Topology optimisation with an implicit function and parameterized cutting surface

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper introduces a new implicit function based method for topology optimization that can: obtain solutions with smooth boundaries, be solved using standard mathematical programming methods and reduce the number of design variables. Using implicit functions for topology optimization is attractive because the solutions have clearly defined, smooth boundaries. Most current methods use the zero level-set of the implicit function to define the boundary. The implicit function is then modified during optimization to move the boundary location and connectivity. The new approach proposed in this paper abandons the zero level-set idea and instead uses a fixed signed-distance implicit function. The definition of the boundary from the fixed implicit function is then modified during optimization. This is achieved by using a cutting surface and defining the boundary as the intersection of the cutting surface and signed-distance function. The cutting surface is parameterized and the parameters become the design variables during optimization. Thus, the optimization problem can be solved using mathematical programming and the number of parameters used to define the cutting surface is less than the number of elements in the analysis mesh. The new method is demonstrated using minimization of compliance, minimization of volume and complaint mechanism problems. The results show that the method can obtain good solutions to well-known problems with smooth, clearly defined boundaries and that this can be achieved using significantly fewer design variables compared with element-based methods.
Original languageEnglish
Title of host publicationECCOMAS 2016
Subtitle of host publicationVII European Congress on Computational Methods in Applied Sciences and Engineering
EditorsM. Papadrakakis, V. Papadopoulos, G. Stefanou, V. Plevris
PublisherEccomas
Pages3553-3565
Number of pages12
ISBN (Print)978-618-82844-0-1
Publication statusPublished - 5 Jun 2016
EventECCOMAS Congress 2016: VII European Congress on Computational Methods in Applied Sciences and Engineering - Crete, Greece
Duration: 5 Jun 201610 Jun 2016
https://www.eccomas2016.org

Conference

ConferenceECCOMAS Congress 2016
Abbreviated titleECCOMAS 2016
CountryGreece
Period5/06/1610/06/16
Internet address

Fingerprint

Shape optimization
Mathematical programming

Keywords

  • Topology optimization
  • implicit functions
  • mathematical programming
  • design space reduction

Cite this

Dunning, P. D. (2016). Topology optimisation with an implicit function and parameterized cutting surface. In M. Papadrakakis, V. Papadopoulos, G. Stefanou, & V. Plevris (Eds.), ECCOMAS 2016: VII European Congress on Computational Methods in Applied Sciences and Engineering (pp. 3553-3565). Eccomas.

Topology optimisation with an implicit function and parameterized cutting surface. / Dunning, Peter Donald.

ECCOMAS 2016: VII European Congress on Computational Methods in Applied Sciences and Engineering. ed. / M. Papadrakakis; V. Papadopoulos; G. Stefanou; V. Plevris. Eccomas, 2016. p. 3553-3565.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Dunning, PD 2016, Topology optimisation with an implicit function and parameterized cutting surface. in M Papadrakakis, V Papadopoulos, G Stefanou & V Plevris (eds), ECCOMAS 2016: VII European Congress on Computational Methods in Applied Sciences and Engineering. Eccomas, pp. 3553-3565, ECCOMAS Congress 2016, Greece, 5/06/16.
Dunning PD. Topology optimisation with an implicit function and parameterized cutting surface. In Papadrakakis M, Papadopoulos V, Stefanou G, Plevris V, editors, ECCOMAS 2016: VII European Congress on Computational Methods in Applied Sciences and Engineering. Eccomas. 2016. p. 3553-3565
Dunning, Peter Donald. / Topology optimisation with an implicit function and parameterized cutting surface. ECCOMAS 2016: VII European Congress on Computational Methods in Applied Sciences and Engineering. editor / M. Papadrakakis ; V. Papadopoulos ; G. Stefanou ; V. Plevris. Eccomas, 2016. pp. 3553-3565
@inproceedings{80960d987df647469f20b5af56e010d5,
title = "Topology optimisation with an implicit function and parameterized cutting surface",
abstract = "This paper introduces a new implicit function based method for topology optimization that can: obtain solutions with smooth boundaries, be solved using standard mathematical programming methods and reduce the number of design variables. Using implicit functions for topology optimization is attractive because the solutions have clearly defined, smooth boundaries. Most current methods use the zero level-set of the implicit function to define the boundary. The implicit function is then modified during optimization to move the boundary location and connectivity. The new approach proposed in this paper abandons the zero level-set idea and instead uses a fixed signed-distance implicit function. The definition of the boundary from the fixed implicit function is then modified during optimization. This is achieved by using a cutting surface and defining the boundary as the intersection of the cutting surface and signed-distance function. The cutting surface is parameterized and the parameters become the design variables during optimization. Thus, the optimization problem can be solved using mathematical programming and the number of parameters used to define the cutting surface is less than the number of elements in the analysis mesh. The new method is demonstrated using minimization of compliance, minimization of volume and complaint mechanism problems. The results show that the method can obtain good solutions to well-known problems with smooth, clearly defined boundaries and that this can be achieved using significantly fewer design variables compared with element-based methods.",
keywords = "Topology optimization, implicit functions, mathematical programming, design space reduction",
author = "Dunning, {Peter Donald}",
year = "2016",
month = "6",
day = "5",
language = "English",
isbn = "978-618-82844-0-1",
pages = "3553--3565",
editor = "M. Papadrakakis and V. Papadopoulos and G. Stefanou and V. Plevris",
booktitle = "ECCOMAS 2016",
publisher = "Eccomas",

}

TY - GEN

T1 - Topology optimisation with an implicit function and parameterized cutting surface

AU - Dunning, Peter Donald

PY - 2016/6/5

Y1 - 2016/6/5

N2 - This paper introduces a new implicit function based method for topology optimization that can: obtain solutions with smooth boundaries, be solved using standard mathematical programming methods and reduce the number of design variables. Using implicit functions for topology optimization is attractive because the solutions have clearly defined, smooth boundaries. Most current methods use the zero level-set of the implicit function to define the boundary. The implicit function is then modified during optimization to move the boundary location and connectivity. The new approach proposed in this paper abandons the zero level-set idea and instead uses a fixed signed-distance implicit function. The definition of the boundary from the fixed implicit function is then modified during optimization. This is achieved by using a cutting surface and defining the boundary as the intersection of the cutting surface and signed-distance function. The cutting surface is parameterized and the parameters become the design variables during optimization. Thus, the optimization problem can be solved using mathematical programming and the number of parameters used to define the cutting surface is less than the number of elements in the analysis mesh. The new method is demonstrated using minimization of compliance, minimization of volume and complaint mechanism problems. The results show that the method can obtain good solutions to well-known problems with smooth, clearly defined boundaries and that this can be achieved using significantly fewer design variables compared with element-based methods.

AB - This paper introduces a new implicit function based method for topology optimization that can: obtain solutions with smooth boundaries, be solved using standard mathematical programming methods and reduce the number of design variables. Using implicit functions for topology optimization is attractive because the solutions have clearly defined, smooth boundaries. Most current methods use the zero level-set of the implicit function to define the boundary. The implicit function is then modified during optimization to move the boundary location and connectivity. The new approach proposed in this paper abandons the zero level-set idea and instead uses a fixed signed-distance implicit function. The definition of the boundary from the fixed implicit function is then modified during optimization. This is achieved by using a cutting surface and defining the boundary as the intersection of the cutting surface and signed-distance function. The cutting surface is parameterized and the parameters become the design variables during optimization. Thus, the optimization problem can be solved using mathematical programming and the number of parameters used to define the cutting surface is less than the number of elements in the analysis mesh. The new method is demonstrated using minimization of compliance, minimization of volume and complaint mechanism problems. The results show that the method can obtain good solutions to well-known problems with smooth, clearly defined boundaries and that this can be achieved using significantly fewer design variables compared with element-based methods.

KW - Topology optimization

KW - implicit functions

KW - mathematical programming

KW - design space reduction

M3 - Conference contribution

SN - 978-618-82844-0-1

SP - 3553

EP - 3565

BT - ECCOMAS 2016

A2 - Papadrakakis, M.

A2 - Papadopoulos, V.

A2 - Stefanou, G.

A2 - Plevris, V.

PB - Eccomas

ER -