Showing papers in "Computers & Structures in 2022"

TL;DR: In this article , a two-dimensional inverse differential quadrature method is proposed to approximate the solution of high-order system of differential equations, circumventing the error arising from high sensitivity to noise associated with high order numerical differentiation operations during direct approximation.

TL;DR: In this paper, a recurrent neural network (RNN) based model is developed as a surrogate to predict nonlinear plastic response under multiaxial loading, which has widespread application such as in the simulation of metal forming, large scale plasticity, and part life prediction.

TL;DR: In this paper , a damage localization strategy that efficiently exploits vibration and temperature data to account for the effects of temperature fluctuations on the structural response is proposed, where deep learning techniques are used to handle the damage localization task as a supervised classification, conditioned on temperature data.

TL;DR: In this paper , a recurrent neural network (RNN) based model is developed as a surrogate to predict nonlinear plastic response under multiaxial loading and regularization is employed to maintain non-negative plastic power density throughout the loading history.

TL;DR: In this article , an improved slime mold algorithm (ISMA) is proposed and applied to the size optimization of truss structures with natural frequency constraints, where an elitist strategy is adopted to replace the generational replacement strategy of the classical SMA.

TL;DR: In this paper , a simple and accurate coupled peridynamics and smoothed particle hydrodynamics (SPH) strategy based on virtual particles and repulsive forces is proposed to simulate fluid-structure interaction (FSI) problems with large deformation and fracturing.

TL;DR: In this paper , Atroshchenko et al. proposed to enrich the PHT-splines field approximation with a set of plane-waves propagating in different directions to capture oscillatory behaviour of the solution and achieve smaller error on coarser meshes.

TL;DR: In this article , a meta-learning-based surrogate modeling framework is presented, which consists of two phases: meta-training and a few-shot learning phase, where the meta-model represents a family of tasks and the adaptation of this model to a new task with few data points is the results of the first and second phase, respectively.

TL;DR: In this article , a surrogate neural network-based model was proposed to estimate the final lens profiles from tens of millions of function evaluations, which can provide a valuable tool for optimizing tolerance design and intelligent component matching for many similar assembly processes.

TL;DR: In this article, the optimal precast beam design was determined using a genetic algorithm using a parametric study, in which a probabilistic-based natural selection was introduced, selecting parental chromosomes for reproduction using inherited probabilities calculated by the ranking of each design among populations.

TL;DR: In this paper , the optimal precast beam design using a genetic algorithm was determined using a parametric study, and two new features for enhancing genetic algorithms were developed in the present study.

TL;DR: In this paper , an AI-based framework for the seismic retrofitting cost optimization of existing reinforced concrete (RC) frame structures is proposed, which is oriented to the minimization of retrofitting-related costs, simultaneously controlling the associated expected annual loss (EAL).

TL;DR: In this paper , a distribution-based global sensitivity analysis based on the Kullback-Leibler divergence derived directly from generalized polynomial chaos expansion (PCE) is presented.

TL;DR: In this paper , a single step optimization approach for topology, size and shape of trusses subjected to multiple static and free vibration constraints is developed, where a topology pseudo-area variable based on a penalty parameter is discretely assigned by either 10−3 or 1 which represents the absence or attendance of a truss member.

TL;DR: In this article , a semi-numerical method for determining the dynamics of micro-resonators with finite width immersed in incompressible viscous fluids is presented, which is based on the Kirchhoff plate theory and a boundary integral formulation of the Stokes equations.

TL;DR: In this article , a multimode coupled nonlinear flutter analysis method for long-span bridges with consideration of vibration-amplitude-dependent flutter derivatives is proposed to determine the real iterative frequency as well as the participation of each structural mode shape to the flutter response for a given wind speed and vibration amplitude.

TL;DR: In this paper , a unified damage model is proposed to describe the level of damage as well as the spatial distribution of damage, and two additional types of damage models are incorporated into the fail-safe design problem.

TL;DR: In this paper, a multimode coupled nonlinear flutter analysis method for long-span bridges with consideration of vibration-amplitude-dependent flutter derivatives is proposed to determine the real iterative frequency and participation of each structural mode shape to the flutter response for a given wind speed and vibration amplitude.

TL;DR: In this article , a Siamese convolutional neural network is used to learn a learnable mapping of raw vibration measurements onto a low-dimensional space, wherein damage locations can be easily identified.

TL;DR: In this paper , the authors proposed a real-time prediction method for key monitoring physical parameters (KMPPs) for early warning of fire-induced building collapse using machine learning, which can identify unknown and uncertain parameters and predict the hard-to-measure KMPPs with high accuracy and efficiency.

TL;DR: In this paper, two additional types of damage models are proposed and incorporated into the fail-safe design problem to minimize structural mass and include local stress constraints and limits on eigenfrequencies.

TL;DR: In this article , the authors adopt the topology optimization of binary structures (TOBS) method to solve structural optimization problems that consider buckling constraints and design-dependent loads, such as fluid pressure loading, a characteristic of submerged structures.

TL;DR: In this paper , a data-driven computing algorithm integrated with model reduction technique is proposed to conduct instability analysis of thin composite structures, where the weight coefficient settings are determined by the locally tangent linear material behavior of the data sets.

TL;DR: In this article, a level set-based topology optimization method for shell structures is presented to minimize both compliance and stress under a volume constraint, where trimmed quadrilateral shell meshes are generated by cutting a background quad-ilateral shell mesh with the zero-isolines of a level-set function.

TL;DR: In this article , a data-driven deep neural network (DNN) based approach is presented to accelerate finite element analysis (FE2) by using a probabilistic approach for surrogates' development.

TL;DR: In this article, the authors adopt the topology optimization of binary structures (TOBS) method to solve structural optimization problems that consider buckling constraints and design-dependent loads, such as fluid pressure loading, a characteristic of submerged structures.

TL;DR: In this article, the geometry, the pith, the knots and the local fiber orientations in timber boards were reconstructed based on X-ray computed tomography scans. But the results were limited to a single board.

TL;DR: In this article , the authors investigated the use of splitting ratio, γ , in the ρ ∞ -Bathe method to reach a higher-order accuracy in the finite element solutions of structural dynamics and heat transfer problems.

TL;DR: In this paper, variable-order shell elements that require only the shell mid-surface to be discretized, contain no rotational degree of freedom and adopt the full three-dimensional constitutive relationship are presented.

TL;DR: In this paper , a higher-order accurate, added-mass-stable fluid-structure interaction scheme centered around a split-step fluid solver is proposed to solve the coupled problem.