The use of a Bayesian weighted least-squares approach to accelerate empirical engine model generation

Research output: Contribution to journalArticle

  • 2 Citations

Abstract

An alternative to the commonly used design-of-experiments (DoE) technique based on the use of prior knowledge to speed up the model generation phase of the test programme is presented. Data are gathered in areas where the prior knowledge is least accurate and is blended into an overall empirical model using a Bayesian weighted least-squares (BWLS) approach. The work demonstrates the potential of the BWLS mechanism to reduce the test effort by the incorporation of prior knowledge from neighbouring operating points. Where the prior knowledge is a good representation of the current operating point the number of test points required to allow model convergence may be half that required by a DoE approach. Conversely, the Bayesian technique with inaccurate prior knowledge can still generate an accurate model at the expense of an increased number of test points over a standard experimental design, because the prior will incorrectly influence the model. The algorithm was tested with and without an outlier detection mechanism. This mechanism proved effective in detecting and rejecting simulated outliers and retesting the spurious data point.
LanguageEnglish
Pages366-394
JournalProceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering
Volume225
Issue numberD3
DOIs
StatusPublished - Feb 2011

Fingerprint

Engines
Design of experiments

Keywords

  • model-based engine calibration
  • engine modelling
  • Bayesian statistics

Cite this

@article{4f8647c794d34942b14a2dfae9b6a7b4,
title = "The use of a Bayesian weighted least-squares approach to accelerate empirical engine model generation",
abstract = "An alternative to the commonly used design-of-experiments (DoE) technique based on the use of prior knowledge to speed up the model generation phase of the test programme is presented. Data are gathered in areas where the prior knowledge is least accurate and is blended into an overall empirical model using a Bayesian weighted least-squares (BWLS) approach. The work demonstrates the potential of the BWLS mechanism to reduce the test effort by the incorporation of prior knowledge from neighbouring operating points. Where the prior knowledge is a good representation of the current operating point the number of test points required to allow model convergence may be half that required by a DoE approach. Conversely, the Bayesian technique with inaccurate prior knowledge can still generate an accurate model at the expense of an increased number of test points over a standard experimental design, because the prior will incorrectly influence the model. The algorithm was tested with and without an outlier detection mechanism. This mechanism proved effective in detecting and rejecting simulated outliers and retesting the spurious data point.",
keywords = "model-based engine calibration, engine modelling, Bayesian statistics",
author = "Brace, {C. J.} and S. Akehurst and Ward, {M C}",
year = "2011",
month = "2",
doi = "10.1177/09544070JAUTO1562",
language = "English",
volume = "225",
pages = "366--394",
journal = "Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering",
issn = "0954-4070",
publisher = "Sage Publications",
number = "D3",

}

TY - JOUR

T1 - The use of a Bayesian weighted least-squares approach to accelerate empirical engine model generation

AU - Brace,C. J.

AU - Akehurst,S.

AU - Ward,M C

PY - 2011/2

Y1 - 2011/2

N2 - An alternative to the commonly used design-of-experiments (DoE) technique based on the use of prior knowledge to speed up the model generation phase of the test programme is presented. Data are gathered in areas where the prior knowledge is least accurate and is blended into an overall empirical model using a Bayesian weighted least-squares (BWLS) approach. The work demonstrates the potential of the BWLS mechanism to reduce the test effort by the incorporation of prior knowledge from neighbouring operating points. Where the prior knowledge is a good representation of the current operating point the number of test points required to allow model convergence may be half that required by a DoE approach. Conversely, the Bayesian technique with inaccurate prior knowledge can still generate an accurate model at the expense of an increased number of test points over a standard experimental design, because the prior will incorrectly influence the model. The algorithm was tested with and without an outlier detection mechanism. This mechanism proved effective in detecting and rejecting simulated outliers and retesting the spurious data point.

AB - An alternative to the commonly used design-of-experiments (DoE) technique based on the use of prior knowledge to speed up the model generation phase of the test programme is presented. Data are gathered in areas where the prior knowledge is least accurate and is blended into an overall empirical model using a Bayesian weighted least-squares (BWLS) approach. The work demonstrates the potential of the BWLS mechanism to reduce the test effort by the incorporation of prior knowledge from neighbouring operating points. Where the prior knowledge is a good representation of the current operating point the number of test points required to allow model convergence may be half that required by a DoE approach. Conversely, the Bayesian technique with inaccurate prior knowledge can still generate an accurate model at the expense of an increased number of test points over a standard experimental design, because the prior will incorrectly influence the model. The algorithm was tested with and without an outlier detection mechanism. This mechanism proved effective in detecting and rejecting simulated outliers and retesting the spurious data point.

KW - model-based engine calibration

KW - engine modelling

KW - Bayesian statistics

UR - http://www.scopus.com/inward/record.url?scp=79951666260&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1177/09544070jAUTO1562

U2 - 10.1177/09544070JAUTO1562

DO - 10.1177/09544070JAUTO1562

M3 - Article

VL - 225

SP - 366

EP - 394

JO - Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering

T2 - Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering

JF - Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering

SN - 0954-4070

IS - D3

ER -