General perspectives on model construction and evaluation for stochastic estimation, with application to a blunt trailing edge wake

General perspectives on model construction and evaluation for stochastic estimation, with... The technique of stochastic estimation is examined as a specific application of linear least squares modelling. Factors that are relevant to the objectives of estimation in fluids, such as the number of sensors, the use of multiple time lags, and the strength of linear correlations, are discussed in the context of a general regression formulation. We consolidate the established findings of several research fields in order to outline clearly the potential pitfalls and reasonable performance expectations of these empirical strategies. Experimental measurements of velocity and fluctuating pressure in the wake of a blunt trailing edge body are used for quantitative illustration of key considerations for model construction and performance evaluation. It is emphasized that estimator accuracy is influenced strongly by the physical relationships among the measured variables, in addition to their correlation with the estimated variable. The evaluation of several performance metrics on an independent test set provides valuable information for the selection of a suitably complex model. In particular, “variance inflation” is interpreted as an indicator of the potential amplification of noise by a stochastic estimator. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Experiments in Fluids Springer Journals

General perspectives on model construction and evaluation for stochastic estimation, with application to a blunt trailing edge wake

Loading next page...
 
/lp/springer_journal/general-perspectives-on-model-construction-and-evaluation-for-0LjjRJUh2D
Publisher
Springer Berlin Heidelberg
Copyright
Copyright © 2014 by Springer-Verlag Berlin Heidelberg
Subject
Engineering; Engineering Fluid Dynamics; Fluid- and Aerodynamics; Engineering Thermodynamics, Heat and Mass Transfer
ISSN
0723-4864
eISSN
1432-1114
D.O.I.
10.1007/s00348-014-1756-y
Publisher site
See Article on Publisher Site

Abstract

The technique of stochastic estimation is examined as a specific application of linear least squares modelling. Factors that are relevant to the objectives of estimation in fluids, such as the number of sensors, the use of multiple time lags, and the strength of linear correlations, are discussed in the context of a general regression formulation. We consolidate the established findings of several research fields in order to outline clearly the potential pitfalls and reasonable performance expectations of these empirical strategies. Experimental measurements of velocity and fluctuating pressure in the wake of a blunt trailing edge body are used for quantitative illustration of key considerations for model construction and performance evaluation. It is emphasized that estimator accuracy is influenced strongly by the physical relationships among the measured variables, in addition to their correlation with the estimated variable. The evaluation of several performance metrics on an independent test set provides valuable information for the selection of a suitably complex model. In particular, “variance inflation” is interpreted as an indicator of the potential amplification of noise by a stochastic estimator.

Journal

Experiments in FluidsSpringer Journals

Published: Jun 19, 2014

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off