심층신경망과 피어슨 상관계수를 활용한 터보팬 엔진 잔여유효수명 예측

홍창우¹^{, *} ; 허견²

1해군본부 소령/연세대학교 전기전자공학부 박사과정
2연세대학교 전기전자공학부 교수

Prognosis of the Remaining Useful Life of a Turbofan Engine Using Deep Neural Network and Pearson Correlation Coefficient

Chang Woo Hong¹^{, *} ; Kyeon Hur²

1Lieutenant commander, Republic of Korea Navy Headquarter (ROKN HQ)/ Ph.D course student, Dept. of Electrical and Electronic Engineering, Yonsei University
2Professor, Dept. of Electrical and Electronic Engineering, Yonsei University

Correspondence to: ^*Chang Woo Hong Tel: +82-2-2123-7436 E-mail: spearw@yonsei.ac.kr

초록

이 논문은 변수의 복잡성을 감소시키는 상관계수 분석을 사용하여 터보 팬 엔진의 잔여 유효수명(RUL)의 예측 정확도를 향상시키기 위한 심층 인공신경망 구조를 제안한다. 제안된 모델은 CNN(convolutional neural network), LSTM(long short-term memory) 및 양방향 LSTM(Bidirectional LSTM)을 적절히 배열하여 각 알고리즘의 성능을 최대화한다. 또한 제안된 모델의 학습 능력을 향상시키기 위해 residual network 및 dropout 기술을 적용한다. 제안된 방법의 성능은 기존 방법과의 비교 및 검증을 통해 낮은 상관 데이터를 제외한 제안된 방법의 유효성이 제시된다.

Abstract

This paper proposes a deep neural network architecture to improve the prognosis accuracy of remaining useful life (RUL) of the turbofan engine by using the correlation coefficient analysis that reduces the variable complexity. The proposed model maximizes the performance of each algorithm by appropriately stacking convolutional neural network (CNN), long short-term memory (LSTM), and bidirectional LSTM. It also includes residual network and dropout technique to improve the learning performance of the proposed model. The performance of the proposed method is compared and verified with that of conventional method, which suggests the validity of the proposed method excluding the low correlated data.

Keywords:

Deep Neural Network, Pearson Correlation Coefficient, Prognostics and Health Management, Turbofan Engine

키워드:

심층 신경망, 피어슨 상관계수, 건전성 예측관리, 터보팬 엔진

References

L. Pintelon, A. Parodi-Herz, “Maintenance: an evolutionary perspective,” Springer, 2008.
N.H. Kim, D. An, J.H. Choi. “Prognostics and health management of engineering systems,” Springer, 2017. [https://doi.org/10.1007/978-3-319-44742-1]
G. S. Babu, P. Zhao, and X.-L. Li. “Deep convolutional neural network based regression approach for estimation of remaining useful life,” In International conference on database systems for advanced applications, Springer, 2016, pp. 214-228. [https://doi.org/10.1007/978-3-319-32025-0_14]
S. Zheng, K. Ristovski, A. Farahat, and C. Gupta. “Long short-term memory network for remaining useful life estimation,” 2017 IEEE international conference on prognostics and health management (ICPHM), 2017, pp. 88-95. [https://doi.org/10.1109/ICPHM.2017.7998311]
J. Zhang, P. Wang, R. Yan, and R. X. Gao. “Long short-term memory for machine remaining life prediction,” Journal of manufacturing systems, Vol. 48, July 2018, pp. 78–86. [https://doi.org/10.1016/j.jmsy.2018.05.011]
C. Zhang, P. Lim, A. K. Qin, and K. C. Tan. “Multiobjective deep belief networks ensemble for remaining useful life estimation in prognostics,” IEEE transactions on neural networks and learning systems, Vol. 28, October 2016, pp. 2306–2318. [https://doi.org/10.1109/TNNLS.2016.2582798]
A. L. Ellefsen, E. Bjørlykhaug, V. Æsøy, S. Ushakov, and H. Zhang. “Remaining useful life predictions for turbofan engine degradation using semi-supervised deep architecture,” Reliability Engineering & System Safety, Vol. 183, March 2019, pp. 240–251. [https://doi.org/10.1016/j.ress.2018.11.027]
Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, “Gradient-based learning applied to document recognition,” Proceedings of the IEEE, Vol. 86, November 1998, pp. 2278–2324. [https://doi.org/10.1109/5.726791]
S. Hochreiter and J. Schmidhuber. “Long short-term memory,” Neural computation, Vol. 9, November 1997, pp. 1735–1780. [https://doi.org/10.1162/neco.1997.9.8.1735]
M. Schuster and K. K. Paliwal. “Bidirectional recurrent neural networks,” IEEE transactions on signal processing, Vol. 45, November 1997, pp. 2673–2681. [https://doi.org/10.1109/78.650093]