International Journal of Computational Methods

Previous studies have demonstrated the superior damage identification performance of the double-window principal component analysis (DWPCA) method over traditional PCA methods and other traditional techniques, such as wavelet and regression analysis. DWPCA uses temporal windows to discriminate structural states and spatial windows to exclude damage-insensitive responses, making it more effective for damage identification. However, determining the optimal temporal window scale and its impact on damage identification performance still remains unclear. In this study, different scales of temporal windows, including yearly, seasonal and monthly windows, are employed to obtain corresponding damage features, i.e., eigenvectors derived from DWPCA. These damage-sensitive eigenvectors from various temporal windows are then used as inputs for artificial intelligence (AI) algorithms to localize and quantify damages. In this paper two types of AI algorithms are employed: random forest (RF) and bidirectional gated recurrent unit (BiGRU). A numerical study using a benchmark model is used to evaluate the contribution of the eigenvector of each temporal scale to damage identification. The results demonstrate that the combined DWPCA eigenvectors ${\psi }_1^{\mathrm{\text{combined}}}$ from the three temporal windows effectively enhance the AI-based damage identification capability. Besides, AI algorithm with ${\psi }_1^{\mathrm{\text{combined}}}$ can have high accuracy exceeding 95% under limited training data sets and strong noise. Additionally, when DWPCA eigenvectors from monthly or seasonal windows as inputs, which is both sensitive to damages and noise, the BiGRU also achieves high accuracy of over 90% for damage identification, due to its advantages in feature extraction. These findings suggest that the proposed approach has significant potential for real-life structural health monitoring applications.

Particle generation is important for particle-based methods. The reasonable initial particle distribution has a profound impact on the computational accuracy and the stability of particle-based methods. We propose a particle generation method that generates the body-fitted and uniform particles fast. This method first generates initial body-fitted and locally uniformly distributed particles on the geometric facets, following the spatial partitioning, particle hashing, particle sorting, and particle merging. The surface particles become body-fitted and uniform. In addition, we combine several other generation methods of boundary-level particles and inner particles to provide a unified framework for the pre-processing process of the smoothed particle hydrodynamics (SPH) method. Some particle generation examples of complex geometric models demonstrate the feasibility and effectiveness of this method. The proposed method can generate large-scale particles of complex geometric models, improve the handling capacity of pre-processing techniques, and keep up with the scale of solvers.

In this paper, the authors discuss the development of six low-order triangular elements for the linear static analysis of thick and thin shell structures. The elements combine two triangular plane stress elements with three triangular thick plate bending elements to produce three-node flat shell elements with 18 degrees of freedom. The two plane stress components are based on the free formulation and the template formulation, while the three plate bending components are based on the free formulation, the deviatoric strain formulation and the hybrid-Trefftz formulation. The theoretical backgrounds and formulations of all elements are briefly discussed to provide a concise summary, while further details are given in the quoted references. The elements are used to analyze five standard shell benchmark examples, and the results obtained are compared to previous work in this area. The elements provide accurate and robust results for the benchmarks, with only small differences for the results as the meshes are refined. While all the proposed elements can be recommended for general shell analysis, the combination of the template plane stress element and the hybrid-Trefftz plate bending element generally provides the best results.

The modeling of industrial processes that involve phase change (e.g., casting, welding or additive manufacturing) is challenging due to their multiscale and multiphysics nature. The coupling of the fluid and solid mechanics is difficult due to the unknown and evolving fluid–solid interface. To deal with the difficulty of the coupling, a unified approach has been developed where fluid and solid regions are represented in the same computational domain and solved by a single solver. The interaction of fluid and solid regions is therefore automatically captured and the material can locally undergo phase transition. This allows to capture the flow in the melt and the thermal stresses in the solid. The solid material model is currently limited to linear elasticity, but it opens the path to more complex material models with plasticity modeling, which already exist in the literature. The proposed method is based on the Particle Finite Element Method (PFEM), which has been shown to accurately handle both fluid and solid mechanics. In this work, we present the key aspects of this novel unified formulation and the treatment of the fluid–solid interface in the PFEM context. The methodology is presented and verified using a set of tests. A laser spot welding example test case demonstrates the potential of combining the unified formulation with the interface treatment and the phase change capabilities and is used for the validation of the present technique.

The paper proposes a finite element method with high precision and convergence for solving 2D transient heat conduction problems. The temperature gradient with respect to time is approximated using the Houbolt method, a third-order difference scheme, which improves the accuracy and convergence rate and permits the use of large time steps for long-term prediction. The proposed finite element method redistributes the thermal mass matrix to balance it with the thermal stiffness matrix, further enhancing the precision of solutions. Besides, the effective condition number is introduced to determine the position of optimal flexible integration points, ensuring that the error of the solutions is minimal. Finally, intensive transient heat conduction problems are analyzed using the present method to demonstrate good performance in precision and convergence. Compared to FEM using the backward finite difference method, the proposed method achieves more accurate and super convergent temperature solutions. In comparison to FEM with the forward finite difference method, it can use larger time steps while maintaining the same level of accuracy.

In this paper, a new approach named simplified boundary knot method with ghost points (SBKM-G) is proposed to improve the performance of SBKM in solving high-order harmonic-type partial differential equations (PDEs). In contrast to the traditional SBKM where the source points are coincident with the collocation nodes on the boundary, the SBKM-G removes the source/ghost nodes to an arbitrary area like a circle (sphere) which covers the original problem domain in 2D (3D). This modification leads to higher accuracy without sacrificing simplicity and efficiency. In addition, the effect of free parameters in SBKM-G has also been studied. Numerical examples in 2D and 3D show the superiority of SBKM-G to SBKM and the performance of SBKM-G in solving this type of high-order PDEs is highlighted.

The flow and heat transfer characteristics in a three-dimensional cavity filled with a conducting fluid are investigated in this study in the presence of a constant magnetic field. The three-dimensional Navier–Stokes equations and energy equation are solved directly using a self-developed method, SCM–ACM, which combines the spectral collocation method (SCM) with high-precision and exponential convergence, and the artificial compressibility method (ACM) with easy implementation and good numerical stability. In this paper, we examine the effects of Hartmann numbers (Ha) ranging from 0 to 100, magnetic field directions, and Grashof numbers (Gr) ranging from $1\times 10^4$ to $1\times 10^6$ on the structure of the flow and temperature fields, with a Prandtl number (Pr) of 0.71. The results show the Grashof and Hartmann numbers have a significant impact on the flow and temperature distribution in the middle of the cavity, but little effect on that near the walls. As the Grashof number increases, a stable thermal stratification is formed at the center of the cube, and thermal boundary layers are formed near the horizontal walls. The increase in Grashof number enhances the heat transfer rate and increases the temperature difference between the upper hot fluid and the lower cold fluid in the cube. Furthermore, the increase in Grashof number enhances the convective intensity between the isothermal walls, leading to the formation of more vortices, which move toward the corners due to the combined action of centrifugal force and inertia. The Hartmann number has stabilizing effects on the flow and weakens the heat transfer. With a higher Grashof number the magnetic effects become more notable. When $\mathrm{\text{Ha}}>50$, the magnetic effects are no longer significant. The magnetic field $B_Z$ perpendicular to the main circulation plane exhibits the strongest stability for the flow.

The ongoing expansion of high-speed railway systems in seismically active regions necessitates improved methods for evaluating the seismic resilience of the railway infrastructure. This study aimed to develop and establish a seismic performance indicator for coupled vehicle–track–tunnel systems to facilitate the assessment of safety under earthquake conditions. A refined finite element model was created, capturing the intricate dynamical interactions between a high-speed train, the track, and the tunnel, with a particular focus on the heterogeneity of the soil stratum. The model, synthesized through a joint simulation using HyperMesh and ABAQUS, was validated with theoretical solutions and experimental data, enhancing its predictive reliability of the system’s seismic response. The research resulted in the innovation of a two-level seismic performance indicator, offering a quantitative tool for evaluating the integrity and safety of a high-speed railway tunnel during seismic events. Although the performance indicator has shown promise in initial validations, it is emphasized that continued verification and refinement are paramount for ensuring its adaptability across various seismic scenarios. The contribution of this research lies in providing a novel, rigorous approach for the seismic design evaluation, which is critical for advancing the safety standards of high-speed railway infrastructures.

Interfaces between polymers and ceramics or metals have wide applications in mechanical, electronic and chemical engineering. Loading speed and temperature have certain influence on interface properties, which affect their application in the service environment. In this paper, the interface intrinsic strength of polypropylene (PP)/silicon oxide (SiO${}_2)$ and polyethylene (PE)/aluminum are studied based on molecular dynamic (MD) tensile method at low temperature and loading speed first. Then changes of interface properties are studied by increasing loading speed and temperature, respectively. The results indicate that interface strength increases or decreases with increasing loading speed or temperature. The interface damage mechanism related to the evolution of free volume is revealed.

Pictograms convey meaning through graphic symbols; however, static pictograms have limitations, prompting research into animated pictograms (motion pictograms). While motion pictograms offer intuitive comprehension, their limited adoption results from higher production costs than their static counterparts. A potential solution to enhance motion pictogram adoption involves automating the generation process from still images using artificial intelligence. In this study, a pretrained model identified key constraints and challenges. Experimental results underscore the effectiveness of motion pictograms, emphasizing their advantages and paving the way for broader adoption and improved accessibility.

Mountain tunnels built near faults often suffer from significant permanent deformation and structural dislocation during seismic activity. In this paper, we present a rock-fault-tunnel geological model with a transition area between the hanging wall and the foot wall which allows the free slippery growth inside the area. A time-sequenced load based on design code and fault activity is conducted in this model to simulate dynamic seismic input after fault dislocation. In our case, a reverse fault with a tunnel cross has been created with this method. A 30cm fault dislocation is simulated by putting the displacement boundary of the hanging wall with a compression vector and the seismic wave is input from the bottom boundary as acceleration waveform adjusted to 0.4g. The model simulates the uplift of the hanging wall and the growth of the slip surface, and reveals the extension mechanism of the triangular shear zone of shear rupture of the surrounding rock due to the extrusion of the reverse fault during the propagation of the reverse fault. The seismic wave with a three-way acceleration was input after the dislocation process. The simulation indicates that with the gradual uplift of the hanging wall, the rock body of the fracture zone shows a more significant large deformation flow trend and a more significant horizontal slip flow. Under reverse fault thrust, the width of the shear effect influence zone is around 300m. A decreasing trend of accumulated strain can be found at the interface due to acceleration input. Dislocation-seismic time-sequence loading may underestimate its damage effects.

In order to better understand, describe and design the interfacial properties of pipe joints for pipeline structures in actual projects, this paper theoretically investigated the mechanical properties of adhesively bonded pipeline joint interface under simultaneous actions of axial tensile loading and temperature changes based on the bilinear rigid-softening constitutive model. Analytical solutions for interface relative slips, shear stresses and load-displacement relationship, as well as expressions for ultimate loads and effective bonding lengths, were derived. Different failure processes including softening or debonding stages caused by the bonding lengths were discussed, demonstrating the full-ranged interfacial failure of pipeline joints and explaining their stress transfer mechanism, interface crack propagation and ductile behavior. The finite element analysis for certain cases was also checked to coincide with the solutions. This helps establish a theoretical analysis basis for the application of pipeline joints in actual environments with loading and temperature differences.

In the processes of seismic design of underground structures, selecting a reasonable input ground motion is very important, which can cause severe damage to underground structures. To quantitatively evaluate the seismic damage potential of ground motions on multi-storey underground structures and solve the problem that single intensity measures are inadequate in accurately indicating the seismic damage potential of ground motions, this paper taking the input ground motions in the seismic design of underground structures as the research object, and constructing some composite intensity measures that can effectively characterize the damage potential of input ground motion. Firstly, considering that the underground structures in different characteristic period-type sites will exhibit different seismic responses under the same excitation, the soil–structure system is divided into four-period bands. Then, four representative periods are selected from four-period bands respectively, and the corresponding four soil–structure system numerical models are established. Subsequently, 40 ground motions are selected for elastoplastic numerical analysis, and the results of the numerical analysis were used as sample data to construct composite intensity measures corresponding to soil–structure systems in each period band using the partial least squares regression method. Finally, 100 additional ground motions were used to verify the correlation between the composite intensity measures and the seismic damage of underground structures. The results show that the correlation coefficient between the composite intensity measures and the seismic damage of multi-storey underground structures is better than those of commonly used single intensity measures.