Finite Elements in Analysis and Design最新文献

To obtain the displacement field of stiffened panel structures is very important for the online monitoring of aircraft or aerospace vehicles, etc. New inverse beam-shell elements are proposed in this study for the full-field displacement reconstruction of stiffened panels via strain measured by shell parts and rib parts simultaneously. The shell and rib parts in the stiffened panel are modeled by inverse shell and beam elements respectively constructed by Mindlin's plate theory and Timoshenko beam theory. To avoid the shear locking, a new inverse beam element with a virtual middle node is introduced. Constraints between the inverse shell and beam elements are given to guarantee the consistency of deformation and two typical inverse beam-shell elements are proposed. A sub-area division scheme is introduced which enables the proposed inverse elements for reconstructing the displacement field of 3D structures composed of multiple stiffened panels. Two numerical examples including a cantilever stiffened panel and a two-edge clamped 3D stiffened panel are given to demonstrate the effectiveness of the newly proposed inverse beam-shell element and the sub-area division scheme. An element-selection scheme for the arrangement of strain gauges is also proposed to reduce the measurement data used. Results show the new inverse beam-shell elements can reconstruct displacement fields accurately and the sub-area division scheme introduced guarantees the accuracy of the reconstructed displacement fields of 3D panels even when a relatively small number of strain gauges are used.

Conjugate heat transfer (CHT) problem of flow around a fixed cylinder is examined by using a high-order method which is based on the hybridizable discontinuous Galerkin (HDG) method. The present numerical method based on HDG discretization produces a system of equations in which the energy equation of fluid is coupled with that of solid while the continuity of heat-flux at the fluid-solid interface is automatically satisfied. We Investigate the effect of the conductivity ratio on the temperature distribution inside the cylinder and more importantly, the constraint of heat-flux continuity at the fluid-solid interface. The present high-order solutions are compared with low-order solutions by finite volume method of ANSYS, especially in terms of the constraint of heat-flux continuity at the interface. We show that the present high-order method provides accurate solutions and satisfies the constraint of heat-flux continuity better than ANSYS even with the use of a coarse grid. Furthermore, we have derived a numerical correlation between the Nusselt and the Reynolds number by using the fact that the surface temperature of the cylinder is nearly constant when conductivity ratio is larger than order of hundred. The proposed numerical correlation was found to be close to that from the exiting experiment.

This study presents a multiobjective optimization framework that integrates Artificial Neural Network (ANN) and Non-dominated Sorting Genetic Algorithm-II (NSGA-II) for the optimization of rolling process parameters of tube-to-tubesheet joint (TTT-joint). During the rolling process, both beneficial contact pressure and detrimental tensile residual stress are generated within the joint. The primary objective of this framework is to minimize the tensile residual stress while maximizing the contact pressure in the TTT-joint. To achieve this, a backpropagation ANN model is trained to rapidly estimate the residual stress and contact pressure for various sets of rolling process parameters. For training purposes, a series of nonlinear elastoplastic finite element (FE) simulations are performed to generate the input database for the neural network. A detailed parametric study is performed based on the axisymmetric FE model of the TTT-joint. The trained neural network is then incorporated into the NSGA-II optimization algorithm to find the fitness function and optimized process parameters. The contact pressure and residual stress predicted by the proposed ANN-NSGA-II framework are validated by finite element analysis (FEA) using the optimized parameters. The present analysis established that the proposed methodology can be applied in practical engineering problems to obtain the process parameters that yield the maximum contact pressure and minimum tensile residual stress in the TTT-joint.

When the finite element method (FEM) is adopted for studying strain localization problems, the mesh dependence phenomenon often ensues. The occurrence of mesh dependency will reduce the reliability of FEM simulations, so it is still worth studying. Herein, a constitutive model with decent mesh stability named the multiscale Cosserat (MC) model which contains higher-order rotation variables based on the conventional Cosserat (CC) theory, was introduced. The theory derivation indicates that the MC model has an extra internal length scale vector D_q that can consider the microscopic geometrical characteristics of the simulated material and can easily regress to the conventional Cosserat model when D_q = 0. After revisiting the mesh dependence problem through numerical simulations of plane strain compression tests, the mechanisms and advantages of the CC and MC models in solving the mesh dependence problem were discussed. The analysis demonstrated that the CC theory can alleviate the mesh dependence problem but cannot eliminate it; when the divergence of the computation occurs, due to a stricter accuracy requirement for convergence, the computation result of the MC model tends to stabilize along with the refinement of the elements. The mesh advantage of the MC model is influenced by both the length scales l and D_q. This study can provide new insight into understanding the mesh dependence problem, and the MC model introduced here is a potential model for comprehensively eliminating the influence of mesh dependence problems.

In this study, the vibration stability (i.e., static divergence) and critical velocity of fluid-conveying, ring-stiffened, truncated conical shells are investigated under various boundary conditions. The shell is characterized using Sanders’ theory, while the fluid is modeled using a velocity potential approach with the impermeability condition at the fluid-shell interface. Using linear superposition, the natural frequencies corresponding to each flow velocity are determined by satisfying the dynamic characteristic equation and boundary conditions. Critical velocities are identified where the natural frequencies vanish, indicating static divergence. Parametric studies are conducted to investigate the effect of ring stiffeners on the critical velocities with respect to the semi-cone angle, number of rings, and boundary conditions. The proposed model is validated through comparison with published data. It is found that the rings significantly affect the stability of the cone under different boundary conditions. Instability in stiffened shells occurs at higher critical fluid velocities than in unstiffened shells across all boundary conditions. An increase in the vertex angle leads to a decrease in critical flow discharge.

Rubber components are widely spread in engineering due to their mechanical properties such as high strength, elongation, and dissipation characteristics. Modeling rubber behavior poses challenges because of its complex visco-elastic properties and various nonlinear effects. As high fidelity simulations become increasingly challenging, reduction techniques such as subspace projection and hyper-reduction have emerged, which seek to achieve efficient use of complex models while reducing computational demands.

This article presents a Python-Abaqus co-simulation framework to perform the Energy Conserving Sampling and Weighting (ECSW) hyperreduction on nonlinear finite element hyper-viscoelastic models. A novel approach based on incremental elementary work is formulated to optimize the element selection in ECSW in the attempt of exploiting the rate dependent material characteristics. The successful implementation of the co-simulation framework underscores the beneficial use of commercial code capabilities in the development of nonlinear reduction algorithms. A numerical cantilever beam subjected to dynamic loading is employed to explore the potential of the newly proposed ECSW variant.

The engineering design of metamaterials with selected acoustic properties necessitates adequate prediction of the elastic wave propagation across various domains and specific frequency ranges. This study proposes a systematic approach centered on the finite element characterization of the three-dimensional Green’s function for a representative volume element. The inherent characteristics of broadband waves and singular impulses contribute to notable challenges related to accuracy and high-frequency oscillations, and thus the emphasis is set on providing an exhaustive analysis for this numerical characterization scheme. The study focuses on the broadband wave dispersion and requisite considerations for numerical damping, and evaluates the impact of dissipation and space–time discretization schemes for optimal performance. In contrast to conventional methods that employ a plane wave, the proposed approach does not need extra assumptions on the enforcement of boundary conditions and can effectively consider the influences of length scale from the material configurations. A quasi-equiaxed polycrystalline ice microstructure is utilized as an application example for homogenizing heterogeneous materials, in line with advancements in cryo-ultrasonic testing techniques.

This study introduces a biomimetic ceramic composite armor system, composed of multilayered biomimetic ceramic tiles and fiber back-plates. The ballistic performance of the composite armor against T12A steel projectiles was investigated through experimental and numerical simulation studies. The experimental findings indicate that, while the biomimetic ceramic structure demonstrates weaker ballistic resistance compared to a monolithic ceramic of equal thickness, it effectively inhibits crack propagation, thereby enabling it to withstand the impact of multiple projectiles. Additionally, the interfacial effects within the layers of the biomimetic ceramic structure create a more chaotic stress field inside the T12A steel projectile, resulting in a higher degree of fragmentation of the projectile compared to penetration through monolithic ceramic. A three-dimensional numerical model was established to analyze the impact of projectile velocity and impact points on the ballistic performance of the biomimetic ceramic composite structure. Simulation results reveal that as the initial velocity of the projectile increases, the energy absorption efficiency of the biomimetic ceramic structure improves, whereas the energy absorption efficiency of the UHMWPE laminated board decreases. This phenomenon is associated with the failure mechanism of the UHMWPE laminated board transitioning from tensile failure to shear failure. Moreover, when the impact point is at the corner of the ceramic tile, the residual projectile head is sharper, and the remaining velocity of the projectile after penetrating the biomimetic ceramic composite structure is higher.

Pseudo-unidirectional fiber networks are used in a variety of applications, such as woven fabrics and needling. A method for generating pseudo-unidirectional fiber networks by extruding linear portions of fibers is described here, and consists of two steps: $\left(i\right)$ Initially, a deliberately disorganized pseudo-unidirectional fiber network was generated geometrically from a stochastic algorithm according to the fiber volume ratio and the distribution law of the angles formed between the portions. $\left(ii\right)$ Then, the fibers were flattened and mechanically deformed using a finite element calculation until restoring the pseudo-unidirectional fiber network geometry, having stored elastic energy. Mechanical inter-fiber contact interactions were finally activated in a relaxation step to obtain a disorganized network in mechanical equilibrium. Angular deviation and migration criteria were defined to geometrically characterize the network disorder before and after mechanical rearrangement, and were shown to correlate with the algorithm’s input parameters. Finally, the generated networks were mechanically characterized using a needle-punch, quantifying the transfer fraction. The mechanisms by which the needle carried and broke the fibers are discussed, and simulations demonstrate the influence of initial network disorder on fiber transfer. The particular case of needling involves transferring fibers present in a 2D web in its transverse direction in order to increase the out-of-plane stiffness of the final product. In this case, entanglement seemed to play a decisive role, as it favored fiber transfer.

The behavior of typical corner supported bare and bonded poly-film hardened concrete plates are investigated experimentally using an Instron impact testing machine and evaluated numerically using two well-known phenomenological models of concrete. Before use, the models are critically evaluated and necessary modifications are incorporated. After validation with experimental data the better of the two models was selected and a series of simulations on the impact resistance of bare as well as the ones with bonded poly-film reinforcement to back face only as well as to both faces of the plate are undertaken with increasing magnitudes of impact energy to study the performance under progressively severe impact energy to evaluate the relative effectiveness of different options of the hardening scheme. Based on this study, the effectiveness of the proposed bonded poly-film based hardening scheme is established.