다변량 데이터 분석을 통한 효과적인 전력부하 예측 : Gas Carrier 사례 연구

홍창우¹^{, *} ; 김소연²

1해군 소령/해군사관학교 기계시스템공학과 조교수
2해군 중령/해군사관학교 전기전자공학과 교수

Effective Power Load Prediction Through Multivariate Data Analysis: Gas Carrier Case Study

Chang Woo Hong¹^{, *} ; So Yeon Kim²

1LCDR, ROK Navy/Assistant professor, Dept. of Mechanical System Engineering, Republic of Korea Naval Academy
2CDR, ROK Navy/Professor, Dept. of Electrical Engineering, Republic of Korea Naval Academy

Correspondence to: ^*Chang Woo Hong Dept. of Mechanical System Engineering, Republic of Korea Naval Academy 1 Jungwon-ro, Jinhae-gu, Changwon-si, Gyungsangnam-do, 51704, Republic of Korea Tel: +82-55-907-5307 E-mail: spearw@navy.ac.kr

초록

본 논문에서는 다변량 시계열 데이터의 특징을 보이는 gas carrier의 전력부하를 인공신경망 네트워크 아키텍처를 구성하여 예측한다. 네트워크는 CNN-RNN 기반으로 구성하며 효과적인 학습과 예측 정확도 향상을 위해 차원 축소인 주성분분석을 사용하였으며, CNN 알고리즘에서 dilation rate를 달리하며 학습의 결과를 비교하였다. 학습 결과 주성분분석과 Dilated CNN을 적절히 사용하여 다변량 데이터를 높은 정확도로 예측할 수 있음을 보였다.

Abstract

In this paper, the power load of a gas carrier, which shows the characteristics of multivariate time series data, is predicted by configuring an artificial neural network architecture. The network was constructed based on CNN-RNN, and principal component analysis, a dimensionality reduction, was used to improve effective learning and prediction accuracy, and the learning results were compared by varying the dilation rate in the CNN algorithm. The learning results showed that multivariate data can be predicted with high accuracy by appropriately using PCA and Dilated CNN.

Keywords:

Artificial Neural Network, Dilated CNN, Gas Carrier, Principal Component Analysis

키워드:

인공신경망, 확장된 합성곱 신경망, 가스 운반선, 주성분분석

References

Li, Kevin, et al. “Determinants of ship operators’ options for compliance with IMO 2020.” Transportation Research Part D: Transport and Environment 86 (2020): 102459. [https://doi.org/10.1016/j.trd.2020.102459]
Uyanık, Tayfun, et al. “A data-driven approach for generator load prediction in shipboard microgrid: The Chemical Tanker Case Study." Energies 16.13 (2023): 5092. [https://doi.org/10.3390/en16135092]
Wang, Jiang, et al. “CNN-RNN: A unified framework for multi-label image classification.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016. [https://doi.org/10.1109/CVPR.2016.251]
Zhao, Bin, et al. “A CNN–RNN architecture for multi-label weather recognition.” Neurocomputing 322 (2018): 47-57. [https://doi.org/10.1016/j.neucom.2018.09.048]
Zhou, Xiaokang, Yue Li, and Wei Liang. “CNN-RNN based intelligent recommendation for online medical pre-diagnosis support.” IEEE/ACM Transactions on Computational Biology and Bioinformatics 18.3 (2020): 912-921. [https://doi.org/10.1109/TCBB.2020.2994780]
Yu, Fisher, and Vladlen Koltun. “Multi-scale context aggregation by dilated convolutions.” arXiv preprint arXiv:1511.07122 (2015).