JOURNAL ARTICLE

Non-linear Regression Model for Fitting Experimental Emissions Data

Abstract

Abstract The measurement and reporting of non-volatile Particulate Matter (nvPM) emissions is now integral to aircraft engine emission certification. Akin to other gaseous emissions, CAEP11 sets a new LTO based standard. So far, the question how nvPM emission can be corrected for ambient conditions has to remain open. Addressing this, combustor rig tests may be consulted, where performance parameter such as Air-to-fuel ratio, pressure and temperature can be varied independently. However, results from selected tests may be misleading as nvPM emissions may show measurement uncertainties within actual sensitivities. The paper provides a novel approach utilising experimental data to investigate the prevailing sensitivities at different operation conditions and derive applicable correction factors. This methodology makes use of the fundamental assumption that each change measured in nvPM emissions is explainable with the relative change in combustor AFR, the combustion chamber inlet pressure P3, and combustor inlet temperature T3, for a given combustor design running on the same fuel. Computing the relative changes between all measurements of the same test for all performance and emission parameters enables the fitting of a nonlinear regression model to the experimental data. The fitted function consists of the product of the relative change of the performance parameters to the power of polynomial exponents. The methodology has the potential to be applied to a wide variation of emissions data obtained from different sources as combustor rig tests as well as from engine emission tests. The paper concludes with a first application of the methodology to experimental engine emissions data and a short discussion on AFR sensitivity. In order to set-up a more broadly applicable model nvPM pathways and mechanisms and their dependencies must be better understood helping to identify relevant parameters, e.g. characteristic AFR in the nearfield of the fuel spray nozzle.

Keywords:
Combustor Combustion Environmental science Polynomial regression Computer science Automotive engineering Linear regression Engineering

Metrics

0
Cited By
0.00
FWCI (Field Weighted Citation Impact)
0
Refs
0.13
Citation Normalized Percentile
Is in top 1%
Is in top 10%

Topics

Vehicle emissions and performance
Physical Sciences →  Engineering →  Automotive Engineering
Advanced Aircraft Design and Technologies
Physical Sciences →  Environmental Science →  Global and Planetary Change
Advanced Combustion Engine Technologies
Physical Sciences →  Chemical Engineering →  Fluid Flow and Transfer Processes

Related Documents

JOURNAL ARTICLE

Model fitting, parameter estimation, linear and non-linear regression

Giuseppe A. Sagnella

Journal:   Trends in Biochemical Sciences Year: 1985 Vol: 10 (3)Pages: 100-103
BOOK-CHAPTER

Fitting distributions to data using linear regression

Elsevier eBooks Year: 2001 Pages: 360-362
BOOK-CHAPTER

Bayesian Linear Regression Model for Curve Fitting

Michael Li

IFIP advances in information and communication technology Year: 2018 Pages: 363-372
© 2026 ScienceGate Book Chapters — All rights reserved.