Fast insight into high-dimensional parametrized simulation data

Daniel Butnaru, Benjamin Peherstorfer, Hans Joachim Bungartz, Dirk Pfluger

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

Abstract

Numerical simulation has become an inevitable tool in most industrial product development processes with simulations being used to understand the influence of design decisions (parameter configurations) on the structure and properties of the product. However, in order to allow the engineer to thoroughly explore the design space and fine-tune parameters, many - usually very time-consuming - simulation runs are necessary. Additionally, this results in a huge amount of data that cannot be analyzed in an efficient way without the support of appropriate tools. In this paper, we address the two-fold problem: First, instantly provide simulation results if the parameter configuration is changed, and, second, identify specific areas of the design space with concentrated change and thus importance. We propose the use of a hierarchical approach based on sparse grid interpolation or regression which acts as an efficient and cheap substitute for the simulation. Furthermore, we develop new visual representations based on the derivative information contained inherently in the hierarchical basis. They intuitively let a user identify interesting parameter regions even in higher-dimensional settings. This workflow is combined in an interactive visualization and exploration framework. We discuss examples from different fields of computational science and engineering and show how our sparse-grid-based techniques make parameter dependencies apparent and how they can be used to fine-tune parameter configurations.

Original languageEnglish (US)
Title of host publicationProceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012
Pages265-270
Number of pages6
Volume2
DOIs
StatePublished - Dec 1 2012
Event11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012 - Boca Raton, FL, United States
Duration: Dec 12 2012Dec 15 2012

Other

Other11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012
CountryUnited States
CityBoca Raton, FL
Period12/12/1212/15/12

Fingerprint

simulation
Product development
Interpolation
Visualization
Derivatives
Engineers
Computer simulation
workflow
visualization
engineer
engineering
regression
science

Keywords

  • feature identification
  • sparse grids
  • surrogate modeling
  • visual analytics

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Education

Cite this

Butnaru, D., Peherstorfer, B., Bungartz, H. J., & Pfluger, D. (2012). Fast insight into high-dimensional parametrized simulation data. In Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012 (Vol. 2, pp. 265-270). [6406761] https://doi.org/10.1109/ICMLA.2012.189

Fast insight into high-dimensional parametrized simulation data. / Butnaru, Daniel; Peherstorfer, Benjamin; Bungartz, Hans Joachim; Pfluger, Dirk.

Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012. Vol. 2 2012. p. 265-270 6406761.

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

Butnaru, D, Peherstorfer, B, Bungartz, HJ & Pfluger, D 2012, Fast insight into high-dimensional parametrized simulation data. in Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012. vol. 2, 6406761, pp. 265-270, 11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012, Boca Raton, FL, United States, 12/12/12. https://doi.org/10.1109/ICMLA.2012.189
Butnaru D, Peherstorfer B, Bungartz HJ, Pfluger D. Fast insight into high-dimensional parametrized simulation data. In Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012. Vol. 2. 2012. p. 265-270. 6406761 https://doi.org/10.1109/ICMLA.2012.189
Butnaru, Daniel ; Peherstorfer, Benjamin ; Bungartz, Hans Joachim ; Pfluger, Dirk. / Fast insight into high-dimensional parametrized simulation data. Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012. Vol. 2 2012. pp. 265-270
@inproceedings{8342db3a9e5d4ad78628cbedef0269a4,
title = "Fast insight into high-dimensional parametrized simulation data",
abstract = "Numerical simulation has become an inevitable tool in most industrial product development processes with simulations being used to understand the influence of design decisions (parameter configurations) on the structure and properties of the product. However, in order to allow the engineer to thoroughly explore the design space and fine-tune parameters, many - usually very time-consuming - simulation runs are necessary. Additionally, this results in a huge amount of data that cannot be analyzed in an efficient way without the support of appropriate tools. In this paper, we address the two-fold problem: First, instantly provide simulation results if the parameter configuration is changed, and, second, identify specific areas of the design space with concentrated change and thus importance. We propose the use of a hierarchical approach based on sparse grid interpolation or regression which acts as an efficient and cheap substitute for the simulation. Furthermore, we develop new visual representations based on the derivative information contained inherently in the hierarchical basis. They intuitively let a user identify interesting parameter regions even in higher-dimensional settings. This workflow is combined in an interactive visualization and exploration framework. We discuss examples from different fields of computational science and engineering and show how our sparse-grid-based techniques make parameter dependencies apparent and how they can be used to fine-tune parameter configurations.",
keywords = "feature identification, sparse grids, surrogate modeling, visual analytics",
author = "Daniel Butnaru and Benjamin Peherstorfer and Bungartz, {Hans Joachim} and Dirk Pfluger",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/ICMLA.2012.189",
language = "English (US)",
isbn = "9780769549132",
volume = "2",
pages = "265--270",
booktitle = "Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012",

}

TY - GEN

T1 - Fast insight into high-dimensional parametrized simulation data

AU - Butnaru, Daniel

AU - Peherstorfer, Benjamin

AU - Bungartz, Hans Joachim

AU - Pfluger, Dirk

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Numerical simulation has become an inevitable tool in most industrial product development processes with simulations being used to understand the influence of design decisions (parameter configurations) on the structure and properties of the product. However, in order to allow the engineer to thoroughly explore the design space and fine-tune parameters, many - usually very time-consuming - simulation runs are necessary. Additionally, this results in a huge amount of data that cannot be analyzed in an efficient way without the support of appropriate tools. In this paper, we address the two-fold problem: First, instantly provide simulation results if the parameter configuration is changed, and, second, identify specific areas of the design space with concentrated change and thus importance. We propose the use of a hierarchical approach based on sparse grid interpolation or regression which acts as an efficient and cheap substitute for the simulation. Furthermore, we develop new visual representations based on the derivative information contained inherently in the hierarchical basis. They intuitively let a user identify interesting parameter regions even in higher-dimensional settings. This workflow is combined in an interactive visualization and exploration framework. We discuss examples from different fields of computational science and engineering and show how our sparse-grid-based techniques make parameter dependencies apparent and how they can be used to fine-tune parameter configurations.

AB - Numerical simulation has become an inevitable tool in most industrial product development processes with simulations being used to understand the influence of design decisions (parameter configurations) on the structure and properties of the product. However, in order to allow the engineer to thoroughly explore the design space and fine-tune parameters, many - usually very time-consuming - simulation runs are necessary. Additionally, this results in a huge amount of data that cannot be analyzed in an efficient way without the support of appropriate tools. In this paper, we address the two-fold problem: First, instantly provide simulation results if the parameter configuration is changed, and, second, identify specific areas of the design space with concentrated change and thus importance. We propose the use of a hierarchical approach based on sparse grid interpolation or regression which acts as an efficient and cheap substitute for the simulation. Furthermore, we develop new visual representations based on the derivative information contained inherently in the hierarchical basis. They intuitively let a user identify interesting parameter regions even in higher-dimensional settings. This workflow is combined in an interactive visualization and exploration framework. We discuss examples from different fields of computational science and engineering and show how our sparse-grid-based techniques make parameter dependencies apparent and how they can be used to fine-tune parameter configurations.

KW - feature identification

KW - sparse grids

KW - surrogate modeling

KW - visual analytics

UR - http://www.scopus.com/inward/record.url?scp=84873591168&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873591168&partnerID=8YFLogxK

U2 - 10.1109/ICMLA.2012.189

DO - 10.1109/ICMLA.2012.189

M3 - Conference contribution

AN - SCOPUS:84873591168

SN - 9780769549132

VL - 2

SP - 265

EP - 270

BT - Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012

ER -