of the geometry. The comparison between the numerical results and experimental data shows that acoustic levels are in good agreement for frequencies ranging from the Blade Passing Frequency (BPF) to its ninth harmonics. However, the simulations do not solve for the smallest turbulent structures and hence are not able to capture high-frequency broadband noise. The results obtained with and without beam demonstrates that its presence has no effect on the BPF, but it increases the amplitudes of its harmonics. Using only parts of the surfaces for the application of the FWH formulation, the beam is seen to be the main contributor of noise at the BPF harmonics. Finally, the distribution of acoustic sources along the beam is analyzed, and it varies from a quasi-uniform distribution below the blade for the first harmonics to a none-uniform distribution with maximal value below the blade tip for harmonics with higher orders.

When compared with more slender square prisms, relatively little is known about the flow around surface-mounted square blocks with very low aspect ratios (AR = H/D, where H and D refer to the height and width of the block, respectively). This research investigates the mean flow behavior behind and around surface-mounted low-aspect-ratio square blocks by performing systematic low-speed wind tunnel experiments. Using a seven-hole pressure probe and two-dimensional particle image velocimetry (PIV), the mean velocity fields were obtained for blocks of AR = 0.2, 0.4, 0.6, and 0.8, which were fully submerged in a turbulent boundary layer on the ground plane. The experiments were conducted at a Reynolds number (based on D) of Re = 7.4 × 104. Two-dimensional PIV was used to obtain measurements in the vertical symmetry plane upstream, above, and in the wake of the blocks at an incidence angle of α = 0°. Using a seven-hole probe, measurements were made in the vertical transverse plane at three locations downstream of the blocks. Additionally, for AR = 0.2 and 0.4, α was varied from 0 to 45° in increments of 5°. The mean wake of the blocks is characterized by two pairs of time-averaged counter-rotating streamwise vortices. The presence of the second pair of vortices affects the downwash in the wake of block in a manner unique to low-aspect-ratio prisms. A horseshoe vortex, a free-end vortex, and a near-wake vortex have been identified as structures of interest in the vertical symmetry plane of the blocks at α = 0°.

This study investigates the effects of tip porosity on tip vortex formation and noise generation for a rotor for small unmanned aerial vehicles (UAVs). The rotor-tip porosity reduces the strength of the tip vortex, causing it to dissipate rapidly. The shedding of weakened tip vortex appears to induce a reduction in tonal noise arising from the harmonics of the blade passing frequency. The tip-vortex mitigation by tip porosity results in a decreased turbulence intensity in the rotor wake, leading to a reduction in broadband noise.

This paper investigated the effects of cutting trailing edge serration on airfoil aerodynamic noise based on the acoustic analogy method. Large eddy simulation (LES) was used to capture the transient near-field information of the airfoil, and the Ffowcs Williams Hawkings (FW-H) equation is employed to predict the far-field noise. The research subject is the NACA0012 airfoil, with the length and width of serration ranging from 0.01 to 0.03 and 0.02 to 0.04 times the airfoil chord length (c), respectively. Acoustic experimental data are used to validate the accuracy of the simulation method. The angle of attack of incoming flow is 0°, with Reynolds and Mach numbers of 6.6 × 106 and 0.147, respectively. The research indicates that within the specified parameter range, the noise reduction effect of the serration perpendicular to the incoming flow direction is proportional to both the height and width of the serration. The maximum noise reduction is 3.3 dB. The cutting trailing edge serration leads to an increase in the high-frequency component of the streamwise noise. In the presence of cutting trailing edge serration, the overall sound pressure level of the streamwise noise increases. Compared to embedded trailing edge serration, cutting trailing edge serration have a lower limited height in terms of noise reduction.

The boundary-layer transition effects on the flow and noise characteristics for a high lift configuration was investigated. Improved delayed detached eddy simulation (IDDES) is performed to resolve the turbulent structures and model the transition procedure by coupling k-ω-γ transition model. For comparison, IDDES coupled with shear-stress transport (SST) model is also performed. From the numerical results, IDDES coupled with transition model (IDDES-Tr) can simulate the boundary layer transition and thus exhibit a different distribution of flow variables such as the velocity and Cf. Therefore, the pressure fluctuations and the location of separation show significant differences. The far-field directivities are integrated for different integral surfaces by solving the Ffowcs Williams and Hawkings (FW-H) equation. At all angles, IDDES coupled with full turbulence model (IDDES-FT) indicates stronger noise and the maximum increase can reach 3.2 dB, due to the weaker fluctuations of IDDES-Tr on the main wing and flap surface which are the dominant noise sources in this study.

A dielectric barrier discharge plasma actuator (DBD-PA) was applied to combustion of a premixed gas to control the flame. By using a coaxial DBD-PA with premixed propane and air, combustion could be sustained even at low equivalence ratios that would otherwise cause blowout. Possible reasons for this are flow induced by the DBD-PA and the production of chemically active species. We first visualized the induced flow using incense particles. Next, in order to identify the chemically active species generated by the DBD-PA, optical emission spectroscopy was performed. The results obtained in this study allowed the overall combustion characteristics to be determined in detail.

In the initial region of a round jet, there are periodic vortex rings, which merge to form a large-scale vortex ring that subsequently collapses and becomes turbulent. This study focuses on the merging of vortex rings in a round jet, and aims to clarify the conditions for the regular merging of vortex rings. Acoustic excitation using a sinusoidal wave was used to facilitate the regular merging of the vortex rings. To investigate the merging process of vortex rings, the streamwise cross-section of the excited jets with varying Reynolds numbers was visualized using a laser-light sheet. As a result, it was found that there is a specific range of Strouhal number within which four vortex rings merge. There is a suitable frequency for the merging by acoustic excitation. The reason is discussed for this phenomenon and it is found there is a suitable distance between vortex rings for successful merging.

A resonator array can broaden the effective sound absorption bandwidth by using the structure with multiple cavities in parallel (or in series) in space. However, this kind of fixed mechanical structure cannot flexibly adapt to different noise sources, which severely limits the development prospect of the resonator array in practical applications. In this paper, a time-varying resonator is proposed based on a shunted electrical circuit, realizing the equivalent effect of cavities in parallel in space. Based on the working principle of spatial parallel and temporal series resonators, theoretical models are established, and then two designs for achieving broadband effects are analyzed theoretically. Through time-domain numerical calculation, the sound absorption performance of two different designs with the same cavity volume is obtained. Compared with the spatial parallel resonator, the proposed temporal series resonator has the advantages of tunable frequency and high adaptability of the varying source while achieving similar or even better sound absorption performance.

This work aims to develop an AI-based shape optimization method for the aerodynamic drag reduction of a highly streamlined high speed train (HST) head/tail. This method includes five major modules, i.e., parameterization, design space pre-optimization, reduced order model, machine-learning-based optimization of control variables and adjoint optimization. The aerodynamic performances, such as drag and lift coefficients, of the HST are evaluated numerically via Reynolds-averaged Navier–Stokes equations. This method may deal with a very large number of control points, in the order of 1500, and control variables, up to 40. It has been demonstrated that, given a speed of 400 km/h, the method may reduce the total drag of the HST with 3 carriages up to 9.36%, exhibiting a great potential in the shape optimization for enhancing aerodynamic performance of HSTs.

The dynamics of the cavitation bubble differ depending on the type of boundary near the bubble because the boundary affects the pressure distribution in the vicinity of the bubble. The asymmetric contraction speed of an interface of a collapsing cavitation bubble due to the asymmetric pressure distribution induces a high-speed bubble jet. The cavitation bubble near a rigid solid causes a bubble jet towards the solid surface during the bubble collapse, leading to cavitation erosion on the solid. Recently, a few studies have investigated a gas-entrapping microtextured surface to alleviate cavitation erosion. They confirmed that a bubble jet induced by a collapsing cavitation bubble near air in a pocket directs away from the surface, entrapping the air. In this study, we numerically investigate the hydrodynamic loading on a wall with an air pocket and the alleviation of cavitation erosion by the entrapped air on the solid surface. The presence of an air pocket mitigates the pressure on the wall during the bubble expansion and contraction. However, high pressure is still observed for a small distance between the air pocket and the bubble. The direction of the bubble jet also agrees well with the previous studies, and the tendency of the bubble jet direction is explained by the interaction of the cavitation bubble and the entrapped air in the pocket.

The present study investigates the impact of wormlike micellar droplets on dry solid horizontal surfaces experimentally. Here, the wormlike micellar sample is an aqueous solution of Cetyltrimethylammonium bromide (CTAB) and Potassium bromide (KBr). The effects of impact speed, surface roughness, and the viscosity and elasticity of the liquid phase on the spreading and receding of the droplet are studied in detail. The droplets are formed via 14G and 16G needles and the speed of impact is in the range of 2–3 m/s. The solid surfaces are plates of plexiglass and stainless steel. The results are compared with the impact of equivalent Newtonian droplets to investigate the effect of rheological properties. It is found that fluid elasticity causes a decrease in the spreading and an increase in the receding of droplets. The stress analysis suggests that this normal force is primarily a result of elongational viscosity, while the influence of the first normal stress difference on these dynamics is minimal.

An artificial intelligence (AI) control system is developed to manipulate a turbulent boundary layer over a flat plate at a momentum-thickness-based Reynolds number Reθ of 1450. The system comprises 11 distributed synthetic minijets (actuation unit), 9 wall wires (sensing unit), and a genetic-algorithm-based control unit for the unsupervised learning of optimal forcing. While the exit velocities and excitation frequencies of the synthetic jets are made the same, optimally determined from conventional uniform forcing, the optimal phase shifts between the jets are obtained from the learning process of the AI control experiments so that the cost function or spanwise averaged friction drag estimated from the 9 wall-wires is minimized. Two distinct optimal solutions, albeit yielding the same cost function, are found after the learning process is converged, both yielding a drag reduction (DR) substantially exceeding conventional uniform forcing. The DR mechanism is discussed. This study points to the great potential of AI in finding the optimal solution even when the control parameter number is large.

Up to now, researchers have been proposing various approaches to simulate two-phase flow in order to investigate physical phenomena such as wave breaking and wind-wave interaction. A fundamental aspect of simulating two-phase flow involves defining the physical characteristics of the two distinct fluids present; a technique known as interface construction is widely used to tackle the problem. Level-set and volume-of-fluid methods represent two widely employed methods for establishing an interface between multiple fluids while conserving mass. In this research, a hybrid approach is proposed as a coupled level-set and volume-of-fluid method, effectively combining the strengths of both methods. This research confirmed the reliability of the algorithm through validation with a benchmark case, which was conducted with a Zalesak's rotating notched disc. The simulation was validated by comparing the interface shape with the original one, and the level-set norm and mass norm order were found to be approximately 10–4 and 10–15, respectively. Moreover, an immersed boundary method was formulated to explore the interaction between dam-break flow and an obstacle.

This study analyzed the characteristics of an impinging jet on a concave surface, using thermographic phosphor thermometry (TPT) and numerical method. The experiments were conducted under a jet Reynolds number of 6600, with variations in nozzle diameters (5 and10 mm) and nozzle-to-surface distances (H/d = 2 and 5). The study highlights the accuracy of the RNG k-ε turbulence model in predicting the Nusselt number distribution when compared to alternative models such as SST k-ω and realizable k-ε. The results reveal significant variations in heat transfer characteristics associated with changes in nozzle diameter and H/d (Nozzle to surface distance). A distinctive secondary peak in heat transfer was observed at H/d = 2, attributed to vortex-induced flow detachment and subsequent reattachment. Furthermore, increasing the nozzle diameter was found to enhance jet momentum, turbulence intensity, and overall heat transfer efficiency. This research provides valuable insights into the intricate dynamics of jet impingement on concave surfaces, with potential applications in various engineering fields.

Rapidly climbing aerodynamic drag is one of the major obstacles for developing commercially viable high speed trains (HSTs) up to 400 km/h. This paper proposes an active flow control (AFC) technique based on surrogate model and machine learning algorithm for achieving enhanced HST aerodynamic performance. Five pairs of distributed blowing jets are deployed on the tail car of a HST model, each associated with its own blowing velocity and angle. The ten independent control parameters are optimized in terms of minimizing the drag of the HST model and meanwhile the control power input based on a non-dominate sorting genetic algorithm-II incorporated with an innovative modification of the Kriging model. The results show that the modified Kriging model not only results in a substantially accelerated optimization process, but also proves to be more effective and accurate in global optimization compared to the conventional Kriging model. For a scaled maglev train model which is highly streamlined, a remarkable drag reduction up to 7.3% and net energy savings of 4.6% are achieved at a speed of 108 km/h.