Physics guided deep learning for data-driven aircraft fuel consumption modeling

Simple item page
dc.contributor.authorUzun, Mevlut
dc.contributor.authorDemirezen, Mustafa Umut
dc.contributor.authorInalhan, Gokhan
dc.date.accessioned2021-03-03T12:36:14Z
dc.date.available2021-03-03T12:36:14Z
dc.date.issued2021-02-08
dc.description.abstractThis paper presents a physics-guided deep neural network framework to estimate fuel consumption of an aircraft. The framework aims to improve data-driven models’ consistency in flight regimes that are not covered by data. In particular, we guide the neural network with the equations that represent fuel flow dynamics. In addition to the empirical error, we embed this physical knowledge as several extra loss terms. Results show that our proposed model accomplishes correct predictions on the labeled test set, as well as assuring physical consistency in unseen flight regimes. The results indicate that our model, while being applicable to the aircraft’s complete flight envelope, yields lower fuel consumption error measures compared to the model-based approaches and other supervised learning techniques utilizing the same training data sets. In addition, our deep learning model produces fuel consumption trends similar to the BADA4 aircraft performance model, which is widely utilized in real-world operations, in unseen and untrained flight regimes. In contrast, the other supervised learning techniques fail to produce meaningful results. Overall, the proposed methodology enhances the explainability of data-driven models without deteriorating accuracy.en_UK
dc.identifier.citationUzun M, Umut Demirezen M, Inalhan G. (2021) Physics guided deep learning for data-driven aircraft fuel consumption modeling. Aerospace, Volume 8, Issue 2, February 2021, Article number 44en_UK
dc.identifier.issn2226-4310
dc.identifier.urihttps://doi.org/10.3390/aerospace8020044
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/16432
dc.language.isoenen_UK
dc.publisherMDPIen_UK
dc.rightsAttribution 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/*
dc.subjectBADAen_UK
dc.subjectfuel consumption modelingen_UK
dc.subjectaircraft performance modelingen_UK
dc.subjectneural networksen_UK
dc.subjectmachine learningen_UK
dc.subjectphysics guided deep learningen_UK
dc.titlePhysics guided deep learning for data-driven aircraft fuel consumption modelingen_UK
dc.typeArticleen_UK

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Physics_guided_deep_learning_for_data-driven_aircraft_fueL-2021.pdf
Size:
3.07 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Staff publications (SATM)