Aspects of approximate optimisation: overcoming the curse of dimensionality and design of experiments

Sophie Trichon, M.H.A. Bonte, Jean-Philippe Ponthot, Antonius H. van den Boogaard

Research output: Contribution to conferencePaper

15 Downloads (Pure)

Abstract

Coupling optimisation algorithms to Finite Element Methods (FEM) is a very promising way to achieve optimal metal forming processes. However, many optimisation algorithms exist and it is not clear which of these algorithms to use. This paper investigates the sensitivity of a Sequential Approximate Optimisation algorithm (SAO) proposed in [1-4] to an increasing number of design variables and compares it with two other algorithms: an Evolutionary Strategy (ES) and an Evolutionary version of the SAO (ESAO). In addition, it observes the influence of different Designs Of Experiments used with the SAO. It is concluded that the SAO is very capable and efficient and its combination with an ES is not beneficial. Moreover, the use of SAO with Fractional Factorial Design is the most efficient method, rather than Full Factorial Design as proposed in [1-4].
Original languageUndefined
Number of pages8
Publication statusPublished - 2007

Keywords

  • IR-59672

Cite this

@conference{f8d1481f072b452eb6f1023e92e6d6b2,
title = "Aspects of approximate optimisation: overcoming the curse of dimensionality and design of experiments",
abstract = "Coupling optimisation algorithms to Finite Element Methods (FEM) is a very promising way to achieve optimal metal forming processes. However, many optimisation algorithms exist and it is not clear which of these algorithms to use. This paper investigates the sensitivity of a Sequential Approximate Optimisation algorithm (SAO) proposed in [1-4] to an increasing number of design variables and compares it with two other algorithms: an Evolutionary Strategy (ES) and an Evolutionary version of the SAO (ESAO). In addition, it observes the influence of different Designs Of Experiments used with the SAO. It is concluded that the SAO is very capable and efficient and its combination with an ES is not beneficial. Moreover, the use of SAO with Fractional Factorial Design is the most efficient method, rather than Full Factorial Design as proposed in [1-4].",
keywords = "IR-59672",
author = "Sophie Trichon and M.H.A. Bonte and Jean-Philippe Ponthot and {van den Boogaard}, {Antonius H.}",
year = "2007",
language = "Undefined",

}

Aspects of approximate optimisation: overcoming the curse of dimensionality and design of experiments. / Trichon, Sophie; Bonte, M.H.A.; Ponthot, Jean-Philippe; van den Boogaard, Antonius H.

2007.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Aspects of approximate optimisation: overcoming the curse of dimensionality and design of experiments

AU - Trichon, Sophie

AU - Bonte, M.H.A.

AU - Ponthot, Jean-Philippe

AU - van den Boogaard, Antonius H.

PY - 2007

Y1 - 2007

N2 - Coupling optimisation algorithms to Finite Element Methods (FEM) is a very promising way to achieve optimal metal forming processes. However, many optimisation algorithms exist and it is not clear which of these algorithms to use. This paper investigates the sensitivity of a Sequential Approximate Optimisation algorithm (SAO) proposed in [1-4] to an increasing number of design variables and compares it with two other algorithms: an Evolutionary Strategy (ES) and an Evolutionary version of the SAO (ESAO). In addition, it observes the influence of different Designs Of Experiments used with the SAO. It is concluded that the SAO is very capable and efficient and its combination with an ES is not beneficial. Moreover, the use of SAO with Fractional Factorial Design is the most efficient method, rather than Full Factorial Design as proposed in [1-4].

AB - Coupling optimisation algorithms to Finite Element Methods (FEM) is a very promising way to achieve optimal metal forming processes. However, many optimisation algorithms exist and it is not clear which of these algorithms to use. This paper investigates the sensitivity of a Sequential Approximate Optimisation algorithm (SAO) proposed in [1-4] to an increasing number of design variables and compares it with two other algorithms: an Evolutionary Strategy (ES) and an Evolutionary version of the SAO (ESAO). In addition, it observes the influence of different Designs Of Experiments used with the SAO. It is concluded that the SAO is very capable and efficient and its combination with an ES is not beneficial. Moreover, the use of SAO with Fractional Factorial Design is the most efficient method, rather than Full Factorial Design as proposed in [1-4].

KW - IR-59672

M3 - Paper

ER -