Next Article in Journal
Special Issue: Civil and Military Airworthiness: Recent Developments and Challenges
Next Article in Special Issue
Computer-Assisted Aircraft Anti-Icing Fluids Endurance Time Determination
Previous Article in Journal
In Situ Measurement of Carbon Fibre/Polyether Ether Ketone Thermal Expansion in Low Earth Orbit
Previous Article in Special Issue
Pinned Droplet Size on a Superhydrophobic Surface in Shear Flow
Open AccessArticle

Fast Evaluation of Aircraft Icing Severity Using Machine Learning Based on XGBoost

by Sibo Li 1, Jingkun Qin 2, Miao He 3 and Roberto Paoli 1,4,*
1
Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL 60607, USA
2
Department of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100083, China
3
National Oilwell Varco, Houston, TX 77041, USA
4
Computational Science Division and Leadership Computing Facility, Argonne National Laboratory, Lemont, IL 60439, USA
*
Author to whom correspondence should be addressed.
Aerospace 2020, 7(4), 36; https://doi.org/10.3390/aerospace7040036
Received: 14 February 2020 / Revised: 24 March 2020 / Accepted: 27 March 2020 / Published: 31 March 2020
(This article belongs to the Special Issue Deicing and Anti-Icing of Aircraft)
Aircraft icing represents a serious hazard in aviation which has caused a number of fatal accidents over the years. In addition, it can lead to substantial increase in drag and weight, thus reducing the aerodynamics performance of the airplane. The process of ice accretion on a solid surface is a complex interaction of aerodynamic and environmental variables. The complex relationship makes machine learning-based methods an attractive alternative to traditional numerical simulation-based approaches. In this study, we introduce a purely data-driven approach to find the complex pattern between different flight conditions and aircraft icing severity prediction. The supervised learning algorithm Extreme Gradient Boosting (XGBoost) is applied to establish the prediction framework which makes prediction based on any set of observations. The input flight conditions for the proposed prediction framework are liquid water content, droplet diameter and exposure time. The proposed approach is demonstrated in three cases: maximum ice thickness prediction, icing area prediction and icing severity level evaluation. Performance comparison studies and error analysis are also conducted to verify the effectiveness and performance of the proposed method. Results show that the proposed method has reasonable capability in evaluating aircraft icing severity. View Full-Text
Keywords: aircraft icing; machine learning; OpenFOAM; computational fluid dynamics; data-driven aircraft icing; machine learning; OpenFOAM; computational fluid dynamics; data-driven
Show Figures

Figure 1

MDPI and ACS Style

Li, S.; Qin, J.; He, M.; Paoli, R. Fast Evaluation of Aircraft Icing Severity Using Machine Learning Based on XGBoost. Aerospace 2020, 7, 36.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop