An adjoint-truncation error based approach for goal-oriented mesh adaptation

Ponsin, Jorge; Fraysse, François; Gómez, M.; Cordero-Gracia, M.

doi:10.1016/j.ast.2014.10.021

Cited by 7 publications

(7 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The summary table 1 shows that the same objective function accuracy can be obtained for almost an order of magnitude lower mesh size when the output-based sensor is used, even though a rather crude re-meshing methodology and the topologically inconsistent meshes for truncation error estimation were used. The topological inconsistency between grids is what distinguishes truncation error estimation methodology used in this work as compared to nested grids used by Ponsin and Fraysse [4] as well as Venditti or Fidkowski [2,3]. Table 1: Quantitative comparison of achieved objective accuracy between re-meshed grids using output-based sensor and uniformly refined regular hex meshes.…”

Section: Cube With 3d Manufactured Solutionmentioning

confidence: 99%

“…The adjointweighted truncation error known also as an output-based indicator gained popularity as an effective driver for an adaptation process -see e.g. [2,3,4]. Through the adjoint weighting, the mesh adaptation is very effectively targeted to those areas of the computational domain where the objective function is highly sensitive to mesh resolution and the local error estimate is large.…”

Section: Introductionmentioning

confidence: 99%

“…Using geometric multi-grid levels for truncation error estimation has been proposed by Fraysse and Ponsin [4]. In this approach, the truncation error is estimated on the finest grid level using a topologically consistent coarser mesh.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards an Output-Based Re-Meshing for Turbomachinery Applications

Gugala¹,

Meyer²,

Müeller³

2016

Proceedings of the VII European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS Congress 2016)

View full text Add to dashboard Cite

Abstract. A methodology for estimating the truncation error between a fine computational grid and topologically inconsistent coarse grid from a multi-grid sequence is presented. This sensor is then weighted with the adjoint solution to obtain a goal-based refinement sensor which provides local scales for a re-meshing procedure. The truncation-based sensor, as well as an output-based indicator, were implemented in the proprietary Rolls-Royce code Hydra. The output-based re-meshing refinement is first applied to the simple cube case with an inviscid flow. The truncation-based sensor and the output-based sensor are then evaluated for a turbine stator case as a first step towards the turbomachinery application of the re-meshing refinement methodology. The results are presented and discussed.

show abstract

Section: Cube With 3d Manufactured Solutionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards an Output-Based Re-Meshing for Turbomachinery Applications

Gugala¹,

Meyer²,

Müeller³

2016

Proceedings of the VII European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS Congress 2016)

View full text Add to dashboard Cite

show abstract

“…Further studies have to be done to compare these type of methods to the proposed approaches based on the truncation error. Also a combination of both techniques has been proposed for low order methods [Ponsin 2015] and needs to be investigated for high order methods in the future. The following section deals with the validation and scaling of the presented solver.…”

Section: Discussionmentioning

confidence: 99%

Adaptation Strategies for Discontinuous Galerkin Spectral Element Methods by means of Truncation Error Estimations

Kompenhans¹

View full text Add to dashboard Cite

“…Recent developments in the adjoint methodology have eliminated the limitation imposed by the number of variables, because adjoint methodology allows obtaining the gradient by solving only the direct and the adjoint problems which are, both of them, independent on the number of variables [44,47]; thus, it could be considered that the classical interest in having a reduced number of variables has vanished and that the designer can freely use more and more complex geometries defined by larger and larger sets of variables; but the reality is extremely different.…”

Section: From Artisan Skills To the The Current Industrial Techniques: Today's Challengesmentioning

confidence: 99%

Geometric manipulation of NURBS surfaces for computational meshes

García¹

View full text Add to dashboard Cite

This work tackles the problem of the geometric design optimization from an innovative point of view. The developed methodology considers directly the parameters which define the geometry as design variables. State of the art techniques use deformation functions which hamper the full process automation and pose an extra difficulty with intersecting and trimming geometries. This new technique eliminates all these obstacles First of all, I would like to express my deep gratitude to my advisers Prof. Marta Cordero and Prof. Mariola Gómez for their continuous support, patience and recommendations along all these years.I will always be in debt with Juan Ramón Pérez Garraleta, he introduced me to the NURBS world, and he showed me the magic behind the most simple solutions.I also want to address my gratitude to the many colleagues I met working in different research activities during the last years. Eusebio, Esther, Geza, Heiko and Jacques, you illustrated me how to fight for each project; you made me feel part of something bigger while working with you. Special acknowledgment is made to Carlos, Maria Chiara and Ramón, who where always ready for interesting discussions that led to a real friendship. I am specially thankful for all the good friends who always helped me to disconnect and to enjoy life.

show abstract

An adjoint-truncation error based approach for goal-oriented mesh adaptation

Cited by 7 publications

References 20 publications

Towards an Output-Based Re-Meshing for Turbomachinery Applications

Towards an Output-Based Re-Meshing for Turbomachinery Applications

Adaptation Strategies for Discontinuous Galerkin Spectral Element Methods by means of Truncation Error Estimations

Geometric manipulation of NURBS surfaces for computational meshes

Contact Info

Product

Resources

About