Journal of Aerospace System Engineering (항공우주시스템공학회지)

The Society for Aerospace System Engineering (항공우주시스템공학회)
권호 표지
DOI QR코드

DOI QR Code

Anomaly Detections Model of Aviation System by CNN

합성곱 신경망(CNN)을 활용한 항공 시스템의 이상 탐지 모델 연구

  • Received : 2023.05.25
  • Accepted : 2023.07.11
  • Published : 2023.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, Urban Aircraft Mobility (UAM) has been attracting attention as a transportation system of the future, and small drones also play a role in various industries. The failure of various types of aviation systems can lead to crashes, which can result in significant property damage or loss of life. In the defense industry, where aviation systems are widely used, the failure of aviation systems can lead to mission failure. Therefore, this study proposes an anomaly detection model using deep learning technology to detect anomalies in aviation systems to improve the reliability of development and production, and prevent accidents during operation. As training and evaluating data sets, current data from aviation systems in an extremely low-temperature environment was utilized, and a deep learning network was implemented using the convolutional neural network, which is a deep learning technique that is commonly used for image recognition. In an extremely low-temperature environment, various types of failure occurred in the system's internal sensors and components, and singular points in current data were observed. As a result of training and evaluating the model using current data in the case of system failure and normal, it was confirmed that the abnormality was detected with a recall of 98 % or more.

최근 미래의 운송시스템으로 도심교통항공(Urban Aircraft Mobility)이 주목받고 있으며 소형 드론도 다양한 산업에서 역할을 하고 있다. 다양한 종류의 항공 시스템 고장은 추락으로 막대한 재산 및 인명 피해로 이어질 수 있다. 항공 시스템이 많이 활용되는 무기체계에서도 고장은 임무 실패의 결과를 유발한다. 본 논문에서는 항공 시스템의 이상(Anomaly)을 탐지하여 개발 및 생산 간 시스템의 신뢰도를 높이고 운용 중 사고를 예방할 수 있도록 딥러닝 기술을 활용한 이상 탐지 모델을 연구했다. 모델 훈련 및 평가 데이터로 극저온 환경에서 시스템의 전류 데이터를 활용하였으며 이미지 인식에 많이 활용되는 딥러닝 기법 합성곱 신경망(CNN; Convolutional Neural Network)을 활용하여 딥러닝 네트워크를 구현했다. 시험 대상 시스템은 극저온 환경에서 다양한 형태의 고장이 유발되었고 전륫값의 특이점이 나타났다. 시스템 정상 및 고장 데이터를 활용하여 모델을 훈련 시키고 평가한 결과 98% 이상의 재현율(Recall)로 이상 탐지하는 것을 확인했다.

Keywords

References

  1. Y. Kim, I. Choi "Analysis of Human Casualties on the Ground in Urban Area due to UAM Crash," Journal of Advanced Navigation Technology, vol. 26 no. 5, pp. 281-288, Oct 2022  https://doi.org/10.12673/JANT.2022.26.5.281
  2. UAM Team Korea, "K-UAM Concept of Operations 1.0," Sep. 2021. 
  3. J. Seo, J. Park, J. Yoo, H. Park, "Anomaly Detection System in Mechanical Facility Equipment: Using Long Short-Term Memory Variational Autoencoder," J Korean Soc Qual Manag, vol. 49, no. 4, pp. 581-594, Dec 2021. 
  4. K. Choi, J. YI, C. Park, S. Yoon, "Deep Learning for Anomaly Detection in Time-Series Data: Review, Analysis, and Guidelines," IEEE Access, vol. 9, pp. 120043-120065, 2021.  https://doi.org/10.1109/ACCESS.2021.3107975
  5. E. Lee, H. Bae, H. Kim, H. Han, Y. Lee, J. Son, "Trends in AI Technology for Smart Manufacturing in the future," Electronics and Telecommunications Trends, vol. 35, no. 1, pp. 60-70, Feb 2020  https://doi.org/10.22648/ETRI.2020.J.350106
  6. A. Geron, Hands-On Machine Learning with Scikit-Learn & Tensorflow, O'Reilly Media, Oct. 2019 
  7. J. Oh, J. Kim, H. Lee, K. Choi, "The Fundamental Design and Analysis of UAV Avionics System for Maritime Operation," KSAS 2018 Fall Conference, pp. 1317-1319, Nov 2018. 
  8. Q. Wen, L. Sun, F. Yang, X. Song, J. Gao, X. Wang, H. Xu "Time Series Data Augmentaion for Deep Learning: A Survey" IJCAI 2021, March 2021. 
  9. L. Basora, X. Olive, T. Dubot "Recent Advances in Anomaly Detection Methods Applied to Aviation" Aerospace 2019, vol. 6, no 11:117, Oct 2019. 
  10. N. Park "Research Trends on Deep Learning for Anomaly Detection of Aviation Safety," Electronics and Telecommunications Trends, vol. 36, no. 5, pp. 82-91, Oct 2021  https://doi.org/10.22648/ETRI.2021.J.360509
  11. S. Jeong, J. Shin, S. Min, J. Hur "Failure Diagnostics of Thermal Imaging Camera by the LSTM," Journal of the Korea Academia-Industrial, vol. 23, no. 12, pp. 355-363, Dec 2022.  https://doi.org/10.5762/KAIS.2022.23.12.355
  12. E. Kim, S. Shin "Audio-Based Drone Defect Detection Using Recurrence Plot and Deep Learning," The Journal of Korea Institute of Communication and Information Sciences, vol. 48, no. 1, pp. 114-122, Jan 2023.  https://doi.org/10.7840/kics.2023.48.1.114
  13. M. Kim "Anomaly Diagnosis of Rotational Machinery Using Time-Series Vibration Data Based on Time-Distributed CNN-LSTM" Journal of Korea Multimedia Society, vol. 25, no. 11, pp. 1547-1556, Nov 2022.  https://doi.org/10.9717/KMMS.2022.25.11.1547
  14. Y. LeCun, Y. Bengio, G. Hinton "Deep Learning," Nature, vol. 521, pp. 436-444, May 2015  https://doi.org/10.1038/nature14539
  15. M. Lee, H. Ahn "A Time Series Graph based Convolutional Neural Network Model for Effective Input Variable Pattern Learning : Application to the Prediction of Stock Market" J Intell Inform Syst, vol. 24, no. 1, pp. 167-181, March 2018. 
  16. M. Saripuddin, A. Suliman, S. S. Sameon, B. N. Jorgensen "Random Undersampling on Inbalance Time Series Data for Anomaly Detection" 4th International Conference on Machine Learning and Machine Intelligence, pp. 151-156, Sep 2021. 
  17. S. Mishra "Handling Imbalanced Data: SMOTE vs. Random Undersampling" IRJET, vol. 4 no, 8, pp. 317-320, Aug 2017.