Acta Mechanica Sinica最新文献

Accurate estimation of a truck’s mass and center of gravity (CG) is critical for optimizing safety and performance but remains challenging due to dynamic uncertainties in weight distribution and road interactions. This study introduces a data-driven mechanics framework integrating four hybrid machine learning (ML) models-tuna search-optimized support vector machine, cuckoo search-optimized BP neural networks, sparrow search algorithm-optimized extreme learning machine, and whale search-optimized XGBoost-to enable estimation. A 17-degree-of-freedom multibody dynamics model, incorporating suspension kinematics via a semirecursive formulation, generates simulation datasets linking real-time tuck states (pitch, roll) to mass and CG. Search algorithms leverage physics-derived truck state data to initialize ML hyperparameters, enhancing training efficiency. Validation against multibody benchmarks confirms accuracy, while robustness is demonstrated across driving scenarios and noise. By unifying data-driven ML with physics-based mechanics, this approach advances parameter estimation, bridging truck dynamics with computational intelligence for automotive design.

Transient heat transfer is a common phenomenon in the development and application of particle reinforced composites (PRCs), and temperature fluctuations often produce complex thermal stress, affecting the material’s service life. However, hybrid finite element methods (FEMs) for analyzing transient heat conduction and thermal stress in PRCs are not widely used, as most remain confined to steady-state scenarios. Based on the generalized multivariate variational splitting principle, this study presents a novel arbitrary polygonal hybrid finite element method (PT-FEM) to solve 2D transient heat conduction and heat stress in PRCs by introducing higher-order temperature and heat flux fields into the element domain. The introduction of higher-order temperature fields allows the method to quickly compute transient temperature fields with high accuracy, followed by real-time transfer of this temperature field to a hybrid stress finite element for thermal stress analysis. Compared with traditional FEM, PT-FEM allows the use of polygonal meshes with an arbitrary number of edges in the computational mesh, and the higher-order temperature field and thermal stress field no longer depend on shape functions. It shows the viability and efficiency of this method by contrasting several numerical examples with traditional FEM. It also highlights the adaptability of arbitrary polygonal elements in meshing PRCs and provides a new approach for analyzing transient heat conduction and thermal stress.

Physics-based reduced-order hemodynamic models have garnered significant interest because of their ability to capture whole-body cardiovascular fluctuations. However, coordinating the numerous interdependent parameters within these models remains a long-standing challenge, and the demand for the personalization of these models persists. We constructed a complex whole-body model of blood circulation (containing the heart, arterial trunk, and branches) and utilized genetic algorithms to automatically and efficiently coordinate the model parameters. Additionally, we introduced a “pseudo-distance” metric by updating the derivative dynamic time-warping algorithm to evaluate the similarity between the simulated waveforms and the target waveforms. After 40 rapid iterations, a complete match was achieved with the target in terms of the blood pressure and flow waveforms amplitude as well as the time domain, resulting in highly realistic waveform mimicry (i.e., the pseudo-distance approached zero). This model takes about 40 min, far less than the manual modeling that usually takes several months. These results indicate that GAs significantly improve the modeling efficiency of reduced-order models, thus lowering the user threshold.

Synchronized measurements of the velocity and temperature fields in a zero-pressure gradient turbulent boundary layer over a heated flat plate were conducted using a cold-hot dual-wire probe calibrated across various temperatures and velocities. A comparison with direct numerical simulation results demonstrated the accuracy of the cold-hot dual-wire probe in measuring velocity within flow fields characterized by significant temperature fluctuations. Additionally, the theoretical frequency response curve of the cold-wire was used to simulate the temperature fluctuation signals measured by cold-wires of different diameters and lengths. The results suggest that using a 5 µm cold-wire instead of thinner ones is a highly cost-effective choice. Finally, based on the strong correlation between the velocity and temperature fields within the thermal boundary layer, we proposed a method to correct temperature effects using the hot-wire signal alone, without cold-wire temperature measurements. Using hot-wire anemometry, the self-correction method can effectively correct the impact of temperature gradients on mean velocity measurements. However, its effectiveness in managing fluctuating temperatures is less than that of the direct measurement method using a cold-hot dual-wire setup. The velocity root-mean-square values obtained with this method have an error of less than 5% in the logarithmic region. This method is a viable alternative when a cold-wire is unavailable.

The influence of the spanwise computational domain constraint on flow statistics and heat transfer is investigated in spanwise rotating plane Poiseuille flow (RPPF) through direct numerical simulations at a fixed global friction Reynolds number of Re_τ = 180. Simulations are conducted at rotation numbers Ro_τ = 5 and 10, with the spanwise computational domain size L_z varying within {2π, π, 0.36π, 0.25π, 0.18π}. In addition to the linear behaviors of the mean streamwise velocity, Reynolds shear stress, and mean temperature observed near the pressure side, a second linear region in the mean temperature is identified near the suction wall for cases with L_z ≤ 0.36π, signifying a flow laminarization. To support the observations, the interscale transport of turbulent kinetic energy and cluster analysis are performed, compensating the former argument of contractions in roll-cell structures. The results reveal that when L_z ≤ 0.36π, the large-scale inverse energy cascades vanish, and the strength and height of cluster patterns are significantly altered, resulting in a reduction in heat transfer efficiency. Therefore, it is recommended to avoid using an excessively small spanwise computational domain when simulating flow and heat transfer problems in RPPF.

In this work, incorporating the impact of grain boundary sliding, a theoretical analytical model has been proposed for assessing fracture toughness at different temperatures of particle reinforced metal matrix composites. The model can achieve prediction of fracture toughness in varying temperature conditions by utilizing the yield strength and Young’s modulus of the metal matrix at room temperature, along with readily accessible material parameters. And the predictions have achieved good consistency with the measurements of five composites obtained from other scholars’ references. Furthermore, based on the established model, the impact of particle volume fraction and particle size on the fracture toughness of composites, as well as their evolution with temperature were studied within the applicable range of the proposed model. This research not only provides an effective and convenient tool for evaluating the fracture toughness of composites serving in different temperature environments, but also lays the foundation for the strengthening and toughening design of composites.

Escaping ejecta enhanced momentum transfer due to recoil produced by the impact, depending on the complex interaction of the projectile remove with to target. The crushing behavior of the target in the initial stage of impact is often neglected, especially in meteorite-like brittle materials. Whereas, the relationship between crack evolution and stress wave propagation is vague under ultra-high speed impact. The experiments of Al sphere impacting into granite were carried out at velocities between 1800 and 4000 m/s by a two-stage light-gas gun (DBR30), revealing distinct fragmentation characteristics on granite. As the speed increases, transitions from intact to fractured, then fragmented, exhibiting distinct failure modes under shock wave loading. Smoothed particle hydrodynamics-finite element method (SPH-FEM) simulation was employed to describe the geometrical evolution of the projectile and the propagation crack in the target. It was found that the shape of the projectile gradually changes from a cone to spherical as speeds increase. Further crack fractal dimension analysis revealed that the penetration mode transition occurs within 1500–2000 m/s. This method provides a novel framework to evaluate the ultra-high speed penetration while quantifying the penetration mode and crushing effect.

Large-eddy simulation based on a newly developed wall-adaptive local eddy-viscosity model is used to study the viscous oscillatory flow around a cylinder at high Keulegan-Carpenter (KC) and Reynolds (Re) numbers. The efficacy of the new model, which was implemented in OpenFOAM, was first checked via simulations of a standard benchmark flow, namely that around a single cylinder in oscillatory flow at Re = 600 and KC = 2.08. These simulations accurately captured the Honji vortices observed in the measurements of Honji. Thereafter, simulations were performed for values of KC in the range 1.67 ⩽ KC ⩽ 66.7, where it was found that the evolution of the Honji vortices and streaked flow is influenced by the interactions that occur between the outer oscillation flow and the inner separated boundary layer. Further, it was found that these interactions are strongly dependent on this parameter, as are all the dimensionless coefficients for both the mean and fluctuating components of the hydrodynamic forces. At low values of KC and Re, frequency doubling resonance between the vortex shedding (f_vo) and that of the oscillatory flow (f_os) was observed. A new empirical relationship between f_vo/f_os, KC and Re is proposed.

Soft magnetoelectric composites (SMCs) demonstrate tremendous application prospects in soft robots and flexible electronics due to their excellent mechanical and magnetoelectric properties. However, there is a lack of theory in describing the voltage response and revealing the voltage response mechanism of SMCs. Herein, based on Biot-Savart’s law, we propose a theoretical model of a SMC consisting of an elastomeric body and a helix coil. The voltage produced by the SMC is verified with available experiment results. For the maximum voltage and compression time, the relative errors between the theoretical and experiment results are 1.02% and 6.67%, respectively, which demonstrates the effectiveness of the proposed model. Based on the theoretical model, the magnetic characteristic and effect of structural parameters of the SMC are studied. Results show that the magnetic flux density decreases with time during the compression. The turn and diameter of helix coils and the magnetic powder content possess a significant influence on the voltage. The maximum voltage of the SMC with optimal parameters reaches 228.50 µV, which is improved by 697% compared with initial parameters. The compression ratio and compression speed possess a smaller effect on the maximum voltage. Finally, the SMC is integrated into a double-fingered gripper capable of sensing its grasping state and the properties of objects. The developed model and the obtained results provide a comprehensive understanding of SMC behavior, offering valuable insights for optimizing its performance.