CNN-RNN 기반의 DNN을 활용한 DP 선박의 전력부하 예측

홍창우¹^{, *} ; 허견²

1해군본부 소령/연세대학교 전기전자공학부 박사과정
2연세대학교 전기전자공학부 교수

Electric Power Load Forecasting of DP Ships by Using CNN-RNN Based DNN

Chang Woo Hong¹^{, *} ; Kyeon Hur²

1Lieutenant Commander, ROK Navy Headquarter/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

선박의 전력 부하를 예측하는 것은 육상 시스템의 전력 부하를 예측하는 것만큼 중요하다. 특히 환경오염물질 규제에 대한 대응에 따라 발전기의 효율적인 운영 및 PMS 시스템 활용이 중요해지고 있으며, 이는 곧 연비를 증가시키는 방법이다. 이 논문에서는 비선형 전력 부하 특성을 가진 dynamic positioning 선박의 단기 전기 부하를 예측하기 위해 심층 인공 신경망을 사용한다. Dynamic positioning 선박은 함위를 유지하기 위해 스러스터를 사용하며, 이는 큰 전력 부하의 사용이다. 스러스터 부하는 불규칙한 펄스 부하로 상황에 따라 정전을 일으킬 수 있다는 점에서 매우 중요한 요소이다. 제안된 심층 신경망은 CNN과 RNN 알고리즘을 결합하여 다양한 변수를 사용하는 불규칙한 전력 부하를 예측한다. 특히 K-fold cross-validation을 사용하여 제한된 데이터 세트에 대한 모델의 높은 강건도를 달성하였다. 실험 연구는 실제 선박의 항해 시뮬레이션 데이터를 사용하여 전력 부하를 예측하고 서로 다른 CNN-RNN 레이어 깊이에 따른 예측 결과를 비교한다.

Abstract

Electric power load forecasting of a ship is just as important as predicting the power load of a land system. In particular, in response to environmental pollutant regulations, efficient operation of the generator and utilization of the PMS system are becoming critical, increasing fuel efficiency. This paper uses a deep artificial neural network to predict the short-term electrical load of a dynamic positioning vessel with non-linear power load characteristics. Dynamic positioning ships use thrusters to maintain their ships, which is the use of large power loads. Thruster load is a significant factor in that it can cause a power outage depending on the situation as an irregular pulse load. In particular, the high robustness of the model was achieved for a limited data set using K-fold cross-validation. The experimental study predicts the electric power load using the navigation simulation data of a real ship and compares the prediction results according to different CNN-RNN layer depths.

Keywords:

Deep Neural Network, Dynamic Positioning Ship, Electric Power Load Forecasting

키워드:

심층 신경망, DP 선박, 전력부하 예측

Acknowledgments

이 논문은 2021년도 한국해군과학기술학회 하계학술대회 발표 논문임.

References

Kim, Si-Yeon, Hyun-Woo Jung, Jeong-Do Park, Seung-Mook Baek, Woo-Seon Kim, Kyung-Hee Chon, and Kyung-Bin Song. "Weekly maximum electric load forecasting for 104 weeks by seasonal ARIMA model." Journal of the Korean Institute of Illuminating and Electrical Installation Engineers, Vol. 28, no. 1 (2014): 50-56. [https://doi.org/10.5207/JIEIE.2014.28.1.050]
Lee, K. Y., Y. T. Cha, and J. H. Park. "Short-term load forecasting using an artificial neural network." IEEE transactions on power systems, Vol. 7, no. 1 (1992): 124-132. [https://doi.org/10.1109/59.141695]
Hsu, Che-Chiang, and Chia-Yon Chen. "Regional load Forecasting in Taiwan ― Applications of Artificial Neural networks." Energy Conversion and Management, Vol. 44, no. 12 (2003): 1941-1949. [https://doi.org/10.1016/S0196-8904(02)00225-X]
Sulligoi, Giorgio, Andrea Vicenzutti, and Roberto Menis. "All-electric Ship Design: From Electrical Propulsion to Integrated Electrical and Electronic Power Systems." IEEE Transactions on Transportation Electrification, Vol. 2, no. 4 (2016): 507-521. [https://doi.org/10.1109/TTE.2016.2598078]
Ku, Hyun-Keun, Hye-Rim Seo, and Jang-Mok Kim. "Lithium-ion Battery Energy Storage System for Power Quality Improvement in Electrical Propulsion Ships." The Transactions of the Korean Institute of Power Electronics, Vol. 20, no. 4 (2015): 351-355. [https://doi.org/10.6113/TKPE.2015.20.4.351]
Cuculić, Aleksandar, Dubravko Vučetić, Rene Prenc, and Jasmin Ćelić. "Analysis of Energy Storage Implementation on Dynamically Positioned Vessels." Energies, Vol. 12, no. 3 (2019): 444. [https://doi.org/10.3390/en12030444]
Satpathi, Kuntal, VSK Murthy Balijepalli, and Abhisek Ukil. "Modeling and Real-time Scheduling of DC Platform Supply Vessel for Fuel Efficient Operation." IEEE Transactions on Transportation Electrification, Vol. 3, no. 3 (2017): 762-778. [https://doi.org/10.1109/TTE.2017.2744180]
Lauvdal, Trygve, and A. K. Ådnanes. "Power Management System with Fast Acting Load Reduction for DP Vessels." In Dynamic Positioning Conference, pp. 17-18. 2000.
Sainath TN, Vinyals O, Senior A, Sak H. Convolutional, Long Short-term Memory, Fully Connected Deep Neural networks. In2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015 Apr 19 (pp. 4580-4584). IEEE. [https://doi.org/10.1109/ICASSP.2015.7178838]
LeCun, Yann, Léon Bottou, Yoshua Bengio, and Patrick Haffner. "Gradient-based Learning Applied to Document Recognition." Proceedings of the IEEE, Vol. 86, no. 11 (1998): 2278-2324. [https://doi.org/10.1109/5.726791]
Hochreiter, Sepp, and Jürgen Schmidhuber. "Long Short-term Memory." Neural Computation, Vol. 9, no. 8 (1997): 1735-1780. [https://doi.org/10.1162/neco.1997.9.8.1735]