Abstract Accurate far-field boundary conditions (BCs) are crucial for simulating airfoils at high angles of attack, where lift and drag can be comparable and moments become significant. In this study, we investigate the effects of different BCs on a NACA 0012 airfoil in incompressible and steady flow, using the Spalart-Allmaras turbulence model at an angle of attack of $$45^{\circ }$$ 45 ∘ and a high Reynolds number. We apply the impulse form of the lift, drag, and moment equations to a control volume coincident with the square computational domain. Our results confirm that satisfying the lift equation requires modeling the airfoil as a point vortex, but they also reveal the importance of including a point source to capture significant drag effects. Omitting the point source creates artificial blockage at the domain sidewalls reminiscent of wind-tunnel blockage. A “Lagally-Filon” correction for cases with laterally-constrained BCs is derived, demonstrating that it substantially improves agreement with the wind tunnel blockage correction. While combining a point vortex and point source is consistent with both lift and drag, additional minor corrections are needed for the moment equation. These corrections may become essential for designs sensitive to aerodynamic moments, such as vertical-axis wind turbines. Overall, this work emphasizes the need for an appropriate point source term in far-field BCs to achieve more accurate airfoil simulations at high angles of attack.

Abstract Despite the exceptional aerodynamic performance of delta wing-body configurations, the phenomenon of vortex bursting and its influencing factors remain inadequately understood. This research investigates the effects of Mach number, Reynolds number, and body upwash on the angle of attack and location of vortex bursting over a delta wing-body planform within subsonic and transonic flight regimes. Experimental and computational methodologies were employed. Aerodynamic forces and moments were measured using a six-component internal balance in the ASSEF Transonic Wind Tunnel (ATWT), while surface static pressure measurements were acquired with electronically scanned pressure (ESP) modules. The experimental results aligned with previous studies, validating the experimental setup. Numerical simulations were corroborated by experimental surface pressure, force, and moment data. The study initially examines the burst angle of attack, revealing a rise from 12° to 13° as the Mach number increases from 0.3 to 0.9. However, at Mach 0.95, vortex bursting did not occur up to an angle of attack of 15°, indicating a threshold in the transonic regime beyond which vortex bursting is absent at moderate angles of attack. Furthermore, a decrease in Reynolds number leads to a lower burst angle of attack, and the presence of a body significantly reduces this angle compared to wing-only configurations. The study also explores the burst position, observing a nonlinear shift with increasing Mach number. A threshold of 14° for the angle of attack was identified, where further increases in Reynolds number alter the burst position both longitudinally and laterally. The newly identified thresholds provide valuable insights for optimizing the design and improving the understanding of vortex bursting in similar configurations.

Abstract The leading edge vortex (LEV) is one of the most important lift augmentation mechanisms in flapping wing aerodynamics. We propose a methodology that aims to provide a quantitative description of the LEV. The first step of the method consists of the identification of the vortical structures surrounding the wing using the Q criterion. The impact of the employed threshold is shown to be minor, not influencing the observed trends. In the second step, we identify the core of the LEV using a thinning algorithm, discriminating the LEV using the orientation of the locally averaged vorticity vector. Finally, we compute relevant flow quantities along the LEV core, by averaging in planes perpendicular to the local vorticity at the LEV core points. We have applied this methodology to flow data corresponding to a pair of wings performing a flapping motion in forward flight at moderate Reynolds number. We have performed a geometrical characterization of the LEV and we have computed several flow quantities along the LEV core. For the particular configuration under study, we have shown that the LEV, during the first half of the downstroke, develops and grows, increasing its circulation smoothly. Approximately at mid-downstroke, the leading edge vortex starts splitting and its downstream part is advected towards the wake while keeping its circulation rather constant. Finally, we have briefly explored the link between the sectional lift on the wing and the local circulation obtained with the present methodology.

Abstract The quality of the cloud field and temperature field within a large-scale icing wind tunnel is paramount to ensuring the reliability of icing test outcomes. Engineering experience underscore the potential impact of the spray rake configuration within the wind tunnel on the flow field quality within the test section. This study, focusing on the 3 m × 2 m Icing Wind Tunnel (IWT) at China Aerodynamics Research and Development Center, employs a numerical simulation system to investigate how two distinct spray rake configurations affect the uniformity of both the cloud and temperature fields in the test section. The Euler method is utilized to describe the flow field and temperature variations, while the Lagrange method calculates droplet motion. A two-way coupling iteration approach is adopted to solve both the continuous and the discrete phases. Furthermore, a parametric sensitivity analysis is conducted to explore the influence of varying wind speeds and droplet sizes on the uniformity of the cloud field and temperature field. The findings reveal that the middle vertical plate of the spray rake, in the case of the B-type nozzle configuration, exerts a notable influence on the uniformity index of liquid water content within the test section. However, its impact on the droplet temperature field and overall airflow temperature is found to be negligible. This research conclusion offers valuable data-driven insights into the decision of whether to retain the middle vertical plate of the spray rake.

Abstract In this paper, a double time-relaxation kinetic model (DtrKM) is proposed for compressible turbulence modeling on unresolved grids. Within the double time-relaxation framework, DtrKM is extended in the form of generalized Bhatnagar-Gross-Krook model. Based on the first-order Chapman-Enskog expansion, DtrKM connects with the six-variable macroscopic governing equations. The first five governing equations correspond to the conservative laws in mass, momentum and total energy, while the sixth equation governs the evolution of unresolved turbulence kinetic energy $$K_{utke}$$ K utke . The unknowns in DtrKM, including turbulent relaxation time and source term, are modeled via gradient-type assumption and standard dynamic modeling approach. The current kinetic model on unresolved grids correspondingly offers a mesoscopic understanding for one-equation subgrid-scale turbulence kinetic energy $$K_{sgs}$$ K sgs model for compressible large eddy simulation. To solve DtrKM accurately and robustly, a high-accuracy gas-kinetic scheme is developed, which inherits the advantages of well-established gas-kinetic scheme for simulating macroscopic governing equations. Three-dimensional decaying compressible isotropic turbulence and temporal compressible plane mixing layer on unresolved grids are simulated to evaluate the generalized kinetic model. The performance of key turbulent quantities up to second-order statistics confirms that DtrKM is comparable with the widely-used dynamic Smagorinsky model. The DtrKM provides a workable approach for compressible turbulence modeling and simulation on unresolved grids.

Abstract This study investigates the effect of an owl-inspired serrated trailing edge on a symmetric Joukowski airfoil with a 12% thickness during the initial acceleration phase. Large eddy simulations were performed at various Reynolds numbers ( Re = 100,000, 250,000, 400,000, and 500,000) at a Mach number of 0.25 and a 5° angle of attack. At a Reynolds number of 250,000, a detailed analysis reveals that adding serrations is advantageous for enhancing the airfoil’s boundary layer stability and reducing noise with only a minor compromise in aerodynamic efficiency that improves over time and at higher Reynolds numbers. Analysis of aerodynamic and aeroacoustic influence on the flow reveals that laminar separation bubble breakdown on the conventional trailing edge airfoil is triggered by trailing edge sound waves. On the other hand, the serrations promote an earlier transition of the turbulent boundary layer without forming a laminar separation bubble. Across higher Reynolds numbers, the serrated trailing edge’s benefits persist, with earlier noise generation and boundary layer transition benefiting the serrated trailing edge’s non-dimensional force in the y -direction. Overall, adding serrations improves boundary layer stability and reduces trailing edge noise during both the acceleration and post-acceleration phases. These findings underscore the potential of serrated trailing edges for various applications and motivate further optimization efforts.

Abstract The paper considers a two-dimensional laminar flow of a viscous incompressible fluid past a flat plate at high Reynolds numbers. In the framework of the asymptotic theory of viscous-inviscid interaction, the effect of a body moving downstream at a low velocity relative to the plate on the Blasius boundary layer is studied. A special case is investigated, where an external small body, modelled by a potential dipole, moves downstream at a constant speed. This classical problem is formally unsteady in the plate's reference frame; however, as a result of the transition to a moving coordinate system associated with the dipole, it is described by steady solutions of the interaction theory but on a wall moving upstream. The paper proposes a technique to solve this problem containing counterflows, i.e., a layer of fluid flowing upstream is near the surface, while above it, the fluid in the boundary layer flows downstream. It was possible to find an exact analytical solution to the linear problem in the near-wall viscous sublayer for small moment of the potential dipole. The solution contains closed and open separation regions near the line of zero streamwise velocity, even in the linear approximation. At wall velocities below a certain critical value, the disturbances in the viscous sublayer are concentrated mainly downstream of the dipole; at wall speeds above the critical value, the disturbances form an upstream ‘wake’ propagating in the near-wall upstream layer.

Abstract Current research on active flutter suppression considering time delays tends to focus on fixed time delays. To address situations where the control loop may experience time-varying delays with uncertainty, a time-varying-delay Active Disturbance Rejection Control (TVD-ADRC) is proposed. First, a parameterized unsteady aerodynamic reduced-order model (ROM) based on a long short-term memory network is introduced into the aeroservoelastic modeling. This model is applied to predict unsteady aerodynamic forces and aeroservoelastic (ASE) behaviors across a wide range of Mach numbers. Its effectiveness in capturing the characteristics of unsteady aerodynamics is validated through comparisons with the high-fidelity computational fluid dynamics (CFD) simulations. Second, the proposed method integrates ADRC with a delayed input and a time-delay identification module in the controller design. Specifically, the time-varying delay is identified using the cross-correlation function method with a moving window, and this method dynamically updates the time-delay compensation module. Additionally, a genetic algorithm is employed to optimize controller parameters, and the integral of the time-weighted absolute error is selected as the performance evaluation index for the control system. Finally, a three-degree-of-freedom aeroservoelastic system of an airfoil with a trailing-edge control surface is studied for flutter suppression. Flutter control under uncertain time-varying delays during flutter occurrence is investigated, and the impact of the magnitude of the time delay on the effectiveness of the flutter control is analyzed. Simulation results indicate that the proposed TVD-ADRC controller could effectively suppress the aeroelastic instabilities across a wide range of Mach numbers and effectively counteract the negative effects of time-varying delays.

Abstract Acquiring acoustic modes with high quality data in large-scale nacelles is quite challenging in the engine industry because of the complex configuration, high flow speed, tremendous number of acoustic modes, and some other extraordinary interference. A complete procedure for mode detection in the engine industry that is applicable to full-size situations is proposed. Two different array patterns are adopted: a circular array for azimuthal modes in both the intake and bypass ducts, and a rotating linear array for radial modes only in the bypass duct. The azimuthal locations of sensors in the circumferential array are non-uniformly distributed to get more modes than the Nyquist limit. For each individual channel signal, an adaptive resampling method is adopted to reduce the components incoherent with source rotation and frequency shifts caused by shaft speed variation. At high flow speeds, boundary turbulence contaminates acoustic signals of wall-flush mounted sensors. A wavenumber decomposition method is used to separate the acoustic part and the dynamic pressure part in the bypass duct during radial mode detection. Finally, both the azimuthal and radial acoustic modes in bypass and intake ducts are acquired successfully.

Abstract This paper shows that lumped directed-area vectors at edges and dual volumes required to implement the edge-based discretization can be computed without explicitly forming the dual control volume around each node for triangular and tetrahedral grids. It is a simpler implementation because there is no need to form a dual control volume by connecting edge-midpoints, face centroids, and element centroids, and it also reduces the time for computing lumped directed-area vectors for a given grid, especially for tetrahedral grids. The speed-up achieved by the proposed algorithm may not be large enough to greatly impact the overall simulation time, but the proposed algorithm is expected to serve as a major stepping stone towards extending the edge-based discretization to four dimensions and beyond (e.g., space-time simulations). Efficient algorithms for computing lumped directed-area vectors and dual volumes without forming dual control volumes are presented, and their implementations are described and compared with traditional algorithms in terms of complexity as well as actual computing time for a given grid.