Explore chapters and articles related to this topic
A mesoscopic analysis of chloride diffusion with adaptive mesh refinement
Published in Nigel Powers, Dan M. Frangopol, Riadh Al-Mahaidi, Colin Caprani, Maintenance, Safety, Risk, Management and Life-Cycle Performance of Bridges, 2018
Zichao Pan, Dalei Wang, Rujin Ma, Airong Chen
To decrease the computational burden in a mesoscopic analysis of chloride transport problem, the technique of adaptive mesh refinement (AMR) can be introduced. The general philosophy of AMR is to improve the accuracy of numerical solutions by starting the analysis on a coarse initial mesh and then refining the mesh based on the refinement criteria. So far, AMR has been already successfully applied in Calculated Fluid Dynamics (CFD) (Rossi, Cotela, Lafontaine, Dadvand, & Idelsohn 2013, Antepara, Lehmkuhl, Borrell, Chiva, & Oliva 2014) and mechanical behaviors of structures such as fracture (Khoei, Moslemi, Ardakany, Barani, & Azadi 2009, Park, Paulino, Celes, & Espinha 2012, Khoei, Eghbalian, Moslemi, & Azadi 2013) and dynamic response under earthquake (Sun & Li 2016). The overall aim of AMR is to obtain a balance between accuracy and computational efficiency in numerical simulations.
Adaptive mesh modelling of localization in a cohesion-softening soil
Published in G.N. Pande, S. Pietruszczak, H.F. Schweiger, Numerical Models in Geomechanics, 2020
Adaptive mesh refinement (AMR) has an important role to play in optimising storage and run-time requirements for problems which are ‘large’; such as 3-d simulations and problems in which element size requirements vary dramatically (e.g. as in strain localisation).
A fourth-order adaptive mesh refinement algorithm for the multicomponent, reacting compressible Navier–Stokes equations
Published in Combustion Theory and Modelling, 2019
Matthew Emmett, Emmanuel Motheau, Weiqun Zhang, Michael Minion, John B. Bell
Compressible reacting flow, like many systems governed by PDEs, exhibits a range of dynamic scales in both space and time with the finest scales existing in only a small fraction of the total area of interest in the simulation. In this case, the use of a local adaptation of the computational grid can reduce the total number of spatial degrees of freedom necessary to resolve the solution compared to a uniform or static mesh. In this paper, we focus on block-structured adaptive mesh refinement (AMR) in the context of finite-volume spatial discretisations. The first block-structured AMR method for hyperbolic problems is introduced by Berger and Oliger [1]. A conservative version of this methodology for gas dynamics was developed by Berger and Colella [2] and extended to three dimensions by Bell et al. [3]. Block-structured AMR has been applied in a wide range of fields including astrophysics, combustion, magnetohydrodynamics, subsurface flow, and shock physics, and there are a number of public-domain software frameworks available for developing applications. See Dubey et al. [4] for a recent survey of AMR applications and software.
An alternative Vorticity based Adaptive Mesh Refinement (V-AMR) technique for tip vortex cavitation modelling of propellers using CFD methods
Published in Ship Technology Research, 2022
AMR is a dynamic mesh technique, and it refines or coarsens the cells in the specified regions of the computational domain according to the adaptive mesh criteria. The solution quantities are automatically interpolated to the new adapted mesh locations. One of the challenges in the AMR technique is the selection of an appropriate refinement criterion. The refinement criterion can be selected either as a scalar quantity (e.g. pressure) or as a gradient (e.g. vorticity) quantity to create the cells in the tip vortex trajectory. As stated in the study of Yvin and Muller 2016, the shape of the pressure field in the transversal direction looks like a Gaussian function; thus, it is difficult to conclude whether the location of the minimum pressure inside the vortex is in the centre or not. For this reason, it was suggested that the refinement criterion should not be chosen value of pressure itself (Yvin and Muller 2016). Hence, the vorticity based criterion was selected as the refinement criterion for the AMR application in this study. The criterion can be defined with the following Equation; where denotes the strain rate tensor and is the angular rotation rate tensor (or vorticity tensor). According to the magnitude of the criterion, the dominant parameter in the flow field can be determined. When the value of the criterion is positive, the flow field is dominated by the vorticity, whereas the strain rate dominates the flow field when the criterion value is negative (Star CCM+ 14.06).
Space-time mesh refinement method for simulating transient mixed flows
Published in Journal of Hydraulic Research, 2021
Zhonghua Yang, Zhonghao Mao, Guanghua Guan, Wei Gao
In this paper, a robust and conservative space-time adaptive mesh refinement (AMR) method is proposed under the framework of PSM. The AMR method can enhance the resolution of simulation results near strong gradients recursively without much time-consuming penalty, but also reduce computational cost by adopting coarse meshes elsewhere. To determine the flow variables at child meshes during the mesh refinement, a conservative determining method based on the shock capturing schemes is proposed, which proves to be effective in avoiding numerical oscillations. The accuracy and speed-up effect of AMR are validated in various numerical tests. The following conclusions can be drawn: The method proposed in this paper to determine flow states in child meshes during mesh refinement can avoid the occurrence of numerical oscillations under different circumstances.The AMR can precisely compute the propagation of filling-bore and represent the strong gradient at the filling-bore front, which is attributed to the fine meshes adopted near the strong gradient.The value of lmax is suggested to be 4 for transient mixed flow simulation to prevent the nested loops from becoming too time-consuming. For simulation of high speed propagation bores, such as water hammer waves, it is suggested to adopt a smaller lmax.The speed-up effect of AMR is nearly halved while simulating two filling-bores in comparison with simulating one filling-bore, because the number of fine meshes increases with the number of strong gradients in the AMR.The speed-up effect of AMR is related to the ratio of coarse meshes in the whole computational area, which makes it particularly efficient for large scale tunnels.