Acta Mechanica Sinica最新文献

This paper studied the effect of synthetic jets on active flow control around a finite-length square cylinder using the large eddy simulation method. Based on the oncoming flow velocity (U_∞) and the model width d, the corresponding Reynolds number is 2.78 × 10⁴. We explored the impact of the momentum coefficient (C_μ) and the dimensionless jet frequency (f*) on a finite-length square cylinder’s aerodynamic forces and flow field characteristics. The square cylinder has an aspect ratio of 5, with one end mounted on a wall and the other end free. The synthetic jet outlet is deployed at the windward leading edge of the square cylinder. It is found that synthetic jets positioned at the top can effectively suppress the cylinder’s aerodynamic forces. Both the momentum coefficient and dimensionless jet frequency influence the control effectiveness. The maximum reductions in total mean drag coefficients (C_d,mean) and fluctuating lift coefficient (C_l,rms) are 4.01% and 50.7%, respectively. With synthetic jet control, the shear flow at the free end of the square cylinder is significantly suppressed, the separation bubble on the top surface disappears, the shear layer at the free end approaches the top surface of the square cylinder, and the turbulent kinetic energy near the free end is significantly enhanced. This study may offer valuable guidance for related engineering applications.

The Richtmyer-Meshkov (RM) instability occurs when a perturbed interface between two fluids undergoes impulsive acceleration due to a shock wave. In this paper, a numerical investigation of the RM instability during the reshock process is conducted using the two-component discrete Boltzmann method. The influence of reflection distance on the RM instability, including both hydrodynamic and thermodynamic non-equilibrium effects, is explored in detail. The interaction time between the reflected shock wave and the material interface varies with different reflection distances. Larger reflection distances lead to a longer evolution time of the material interface before reshock, resulting in more complex effects on the interface deformation, the mixing extent of the fluid system, and non-equilibrium behaviors after reshock. Additionally, while the reflection distance has a minimal impact on mixing entropy before the secondary impact, a significant difference emerges after the secondary impact. This suggests that the secondary impact enhances the evolution of the RM instability. Furthermore, non-equilibrium behaviors or quantities exhibit complex dynamics due to the influence of the transmitted shock wave, transverse waves, rarefaction waves, material interfaces, and dissipation/diffusion processes.

In predicting the stress distributions of porous materials, the unique pore microstructure significantly impacts the stress distribution results. Traditional finite element methods (FEMs) typically require a large number of meshes to achieve a certain level of accuracy, leading to more degrees of freedom to be computed thus reducing efficiency. This paper proposes a three-dimensional Voronoi cell FEM (3D VCFEM) for porous materials that considers both elasticity and thermal strain using the three-dimensional hybrid stress element method and variational principles. Based on the principle of minimum complementary energy, modified complementary energy functionals are derived for cases with and without thermal strain. After scaling the elements, using the Delaunay division method to subdivide the pore element into multiple Delaunay tetrahedra, integration is performed using the Hammer numerical integration method. A three-dimensional stress function considering ellipsoidal shapes is constructed. By assuming and solving a higher-order stress field within the element, the stress calculation results of VCFEM are compared with those from MSC MARC to verify the accuracy and efficiency of VCFEM, and the innovative features and significant contributions of the three-dimensional VCFEM are also elaborated upon.

In the paper, we investigate the outer-layer similarity and coherent structures comparing the smooth-wall and rough-wall turbulent boundary layers (TBLs) in the streamwise-wall-normal plane at a low Reynolds number. Global quantities derived from planar velocity fields, measured via two-dimensional particle image velocimetry, are comprehensively assessed under varied surface conditions. The verification of Townsend’s outer-layer similarity is explored based on the comparisons of the mean velocity flow, velocity deficit, Reynolds stress, and diagnostic plots of the second-order statistics, even at these relatively low Reynolds numbers. The analysis of energy spectra and spatial correlation demonstrates that while the energy associated with large-scale motions attenuates and the small-scale fluctuations amplify near the wall, an outer energy peak is still observed, and the coherent structures at higher wall-normal positions are enhanced, retaining the basic spatial topology. Proper orthogonal decomposition, deemed as a global method based on the energy distribution, elucidates similarities in the low-order modes alongside distinct differences in the energy patterns, in agreement with the mechanisms observed in pre-multiplied energy spectra. Furthermore, flow structures associated with the energy transport, particularly ejection (Q2), sweep (Q4) events, and clockwise vortices, are identified by applying conditional average and quadrant analysis to the filtered velocity fields. The Q2/Q4 events, modulated by vortices induced by surface roughness, collectively delineate the footprints of large-scale fluctuations. Additionally, the small-scale flow fields over the rough surface display the wider and stronger upwash and downwash motions, which can be regarded as an imprint of high-speed momentum. These findings offer novel insights into the behavior of TBLs over both smooth and rough surfaces.

Shock/shock interaction (SSI) poses significant risks to aircraft performance and safety, while the double wedge type-V SSI serves as a classic case. The present study systematically investigates the control effects of low-frequency and high-frequency SparkJet arrays on the double wedge type-V SSI through Ma_∞ = 6.13 wind tunnel experiments. Two distinct discharge forms were employed, capable of generating low-frequency high-energy SparkJet and high-frequency low-energy SparkJet, respectively. During low-frequency discharges, the produced high-energy SparkJet progressively develops downstream and couples with the separation zone, forming a plasma layer that directly impacts the SSI zone. Wall pressures within the coverage of the plasma layer demonstrate a significant decrease, while pressures outside the coverage exhibit an increase. The greatest observed reduction was 52.49%. Although the discharge energy is relatively low during high-frequency discharges, effective control can still be achieved at specific frequencies. When the discharge frequency is set to 20 kHz, the continuous generation of SparkJets results in a notable expansion of the separation zone and a slight attenuation of the SSI, which is more pronounced compared to that observed at 5 and 10 kHz. In addition, the power spectral density and spectral proper orthogonal decomposition analyses reveal that the type-V SSI exhibits three typical characteristic frequencies: a low-frequency signal marked by the separation shock, a sub-high-frequency signal characterized by both the separation shock and the SSI zone, and multiple high-frequency signals near and within the SSI zone. Under the control of SparkJets at three distinct discharge frequencies, the resultant flow fields are all predominantly characterized by their respective discharge frequencies. This leads to the attenuation of the original low-frequency and sub-high-frequency signals. The underlying control mechanisms and causative factors are further discussed.

The reliable initiation of oblique detonation waves (ODWs) represents a critical factor determining the operational performance of oblique detonation engines (ODEs). While previous research has predominantly focused on idealized semi-infinite wedge configurations, such studies have consistently revealed challenges including wave instability, detonation quenching, and compromised engine efficiency. This study presents a numerical investigation of initiation mechanisms and flow field characteristics in ODWs induced by curved surfaces. The analysis employs two-dimensional, multispecies, compressible Reynolds-averaged Navier-Stokes equations coupled with a detailed acetylene combustion model. Key results demonstrate that curved-surface-induced detonations achieve a substantially wider standing range than wedge-induced counterparts, primarily attributable to sustained compression effects generated by concave geometries. Notably, at small wedge angles, the curved surface configuration significantly reduces the detonation initiation distance. The initial formation phase reveals unstable behavior characterized by the large-angle overdriven detonation wave originating from the downstream steep wall section, which subsequently migrates upstream before stabilizing. Detailed examination of the wave structure identifies four distinct components: a curved shock wave (CSW), an overdriven detonation front, a transmitted shock wave, and a supersonic jet flow. Within this configuration, we observe an alternating reflection pattern of expansion waves and compression waves in the supersonic jet region, arising from Type IVr shock-shock interactions between the overdriven detonation wave and the CSW. Viscous effects analysis shows that the gradual curvature transition between the wall and flat plate effectively attenuates shock wave/boundary layer interactions. Furthermore, increased boundary layer thickness is found to significantly alter the ODW morphology while simultaneously inhibiting upstream propagation of the downstream overdriven detonation wave. These findings provide fundamental insights into the complex fluid dynamics governing ODEs, offering valuable implications for the development of more stable and efficient propulsion systems.

Modeling open-hole failure of composites is a complex task, consisting of a highly nonlinear response with interacting failure modes. Numerical modeling of this phenomenon has traditionally been based on the finite element method, but requires to tradeoff between high fidelity and computational cost. To mitigate this shortcoming, recent work has leveraged machine learning to predict the strength of open-hole composite specimens. Here, we also propose using data-based models to tackle open-hole composite failure from a classification point of view. More specifically, we show how to train surrogate models to learn the ultimate failure envelope of an open-hole composite plate under in-plane loading. To achieve this, we solve the classification problem via support vector machine (SVM) and test different classifiers by changing the SVM kernel function. The flexibility of kernel-based SVM also allows us to integrate the recently developed quantum kernels in our algorithm and compare them with the standard radial basis function kernel. Finally, thanks to kernel-target alignment optimization, we tune the free parameters of all kernels to best separate safe and failure-inducing loading states. The results show classification accuracies higher than 90% for RBF, especially after alignment, followed closely by the quantum kernel classifiers.

We propose a new mechanistic framework to unveil the fundamental mechanisms governing multi-cycle plastic strain recovery in nanocrystalline metals. The model uniquely integrates crystal plasticity in nanograins with grain boundary (GB) chemo-mechanics, explicitly resolving atomic flux driven by chemical potential gradients under evolving stress and free volume distributions. Applied to nanocrystalline copper films, our simulations capture transient (10⁻⁷ s⁻¹) and steady-state (10⁻⁸ s⁻¹) strain recovery rates spanning hours to days, achieving quantitative agreement with experimental kinetics across six orders of time scale. Three key advances emerge: (1) GB-mediated atomic diffusion dominates recovery (contributing > 75% of total strain reversal), while dislocation back-stress in nanograins plays a secondary role; (2) recovery cycles induce microstructural evolution through stress-driven free volume redistribution, generating chaotic GB stress states and localized plasticity accumulation at triple junctions; (3) macroscopic strain recovery masks progressive microplasticity in GB networks, revealing a fatigue precursor mechanism inaccessible to conventional models. This work establishes the first predictive link between atomic-scale GB dynamics and macroscopic time-dependent recovery, providing a transformative tool for designing fatigue-resistant nanocrystalline alloys through GB engineering.

Elucidating the relationship between geometrically necessary dislocations (GNDs) and back stress is essential for modeling the strain hardening behavior of polycrystalline materials. This study employs dislocation dynamics simulations to quantitatively assess the impact of GND distributions on the associated back stress at the mesoscale. In a simple cubic lattice, the stress fields generated by elementary GND boundaries, including variations in boundary sizes, dislocation types, and distribution patterns, are systematically analyzed. By taking into account the fluctuation of surface GND density, the calculation of back stress is established using the elasticity theory of dislocations combined with scaling functions. It has been demonstrated that the surface GND density is a critical parameter that controls the amplitude of back stress. Subsequently, the prediction of back stress in face-centered cubic crystalline grains is validated with more realistic GND distributions. Considering identical initial Frank-Read sources, dislocation pile-ups are predominantly formed in coarse grains, yet the resulting surface GND density remains comparable to that observed in smaller grains. This phenomenon is responsible for the similar back stress values in grains of varying sizes. Finally, the activation of cross-slip inhibits the formation of dislocation pile-ups, leading to a linear decrease in back stress with increasing plastic strain.

The miniaturization of electronic components and the increasing density of solder joint arrays have made the reliability testing and simulation optimization of packaging devices increasingly challenging. Effectively capturing the stress within packaging structures has become a critical issue that needs to be addressed in the field of advanced packaging. This research focuses on wafer-level chip packaging structures, exploring the internal stress evolution under thermal cycling loads and proposing a methodology that integrates experimental and simulation approaches based on embedded silicon-based piezoresistive sensors. By leveraging these sensors for the first time, real-time monitoring of stress variations across different regions of power modules was achieved, offering precise characterization of cumulative stress behavior during thermal cycling. The results indicate that the gradual accumulation of internal stress is predominantly driven by the inherent plastic deformation and creep properties of solder materials under cyclic thermal conditions. Based on this, a unified creep-plasticity constitutive model coupled with damage was developed and compiled into a UMAT subroutine, which was then incorporated into finite element software for simulation. The simulation results closely matched the experimental data, successfully replicating the stress evolution pattern during thermal cycling. This study not only elucidates the underlying mechanisms of stress evolution in advanced packaging structures but also validates the feasibility of using embedded sensor technology and enhanced simulation models to tackle the challenge of stress measurement, providing a novel approach and technical pathway for the reliability design and optimization of packaging structures.