TU Darmstadt / ULB / TUprints

Evaluation of the aircraft fuel economy using advanced statistics and machine learning

Baumann, S. ; Neidhardt, T. ; Klingauf, U. (2024)
Evaluation of the aircraft fuel economy using advanced statistics and machine learning.
In: CEAS Aeronautical Journal, 2021, 12 (3)
doi: 10.26083/tuprints-00023508
Article, Secondary publication, Publisher's Version

[img] Text
Copyright Information: CC BY 4.0 International - Creative Commons, Attribution.

Download (1MB)
Item Type: Article
Type of entry: Secondary publication
Title: Evaluation of the aircraft fuel economy using advanced statistics and machine learning
Language: English
Date: 30 April 2024
Place of Publication: Darmstadt
Year of primary publication: 2021
Place of primary publication: Wien
Publisher: Springer
Journal or Publication Title: CEAS Aeronautical Journal
Volume of the journal: 12
Issue Number: 3
DOI: 10.26083/tuprints-00023508
Corresponding Links:
Origin: Secondary publication DeepGreen

Fuel represents a significant proportion of an airline’s operating costs. Statistical analyses and physical models have been used to monitor and estimate fuel consumption up to now, but these can have considerable inaccuracies. This means that, currently, there are no suitable detection methods for the evaluation of aircraft retrofits, of which some only suggest a fuel efficiency potential in the tenths of a percent range. This article examines suitable assessments of the fuel economy of aircraft and especially aircraft with and without retrofitting. For this purpose, the effects of technical influences such as measurement errors and external uncertainties such as turbulence on the evaluation of the fuel economy are examined in more detail. The focus of the article is on a discussion of possible optimization potentials of conventional statistical evaluation methods, especially regarding possible misinterpretations and spurious correlations. This discussion is exemplarily based on a case study of simulated flight data of an Airbus A320 (with and without improved wing tips (sharklets) as an exemplary retrofit). For this purpose, a suitable simulation environment is presented in which relevant environmental parameters such as wind and turbulence can be set, and measurement errors in the recorded data can be manipulated. It is found that measurement errors as well as turbulence can lead to a bias in key figures that are used for the evaluation of fuel flow signals. The effect of turbulence can partly be mitigated by the use of an improved key figure the authors propose. The investigation is also carried out using a data-based evaluation method to simulate the fuel flow using a machine learning model (random forests), whereby the effects of turbulence and measurement errors significantly influence the fuel flow predicted by the model in the same order of magnitude as potential retrofit measures.

Uncontrolled Keywords: Aircraft fuel economy, Retrofits, Machine learning, Aviation, Fuel efficiency, Data-based models, Grey box modeling, Noise, X-Plane, Simulation
Status: Publisher's Version
URN: urn:nbn:de:tuda-tuprints-235087
Additional Information:

A Correction to this article was published on 27 July 2021

Classification DDC: 600 Technology, medicine, applied sciences > 620 Engineering and machine engineering
Divisions: 16 Department of Mechanical Engineering > Institute of Flight Systems and Automatic Control (FSR)
Date Deposited: 30 Apr 2024 12:23
Last Modified: 30 Apr 2024 12:23
SWORD Depositor: Deep Green
URI: https://tuprints.ulb.tu-darmstadt.de/id/eprint/23508
Actions (login required)
View Item View Item