Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics
Efficient Robust Design Optimization (RDO) strategies coupling a parsimonious uncertainty quantification (UQ) method with a surrogate-based multi-objective genetic algorithm (SMOGA) are investigated for a test problem in computational fluid dynamics (CFD), namely the inverse robust design of an expa...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-09-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/13/10/248 |
| _version_ | 1827705110987276288 |
|---|---|
| author | Aldo Serafino Benoit Obert Paola Cinnella |
| author_facet | Aldo Serafino Benoit Obert Paola Cinnella |
| author_sort | Aldo Serafino |
| collection | DOAJ |
| description | Efficient Robust Design Optimization (RDO) strategies coupling a parsimonious uncertainty quantification (UQ) method with a surrogate-based multi-objective genetic algorithm (SMOGA) are investigated for a test problem in computational fluid dynamics (CFD), namely the inverse robust design of an expansion nozzle. The low-order statistics (mean and variance) of the stochastic cost function are computed through either a gradient-enhanced kriging (GEK) surrogate or through the less expensive, lower fidelity, first-order method of moments (MoM). Both the continuous (non-intrusive) and discrete (intrusive) adjoint methods are evaluated for computing the gradients required for GEK and MoM. In all cases, the results are assessed against a reference kriging UQ surrogate not using gradient information. Subsequently, the GEK and MoM UQ solvers are fused together to build a multi-fidelity surrogate with adaptive infill enrichment for the SMOGA optimizer. The resulting hybrid multi-fidelity SMOGA RDO strategy ensures a good tradeoff between cost and accuracy, thus representing an efficient approach for complex RDO problems. |
| first_indexed | 2024-03-10T15:55:43Z |
| format | Article |
| id | doaj.art-3a44f7fe9d134262867438ac6e14a16d |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-03-10T15:55:43Z |
| publishDate | 2020-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-3a44f7fe9d134262867438ac6e14a16d2023-11-20T15:37:58ZengMDPI AGAlgorithms1999-48932020-09-01131024810.3390/a13100248Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid DynamicsAldo Serafino0Benoit Obert1Paola Cinnella2Enertime, 1 rue du Moulin des Bruyères, 92400 Courbevoie, FranceEnertime, 1 rue du Moulin des Bruyères, 92400 Courbevoie, FranceLaboratoire Dynfluid, Arts et Métiers ParisTech, 151 blvd de l’hopital, 75013 Paris, FranceEfficient Robust Design Optimization (RDO) strategies coupling a parsimonious uncertainty quantification (UQ) method with a surrogate-based multi-objective genetic algorithm (SMOGA) are investigated for a test problem in computational fluid dynamics (CFD), namely the inverse robust design of an expansion nozzle. The low-order statistics (mean and variance) of the stochastic cost function are computed through either a gradient-enhanced kriging (GEK) surrogate or through the less expensive, lower fidelity, first-order method of moments (MoM). Both the continuous (non-intrusive) and discrete (intrusive) adjoint methods are evaluated for computing the gradients required for GEK and MoM. In all cases, the results are assessed against a reference kriging UQ surrogate not using gradient information. Subsequently, the GEK and MoM UQ solvers are fused together to build a multi-fidelity surrogate with adaptive infill enrichment for the SMOGA optimizer. The resulting hybrid multi-fidelity SMOGA RDO strategy ensures a good tradeoff between cost and accuracy, thus representing an efficient approach for complex RDO problems.https://www.mdpi.com/1999-4893/13/10/248robust design optimizationuncertainty quantificationgradient enhanced krigingmethod of momentsmulti-fidelity surrogatecontinuous adjoint |
| spellingShingle | Aldo Serafino Benoit Obert Paola Cinnella Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics Algorithms robust design optimization uncertainty quantification gradient enhanced kriging method of moments multi-fidelity surrogate continuous adjoint |
| title | Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics |
| title_full | Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics |
| title_fullStr | Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics |
| title_full_unstemmed | Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics |
| title_short | Multi-Fidelity Gradient-Based Strategy for Robust Optimization in Computational Fluid Dynamics |
| title_sort | multi fidelity gradient based strategy for robust optimization in computational fluid dynamics |
| topic | robust design optimization uncertainty quantification gradient enhanced kriging method of moments multi-fidelity surrogate continuous adjoint |
| url | https://www.mdpi.com/1999-4893/13/10/248 |
| work_keys_str_mv | AT aldoserafino multifidelitygradientbasedstrategyforrobustoptimizationincomputationalfluiddynamics AT benoitobert multifidelitygradientbasedstrategyforrobustoptimizationincomputationalfluiddynamics AT paolacinnella multifidelitygradientbasedstrategyforrobustoptimizationincomputationalfluiddynamics |