0
Research Papers

Technology Characterization Models and Their Use in Systems Design

[+] Author and Article Information
Robert R. Parker

Design Systems Laboratory,
Department of Mechanical Engineering,
Texas A&M University,
College Station, TX 77845
e-mail: parker.ro@gmail.com

Edgar Galvan

Design Systems Laboratory,
Department of Mechanical Engineering,
Texas A&M University,
College Station, TX 77845
e-mail: e_galvan@tamu.edu

Richard J. Malak

Design Systems Laboratory,
Department of Mechanical Engineering,
Texas A&M University,
College Station, TX 77845
e-mail: rmalak@tamu.edu

1Corresponding author.

Contributed by the Design Automation Committee of ASME for publication in the JOURNAL OF MECHANICAL DESIGN. Manuscript received March 13, 2012; final manuscript received October 24, 2013; published online April 28, 2014. Assoc. Editor: Harrison M. Kim.

J. Mech. Des 136(7), 071003 (Apr 28, 2014) (11 pages) Paper No: MD-12-1155; doi: 10.1115/1.4025960 History: Received March 13, 2012; Revised October 24, 2013

Prior research suggests that set-based design representations can be useful for facilitating collaboration among engineers in a design project. However, existing set-based methods are limited in terms of how the sets are constructed and in their representational capability. The focus of this article is on the problem of modeling the capabilities of a component technology in a way that can be communicated and used in support of system-level decision making. The context is the system definition phases of a systems engineering project, when engineers still are considering various technical concepts. The approach under investigation requires engineers familiar with the component- or subsystem-level technologies to generate a set-based model of their achievable technical attributes, called a technology characterization model (TCM). Systems engineers then use these models to explore system-level alternatives and choose the combination of technologies that are best suited to the design problem. Previously, this approach was shown to be theoretically sound from a decision making perspective under idealized circumstances. This article is an investigation into the practical effectiveness of different TCM representational methods under realistic conditions such as having limited data. A power plant systems engineering problem is used as an example, with TCMs generated for different technical concepts for the condenser component. Samples of valid condenser realizations are used as inputs to the TCM representation methods. Two TCM representation methods are compared based on their solution accuracy and computational effort required: a Kriging-based interpolation and a machine learning technique called support vector domain description (SVDD). The results from this example hold that the SVDD-based method provides the better combination of accuracy and efficiency.

FIGURES IN THIS ARTICLE
<>
Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Illustrative examples of the SVDDFS, KRIGFS and KRIGOS TCM representation methods

Grahic Jump Location
Fig. 2

Nonideal Rankine cycle

Grahic Jump Location
Fig. 3

Systems design problem layout

Grahic Jump Location
Fig. 4

Heat exchanger technologies

Grahic Jump Location
Fig. 5

The (a) mean utility found using each method and (b) pairwise t-test that data with 95% confidence interval. The utility of the TCM methods is that of the feasible design found during discipline-level design.

Grahic Jump Location
Fig. 6

Mean normalized distance between the target attributes and the attributes of the feasible design found during discipline-level design

Grahic Jump Location
Fig. 7

Mean number of system and component level function evaluations for the different methods with 95% confidence interval

Grahic Jump Location
Fig. 8

The (a) mean distance between true points on the frontier to the approximation for the different TCM representation methods and (b) pairwise t-test that data with 95% confidence interval. The test shows statistically significant difference between all the representation methods.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In