Click for next page ( 14

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 13
13 (see data tables, Appendix A). In this formalism, because This has been estimated by combining the overall systematic ratios are used, the systematic error in the measurement of uncertainty in the measurements of emission index with the each specific hydrocarbon actually drops out. For example, if average spread observed in the fuel flow dependence param- the proton transfer for reaction mass spectrometry (PTR-MS) eter for formaldehyde and ethene. The overall spread, how- measurement of styrene is systematically high by 25%, because ever, of observed variability in this parameter when consider- the emission index is being divided by the emission index ing the entire distribution of VOCs measured in this study taken with the same test hour, the bias in this measurement is large (see Figure III-6). The variability in the observations will essentially cancel when the instrument is operating within can be characterized by the 95% confidence limit. Twice the its linear dynamic range. Gaussian half width of the distribution would imply a range of Overall uncertainty estimates for the analytical instru- -86 to -18 s kg-1. We attribute the width of this distribution to ments used in this study are discussed in Appendix E. An genuine engine-to-engine and intra-engine state variability. assessment of analytical instrument contribution to error, A cursory examination of the test results in Appendix A in the context of determining the near-idle fuel flow depen- reveals consistent patterns for all VOC and CO emission dence of the emissions index, yields 5%. In this work, the indices for a single test. This test pools the results from several emission index for CO2 has been assumed to be a constant engines, installed on different airframes, with operational 3160 g kg-1. This assumption is good to 3%. Furthermore, parameters characterized by different on-board sensors, oper- this work assumes that all fuel carbon is in the form of CO2, ating with different bleed air requirements, and with differing which introduces a bias of up to 5%. Taken together, the maintenance histories. This project assumes that each of these combination of analytical instrument error and estimated factors contributes to the observed width in the distribution analysis methodology uncertainty indicates an overall sys- of near-idle fuel flow dependences. The variability in these tematic uncertainty of 12%. data should not be confused with error. In a model estimate of airport HAP emissions, the number of aircraft source engines will be much larger than the 11 studied here. If the influence III.5Fuel Flow Dependence of engine-to-engine variability effectively forms a normal and Variability distribution of idle fuel flow emissions dependence, then the Based on the dominant VOCs as well as the ICAO UHC uncertainty from this source of error in the tabulation of total databank results, this study recommends a near-idle fuel flow airport emissions can be characterized by the uncertainties in dependence of -51 s kg-1 for the CFM56-7B engine variants. the central value and the width of the emissions distribution. The overall uncertainty in this recommended central value For these data, the uncertainty in the central value is much of the near-idle fuel flow dependence parameter is 23%. less than the width of the distribution.