Abstract

Even though deicing or airframe coating technologies continue to develop, aircraft icing is still one of the critical threats to aviation. While the detection of potential icing clouds has been conducted using geostationary satellite data in the US and Europe, there is not yet a robust model that detects potential icing areas in East Asia. In this study, we proposed machine-learning-based icing detection models using data from two geostationary satellites—the Communication, Ocean, and Meteorological Satellite (COMS) Meteorological Imager (MI) and the Himawari-8 Advanced Himawari Imager (AHI)—over Northeast Asia. Two machine learning techniques—random forest (RF) and multinomial log-linear (MLL) models—were evaluated with quality-controlled pilot reports (PIREPs) as the reference data. The machine-learning-based models were compared to the existing models through five-fold cross-validation. The RF model for COMS MI produced the best performance, resulting in a mean probability of detection (POD) of 81.8%, a mean overall accuracy (OA) of 82.1%, and mean true skill statistics (TSS) of 64.0%. One of the existing models, flight icing threat (FIT), produced relatively poor performance, providing a mean POD of 36.4%, a mean OA of 61.0, and a mean TSS of 9.7%. The Himawari-8 based models also produced performance comparable to the COMS models. However, it should be noted that very limited PIREP reference data were available especially for the Himawari-8 models, which requires further evaluation in the future with more reference data. The spatio-temporal patterns of the icing areas detected using the developed models were also visually examined using time-series satellite data.

Highlights

  • Aircraft icing is a dangerous threat that results in many accidents which can cause fatalities and financial losses [1,2]

  • The objectives of this research were to (1) develop icing detection algorithms for a portion of East Asia based on machine learning approaches with COMS Meteorological Imager (MI) and Himawari-8 Advanced Himawari Imager (AHI) products; (2) compare the proposed algorithms with the existing flight icing threat (FIT) and Korea Meteorological Administration (KMA) algorithms; and (3) interpret the properties of the potential icing clouds identified by the algorithms

  • Two machine learning approaches—random forest (RF) and multinomial log-linear (MLL)—were used to develop the icing detection models. Both machine learning-based models resulted in better performance (POD of 68–82% and probability of false detection (POFD) of 16–18%) than those of the existing physical-theory-based models (POD of 12–36% and POFD of 7–27%) when COMS data were used

Read more

Summary

Introduction

Aircraft icing is a dangerous threat that results in many accidents which can cause fatalities and financial losses [1,2]. It is a phenomenon in which supercooled droplets (SCDs) collide with a hard surface forming an ice film. When icing forms on aircraft bodies and wings, the aircraft’s balance is disturbed, resulting in a loss of control. For this reason, detecting and avoiding potential icing areas is crucial for aviation safety. The detection of SCDs, especially in the freezing phase of rain, is usually conducted using a thresholding approach based on the subfreezing temperature range and high relative humidity [6]

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call