Rapid aerodynamic shape optimization under uncertainty using a stochastic gradient approach View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2022-06-30

AUTHORS

Lluís Jofre, Alireza Doostan

ABSTRACT

A common approach in aerodynamic design is to optimize a performance function—provided some constraints—defined by a choice of an aerodynamic model at nominal operating conditions. Practical experience indicates that such a deterministic approach may result in considerably sub-optimal designs when the adopted aerodynamic model does not lead to accurate predictions, or when the actual operating conditions differ from those considered in the design. One approach to address this shortcoming is to consider an average or robust design, wherein the statistical moments of the performance function, given the uncertainty in the operating conditions and the aerodynamic model, is optimized. However, when the number of uncertain inputs is large or the performance function exhibits significant variability, an accurate evaluation of these moments may require a large number of function evaluations at each optimization iteration, rendering the problem significantly expensive. To tackle this difficulty, we consider a variant of the stochastic gradient descent method where in each iteration, a stochastic approximation of the objective, constraints, and their gradients is generated. This is done via a small number of forward/adjoint solutions corresponding to random selections of the uncertainties. The methodology is applied to the robust optimization of the NACA-0012 airfoil subject to operating condition and turbulence model uncertainty. With a cost that is only a small factor larger than that of the deterministic methodology, the stochastic gradient approach significantly improves the performance of the aerodynamic design for a wide range of operating conditions and turbulence models. More... »

PAGES

196

References to SciGraph publications

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s00158-022-03293-y

DOI

http://dx.doi.org/10.1007/s00158-022-03293-y

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1149109951


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0102", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Applied Mathematics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0103", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Numerical and Computational Mathematics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Department of Fluid Mechanics, Universitat Polit\u00e8cnica de Catalunya \u00b7 BarcelonaTech (UPC), 08019, Barcelona, Spain", 
          "id": "http://www.grid.ac/institutes/grid.6835.8", 
          "name": [
            "Department of Fluid Mechanics, Universitat Polit\u00e8cnica de Catalunya \u00b7 BarcelonaTech (UPC), 08019, Barcelona, Spain"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Jofre", 
        "givenName": "Llu\u00eds", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Smead Department of Aerospace Engineering Sciences, University of Colorado, 80309, Boulder, CO, USA", 
          "id": "http://www.grid.ac/institutes/grid.266190.a", 
          "name": [
            "Smead Department of Aerospace Engineering Sciences, University of Colorado, 80309, Boulder, CO, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Doostan", 
        "givenName": "Alireza", 
        "id": "sg:person.01027332134.75", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01027332134.75"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/s00466-020-01870-w", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1129848052", 
          "https://doi.org/10.1007/s00466-020-01870-w"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s00158-020-02599-z", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1131199223", 
          "https://doi.org/10.1007/s00158-020-02599-z"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s00707-017-2049-3", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1092654501", 
          "https://doi.org/10.1007/s00707-017-2049-3"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1023/a:1023283917997", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1032929633", 
          "https://doi.org/10.1023/a:1023283917997"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s11081-006-9974-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1045827911", 
          "https://doi.org/10.1007/s11081-006-9974-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s00158-012-0791-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1041893884", 
          "https://doi.org/10.1007/s00158-012-0791-7"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2022-06-30", 
    "datePublishedReg": "2022-06-30", 
    "description": "A common approach in aerodynamic design is to optimize a performance function\u2014provided some constraints\u2014defined by a choice of an aerodynamic model at nominal operating conditions. Practical experience indicates that such a deterministic approach may result in considerably sub-optimal designs when the adopted aerodynamic model does not lead to accurate predictions, or when the actual operating conditions differ from those considered in the design. One approach to address this shortcoming is to consider an average or robust design, wherein the statistical moments of the performance function, given the uncertainty in the operating conditions and the aerodynamic model, is optimized. However, when the number of uncertain inputs is large or the performance function exhibits significant variability, an accurate evaluation of these moments may require a large number of function evaluations at each optimization iteration, rendering the problem significantly expensive. To tackle this difficulty, we consider a variant of the stochastic gradient descent method where in each iteration, a stochastic approximation of the objective, constraints, and their gradients is generated. This is done via a small number of forward/adjoint solutions corresponding to random selections of the uncertainties. The methodology is applied to the robust optimization of the NACA-0012 airfoil subject to operating condition and turbulence model uncertainty. With a cost that is only a small factor larger than that of the deterministic methodology, the stochastic gradient approach significantly improves the performance of the aerodynamic design for a wide range of operating conditions and turbulence models.", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s00158-022-03293-y", 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1050630", 
        "issn": [
          "1615-147X", 
          "1615-1488"
        ], 
        "name": "Structural and Multidisciplinary Optimization", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "7", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "65"
      }
    ], 
    "keywords": [
      "stochastic gradient approach", 
      "aerodynamic model", 
      "operating conditions", 
      "turbulence model uncertainty", 
      "performance function", 
      "aerodynamic shape optimization", 
      "aerodynamic design", 
      "stochastic gradient descent method", 
      "gradient approach", 
      "stochastic approximation", 
      "gradient descent method", 
      "optimization iterations", 
      "robust optimization", 
      "statistical moments", 
      "adjoint solution", 
      "shape optimization", 
      "actual operating conditions", 
      "nominal operating conditions", 
      "sub-optimal design", 
      "model uncertainty", 
      "uncertain inputs", 
      "descent method", 
      "function evaluations", 
      "deterministic approach", 
      "robust design", 
      "turbulence model", 
      "deterministic methodology", 
      "iteration", 
      "small factor", 
      "uncertainty", 
      "optimization", 
      "accurate prediction", 
      "moment", 
      "approximation", 
      "common approach", 
      "design", 
      "model", 
      "random selection", 
      "large number", 
      "airfoil", 
      "accurate evaluation", 
      "approach", 
      "performance", 
      "small number", 
      "constraints", 
      "wide range", 
      "NACA", 
      "conditions", 
      "methodology", 
      "problem", 
      "solution", 
      "function", 
      "number", 
      "prediction", 
      "input", 
      "gradient", 
      "cost", 
      "practical experience", 
      "range", 
      "method", 
      "shortcomings", 
      "evaluation", 
      "choice", 
      "selection", 
      "difficulties", 
      "significant variability", 
      "variants", 
      "objective", 
      "variability", 
      "factors", 
      "experience"
    ], 
    "name": "Rapid aerodynamic shape optimization under uncertainty using a stochastic gradient approach", 
    "pagination": "196", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1149109951"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s00158-022-03293-y"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s00158-022-03293-y", 
      "https://app.dimensions.ai/details/publication/pub.1149109951"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2022-10-01T06:50", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221001/entities/gbq_results/article/article_927.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s00158-022-03293-y"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/s00158-022-03293-y'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s00158-022-03293-y'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s00158-022-03293-y'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s00158-022-03293-y'


 

This table displays all metadata directly associated to this object as RDF triples.

169 TRIPLES      21 PREDICATES      103 URIs      87 LITERALS      6 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s00158-022-03293-y schema:about anzsrc-for:01
2 anzsrc-for:0102
3 anzsrc-for:0103
4 anzsrc-for:0104
5 schema:author Nc9be47c1736a4581a589a4a5798122fa
6 schema:citation sg:pub.10.1007/s00158-012-0791-7
7 sg:pub.10.1007/s00158-020-02599-z
8 sg:pub.10.1007/s00466-020-01870-w
9 sg:pub.10.1007/s00707-017-2049-3
10 sg:pub.10.1007/s11081-006-9974-7
11 sg:pub.10.1023/a:1023283917997
12 schema:datePublished 2022-06-30
13 schema:datePublishedReg 2022-06-30
14 schema:description A common approach in aerodynamic design is to optimize a performance function—provided some constraints—defined by a choice of an aerodynamic model at nominal operating conditions. Practical experience indicates that such a deterministic approach may result in considerably sub-optimal designs when the adopted aerodynamic model does not lead to accurate predictions, or when the actual operating conditions differ from those considered in the design. One approach to address this shortcoming is to consider an average or robust design, wherein the statistical moments of the performance function, given the uncertainty in the operating conditions and the aerodynamic model, is optimized. However, when the number of uncertain inputs is large or the performance function exhibits significant variability, an accurate evaluation of these moments may require a large number of function evaluations at each optimization iteration, rendering the problem significantly expensive. To tackle this difficulty, we consider a variant of the stochastic gradient descent method where in each iteration, a stochastic approximation of the objective, constraints, and their gradients is generated. This is done via a small number of forward/adjoint solutions corresponding to random selections of the uncertainties. The methodology is applied to the robust optimization of the NACA-0012 airfoil subject to operating condition and turbulence model uncertainty. With a cost that is only a small factor larger than that of the deterministic methodology, the stochastic gradient approach significantly improves the performance of the aerodynamic design for a wide range of operating conditions and turbulence models.
15 schema:genre article
16 schema:isAccessibleForFree true
17 schema:isPartOf N6ef6549009384251bcc2a45e17fcd209
18 Nf91b9775ebca4463afda114bd43fb98a
19 sg:journal.1050630
20 schema:keywords NACA
21 accurate evaluation
22 accurate prediction
23 actual operating conditions
24 adjoint solution
25 aerodynamic design
26 aerodynamic model
27 aerodynamic shape optimization
28 airfoil
29 approach
30 approximation
31 choice
32 common approach
33 conditions
34 constraints
35 cost
36 descent method
37 design
38 deterministic approach
39 deterministic methodology
40 difficulties
41 evaluation
42 experience
43 factors
44 function
45 function evaluations
46 gradient
47 gradient approach
48 gradient descent method
49 input
50 iteration
51 large number
52 method
53 methodology
54 model
55 model uncertainty
56 moment
57 nominal operating conditions
58 number
59 objective
60 operating conditions
61 optimization
62 optimization iterations
63 performance
64 performance function
65 practical experience
66 prediction
67 problem
68 random selection
69 range
70 robust design
71 robust optimization
72 selection
73 shape optimization
74 shortcomings
75 significant variability
76 small factor
77 small number
78 solution
79 statistical moments
80 stochastic approximation
81 stochastic gradient approach
82 stochastic gradient descent method
83 sub-optimal design
84 turbulence model
85 turbulence model uncertainty
86 uncertain inputs
87 uncertainty
88 variability
89 variants
90 wide range
91 schema:name Rapid aerodynamic shape optimization under uncertainty using a stochastic gradient approach
92 schema:pagination 196
93 schema:productId N5a0a1f044a2d4a318c27d340949b55d0
94 N71536cfcb48f47fb80fc8c53b3db21e6
95 schema:sameAs https://app.dimensions.ai/details/publication/pub.1149109951
96 https://doi.org/10.1007/s00158-022-03293-y
97 schema:sdDatePublished 2022-10-01T06:50
98 schema:sdLicense https://scigraph.springernature.com/explorer/license/
99 schema:sdPublisher N158eee7443ff4d0587654c2592e265e1
100 schema:url https://doi.org/10.1007/s00158-022-03293-y
101 sgo:license sg:explorer/license/
102 sgo:sdDataset articles
103 rdf:type schema:ScholarlyArticle
104 N158eee7443ff4d0587654c2592e265e1 schema:name Springer Nature - SN SciGraph project
105 rdf:type schema:Organization
106 N4345be1206624529b37c2b134fde0b1a schema:affiliation grid-institutes:grid.6835.8
107 schema:familyName Jofre
108 schema:givenName Lluís
109 rdf:type schema:Person
110 N5a0a1f044a2d4a318c27d340949b55d0 schema:name dimensions_id
111 schema:value pub.1149109951
112 rdf:type schema:PropertyValue
113 N6ef6549009384251bcc2a45e17fcd209 schema:volumeNumber 65
114 rdf:type schema:PublicationVolume
115 N71536cfcb48f47fb80fc8c53b3db21e6 schema:name doi
116 schema:value 10.1007/s00158-022-03293-y
117 rdf:type schema:PropertyValue
118 Nc45ee312f08840758c6d79e3a045a661 rdf:first sg:person.01027332134.75
119 rdf:rest rdf:nil
120 Nc9be47c1736a4581a589a4a5798122fa rdf:first N4345be1206624529b37c2b134fde0b1a
121 rdf:rest Nc45ee312f08840758c6d79e3a045a661
122 Nf91b9775ebca4463afda114bd43fb98a schema:issueNumber 7
123 rdf:type schema:PublicationIssue
124 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
125 schema:name Mathematical Sciences
126 rdf:type schema:DefinedTerm
127 anzsrc-for:0102 schema:inDefinedTermSet anzsrc-for:
128 schema:name Applied Mathematics
129 rdf:type schema:DefinedTerm
130 anzsrc-for:0103 schema:inDefinedTermSet anzsrc-for:
131 schema:name Numerical and Computational Mathematics
132 rdf:type schema:DefinedTerm
133 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
134 schema:name Statistics
135 rdf:type schema:DefinedTerm
136 sg:journal.1050630 schema:issn 1615-147X
137 1615-1488
138 schema:name Structural and Multidisciplinary Optimization
139 schema:publisher Springer Nature
140 rdf:type schema:Periodical
141 sg:person.01027332134.75 schema:affiliation grid-institutes:grid.266190.a
142 schema:familyName Doostan
143 schema:givenName Alireza
144 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01027332134.75
145 rdf:type schema:Person
146 sg:pub.10.1007/s00158-012-0791-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1041893884
147 https://doi.org/10.1007/s00158-012-0791-7
148 rdf:type schema:CreativeWork
149 sg:pub.10.1007/s00158-020-02599-z schema:sameAs https://app.dimensions.ai/details/publication/pub.1131199223
150 https://doi.org/10.1007/s00158-020-02599-z
151 rdf:type schema:CreativeWork
152 sg:pub.10.1007/s00466-020-01870-w schema:sameAs https://app.dimensions.ai/details/publication/pub.1129848052
153 https://doi.org/10.1007/s00466-020-01870-w
154 rdf:type schema:CreativeWork
155 sg:pub.10.1007/s00707-017-2049-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1092654501
156 https://doi.org/10.1007/s00707-017-2049-3
157 rdf:type schema:CreativeWork
158 sg:pub.10.1007/s11081-006-9974-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1045827911
159 https://doi.org/10.1007/s11081-006-9974-7
160 rdf:type schema:CreativeWork
161 sg:pub.10.1023/a:1023283917997 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032929633
162 https://doi.org/10.1023/a:1023283917997
163 rdf:type schema:CreativeWork
164 grid-institutes:grid.266190.a schema:alternateName Smead Department of Aerospace Engineering Sciences, University of Colorado, 80309, Boulder, CO, USA
165 schema:name Smead Department of Aerospace Engineering Sciences, University of Colorado, 80309, Boulder, CO, USA
166 rdf:type schema:Organization
167 grid-institutes:grid.6835.8 schema:alternateName Department of Fluid Mechanics, Universitat Politècnica de Catalunya · BarcelonaTech (UPC), 08019, Barcelona, Spain
168 schema:name Department of Fluid Mechanics, Universitat Politècnica de Catalunya · BarcelonaTech (UPC), 08019, Barcelona, Spain
169 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...