BIOS: an object-oriented framework for Surrogate-Based Optimization using bio-inspired algorithms

Elias Saraiva Barroso; Leonardo Gonçalves Ribeiro; Marina Alves Maia; I.B.C.M. Rocha; Evandro Parente; Antônio Macário Cartaxo de Melo

doi:10.1007/s00158-022-03302-0

BIOS: an object-oriented framework for Surrogate-Based Optimization using bio-inspired algorithms

Elias Saraiva Barroso, Leonardo Gonçalves Ribeiro, Marina Alves Maia, I.B.C.M. Rocha, Evandro Parente^*, Antônio Macário Cartaxo de Melo

^*Corresponding author for this work

Applied Mechanics

Research output: Contribution to journal › Article › Scientific › peer-review

1 Citation (Scopus)

58 Downloads (Pure)

Abstract

This paper presents BIOS (acronym for Biologically Inspired Optimization System), an object-oriented framework written in C++, aimed at heuristic optimization with a focus on Surrogate-Based Optimization (SBO) and structural problems. The use of SBO to deal with structural optimization has grown considerably in recent years due to the outstanding gain in efficiency, often with little loss in accuracy. This is especially promising when adaptive sampling techniques are used. However, many issues are yet to be addressed before SBO can be employed reliably in most optimization problems. In that sense, continuous experimentation, testing and comparison are needed, which can be more easily carried out in an existing framework. The architecture is designed to implement conventional nature inspired algorithms and Sequential Approximated Optimization (SAO). The system aims to be efficient, easy to use and extensible. The efficiency and accuracy of the system are assessed on a set of benchmarks, and on the optimization of functionally graded structures. Excellent results are obtained.

Original language	English
Article number	203
Number of pages	27
Journal	Structural and Multidisciplinary Optimization
Volume	65
Issue number	7
DOIs	https://doi.org/10.1007/s00158-022-03302-0
Publication status	Published - 2022

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Composite structures
Object-oriented framework
Sequential Approximated Optimization
Structural optimization
Surrogate-Based Optimization

Access to Document

10.1007/s00158-022-03302-0

s00158-022-03302-0Final published version, 4.86 MB

Cite this

@article{8e8e8068aabe48a196ac44a9813a49cd,

title = "BIOS: an object-oriented framework for Surrogate-Based Optimization using bio-inspired algorithms",

abstract = "This paper presents BIOS (acronym for Biologically Inspired Optimization System), an object-oriented framework written in C++, aimed at heuristic optimization with a focus on Surrogate-Based Optimization (SBO) and structural problems. The use of SBO to deal with structural optimization has grown considerably in recent years due to the outstanding gain in efficiency, often with little loss in accuracy. This is especially promising when adaptive sampling techniques are used. However, many issues are yet to be addressed before SBO can be employed reliably in most optimization problems. In that sense, continuous experimentation, testing and comparison are needed, which can be more easily carried out in an existing framework. The architecture is designed to implement conventional nature inspired algorithms and Sequential Approximated Optimization (SAO). The system aims to be efficient, easy to use and extensible. The efficiency and accuracy of the system are assessed on a set of benchmarks, and on the optimization of functionally graded structures. Excellent results are obtained.",

keywords = "Composite structures, Object-oriented framework, Sequential Approximated Optimization, Structural optimization, Surrogate-Based Optimization",

author = "Barroso, {Elias Saraiva} and Ribeiro, {Leonardo Gon{\c c}alves} and Maia, {Marina Alves} and I.B.C.M. Rocha and Evandro Parente and {de Melo}, {Ant{\^o}nio Mac{\'a}rio Cartaxo}",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.",

year = "2022",

doi = "10.1007/s00158-022-03302-0",

language = "English",

volume = "65",

journal = "Structural and Multidisciplinary Optimization",

issn = "1615-147X",

publisher = "Springer",

number = "7",

}

TY - JOUR

T1 - BIOS

T2 - an object-oriented framework for Surrogate-Based Optimization using bio-inspired algorithms

AU - Barroso, Elias Saraiva

AU - Ribeiro, Leonardo Gonçalves

AU - Maia, Marina Alves

AU - Rocha, I.B.C.M.

AU - Parente, Evandro

AU - de Melo, Antônio Macário Cartaxo

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2022

Y1 - 2022

N2 - This paper presents BIOS (acronym for Biologically Inspired Optimization System), an object-oriented framework written in C++, aimed at heuristic optimization with a focus on Surrogate-Based Optimization (SBO) and structural problems. The use of SBO to deal with structural optimization has grown considerably in recent years due to the outstanding gain in efficiency, often with little loss in accuracy. This is especially promising when adaptive sampling techniques are used. However, many issues are yet to be addressed before SBO can be employed reliably in most optimization problems. In that sense, continuous experimentation, testing and comparison are needed, which can be more easily carried out in an existing framework. The architecture is designed to implement conventional nature inspired algorithms and Sequential Approximated Optimization (SAO). The system aims to be efficient, easy to use and extensible. The efficiency and accuracy of the system are assessed on a set of benchmarks, and on the optimization of functionally graded structures. Excellent results are obtained.

AB - This paper presents BIOS (acronym for Biologically Inspired Optimization System), an object-oriented framework written in C++, aimed at heuristic optimization with a focus on Surrogate-Based Optimization (SBO) and structural problems. The use of SBO to deal with structural optimization has grown considerably in recent years due to the outstanding gain in efficiency, often with little loss in accuracy. This is especially promising when adaptive sampling techniques are used. However, many issues are yet to be addressed before SBO can be employed reliably in most optimization problems. In that sense, continuous experimentation, testing and comparison are needed, which can be more easily carried out in an existing framework. The architecture is designed to implement conventional nature inspired algorithms and Sequential Approximated Optimization (SAO). The system aims to be efficient, easy to use and extensible. The efficiency and accuracy of the system are assessed on a set of benchmarks, and on the optimization of functionally graded structures. Excellent results are obtained.

KW - Composite structures

KW - Object-oriented framework

KW - Sequential Approximated Optimization

KW - Structural optimization

KW - Surrogate-Based Optimization

UR - http://www.scopus.com/inward/record.url?scp=85133561259&partnerID=8YFLogxK

U2 - 10.1007/s00158-022-03302-0

DO - 10.1007/s00158-022-03302-0

M3 - Article

AN - SCOPUS:85133561259

SN - 1615-147X

VL - 65

JO - Structural and Multidisciplinary Optimization

JF - Structural and Multidisciplinary Optimization

IS - 7

M1 - 203

ER -

BIOS: an object-oriented framework for Surrogate-Based Optimization using bio-inspired algorithms

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this