Showing papers in "Computer Methods in Applied Mechanics and Engineering in 2022"

TL;DR: In this article , a metaheuristic algorithm called dwarf mongoose optimization algorithm (DMO) is proposed to solve the classical and CEC 2020 benchmark functions and 12 continuous/discrete engineering optimization problems.

TL;DR: In this paper, a bio-inspired optimization algorithm called artificial hummingbird algorithm (AHA) is proposed to solve optimization problems, which simulates the special flight skills and intelligent foraging strategies of hummingbirds in nature.

TL;DR: In this article , a bio-inspired optimization algorithm called artificial hummingbird algorithm (AHA) is proposed to solve optimization problems, which simulates the special flight skills and intelligent foraging strategies of hummingbirds in nature.

TL;DR: In this paper , a bio-inspired algorithm inspired by starlings' behaviors during their stunning murmuration named starling murmuration optimizer (SMO) is presented to solve complex and engineering optimization problems as the most appropriate application of metaheuristic algorithms.

TL;DR: In this article, a hybrid enhanced Monte Carlo simulation (HEMCS) method was proposed to estimate the failure probability with low computation cost and high computational burden. But, the authors focused on developing a novel enhanced MCS approach with an advanced machine learning method for achieving accurate approximation of failure probability using high-efficiency computations.

TL;DR: In this paper , a gradient-enhanced physics-informed neural networks (gPINNs) is proposed for improving the accuracy of PINNs, which leverage gradient information of the PDE residual and embed the gradient into the loss function.

TL;DR: In this paper , a hybrid enhanced Monte Carlo simulation (HEMCS) approach with an advanced machine learning method was proposed to achieve accurate approximation of failure probability with high-efficiency computations.

TL;DR: It is shown that the proposed adaptive sampling methods of RAD and RAR-D significantly improve the accuracy of PINNs with fewer residual points for both forward and inverse problems.

TL;DR: Wang et al. as mentioned in this paper proposed an enhanced hybrid arithmetic optimization algorithm (CSOAOA), integrated with point set strategy, optimal neighborhood learning strategy, and crisscross strategy, to solve complex engineering optimization problems.

TL;DR: In this article , a physics-informed variational formulation of DeepONet (V-DeepONet) is proposed for brittle fracture analysis, which is trained to map the initial configuration of the defect to the relevant fields of interests (e.g., damage and displacements).

TL;DR: In this article , two improved PSO algorithms are proposed to enhance the convergence rate with global optimal results during the structural reliability analysis using a hybrid PSO-based harmony search algorithm (PSO-HS) and enhanced PSO (EPSO).

TL;DR: The backward compatible PINN (bc-PINN) as mentioned in this paper was proposed to solve the Cahn Hilliard and Allen Cahn equations sequentially over successive time segments using a single neural network.

TL;DR: In this paper , the authors explore the idea of multiscale modeling with machine learning and employ DeepONet, a neural operator, as an efficient surrogate of the expensive solver.

TL;DR: In this paper, a generalised phase field-based formulation for predicting fatigue crack growth in metals is presented, where different fatigue degradation functions are considered and their influence is benchmarked against experiments.

TL;DR: In this paper , a generalised phase field-based formulation for predicting fatigue crack growth in metals is presented, which accommodates the so-called AT1, AT2 and phase field cohesive zone (PF-CZM) models.

TL;DR: In this article , a multi-objective Artificial Hummingbird Algorithm (MOAHA) was developed to solve complex multi-Objective optimization problems, including engineering design problems. But despite its superior performance, this algorithm can only solve problems with one objective.

TL;DR: In this paper , the authors investigate the performance of two neural operators and develop new practical extensions that will make them more accurate and robust and importantly more suitable for industrial-complexity applications.

TL;DR: In this paper , an efficient local adaptive Kriging approximation method with single-loop strategy (LAKAM-SLS) was proposed to enhance the computational efficiency of surrogate-based RBDO methods.

TL;DR: In this paper , a series expansion method was adopted to decouple the frequency-dependent terms from the integrand in the boundary element method, including the terms associated with the impedance boundary conditions that were introduced to model the absorption materials.

TL;DR: In this article , a discrete PINN framework based on graph convolutional network (GCN) and variational structure of PDE is proposed to solve forward and inverse partial differential equations (PDEs) in a unified manner.

TL;DR: In this paper , the authors integrate data, physics, and uncertainties by combining neural networks, physics-informed modeling, and Bayesian inference to improve the predictive potential of traditional neural network models.

TL;DR: In this paper , a topology optimization method to design self-support structures for metal additive manufacturing is proposed to avoid part failures of cracking, delamination, or warpage by constraining the process-induced residual stresses.

TL;DR: In this article , a deep neural operator architecture is proposed to learn the implicit mappings between loading conditions and the resultant displacement and/or damage fields, with the neural network serving as a surrogate for a solution operator.

TL;DR: In this paper , the authors proposed a mesh-free peridynamics (PD) discretization scheme that employs a simple collocation procedure and is truly mesh free, i.e., it does not depend on any background integration cells.

TL;DR: In this paper , the authors consider nonlinear stress and displacement fields invoked by material inhomogeneities with sharp phase interfaces and show that the domain decomposition approach is capable to accurately resolve non-linear stress, displacement and energy fields in heterogeneous microstructures obtained from real-world μCT-scans.

TL;DR: In this article , the authors proposed a 3D heat transfer model that considers contact heat transfer and thermal cracking in continuous-discontinuous media, and the model is combined with the 3D Finite Discrete Element (FDEM) for coupled thermo-mechanical calculation, which includes the temperature distribution of the system, the thermal stress caused by the temperature, and finally, the newly generated broken joint element is updated for heat transfer calculation at the next time.

TL;DR: In this article, the authors proposed a 3D heat transfer model that considers contact heat transfer and thermal cracking in continuous-discontinuous media, and the model is combined with the 3D Finite Discrete Element (FDEM) for coupled thermo-mechanical calculation, which includes the temperature distribution of the system, the thermal stress caused by the temperature, and finally, the newly generated broken joint element is updated for heat transfer calculation at the next time.

TL;DR: In this article , a thermodynamically consistent phase field model with new crack driving forces is proposed to simulate the mixed-mode fracture phenomena in rock-like materials, where the governing equations are derived using the volumetric and deviatoric strain split.

TL;DR: In this paper, a modified Newton-Raphson approach is proposed to accommodate for the local nature of the laGPR approximation when solving the global structural problem in a FE-FFT setting.

TL;DR: In this article , a new approach based on distance fields to exactly impose boundary conditions in physics-informed deep neural networks is introduced to improve the training in deep learning for partial differential equations.