Development of aircraft fuel burn modeling techniques with applications to global emissions modeling and assessment of the benefits of reduced vertical separation minimums

Abstract

Given the current level of concern over anthropogenic climate change and the role of commercial aviation in this process, the ability to adequately model and quantify fuel burn and emissions on a system wide scale is of high importance. In particular, the ability to adequately assess the ability of operational alternatives within the commercial aviation system to improve system efficiency and reduce environmental impact is essential. Much work has already been done with this end in mind; however, given the high degree of complexity associated with a large system such as this, there is opportunity for improvements in modeling capability. The work presented in this thesis was conducted with this aim, to build additional functionality and fidelity into an established modeling method. The FAA System for assessing Aviation's Global Emissions (SAGE)'is a well established model for the creation of global inventories of aviation fuel use and emissions. There are, however, two aspects of the model which could benefit from improvements in modeling methodology. The first is the way in which the specific fuel consumption (SFC) is calculated. Previous to this study, SFC was calculated through the methods put forward in EUROCONTROL's Base of Aircraft Data (BADA).

(cont.) These methods are based on aircraft type specific coefficients and perform well in the context of global inventories; however, they lack the necessary functional dependence on ambient and operational variables to adequately assess the effects of the small changes often associated with various operational alternatives. An effort was made to assess the functional dependence of SFC on these variables through statistical analysis of a large body of Computerized Flight Recorder Data (CFDR) and to use this as a basis for improving the modeling of SFC in SAGE. The result of this effort was the introduction of a statistically-derived SFC model into SAGE which contained the desired functional dependence on temperature, pressure, Mach number, and thrust, and thereby improved the fidelity with which fuel bum is modeled. This SFC model, as implemented in SAGE, reduced the average absolute error by 21% as compared to the original BADA model. Additional improvement was made with the introduction of weather information into the SAGE model. Previously, the model assumed standard atmospheric conditions and zero wind. An algorithm was devised which processed and incorporated global assimilated weather data from NASA Goddard into the performance calculations within SAGE.

(cont.) It was found the introduction of dynamic weather contributed greatly to the accuracy of SAGE given that the system wide average true airspeed error is 10% when the assumption of zero winds is used. Finally this improved model was used to quantify the benefits of implementing Reduced Vertical Separation Minimums (RVSM) in US airspace in January of 2005. This was accomplished through a comparison of system wide efficiency during representative time periods prior to and following RVSM implementation. The results of this analysis provide insight into not only the benefits of RVSM, but also the effects of these model improvements and the efficacy of the different efficiency metrics used. It was found that RVSM resulted in an increase in fuel efficiency (nm/kg) of 1.81% (± 0.55%) and an increase in NOx efficiency (nm/kg) of 3.14% (± 1.25%). An additional control comparison was made, during these same time periods, of system efficiency over the North Atlantic and Western Europe where RVSM had been implemented several years prior. Using an efficiency metric which normalized for the difference in winds between the two periods it was found that there was indeed no benefit seen in this control study providing support for the US Domestic RVSM results.