자율운항선박 지원을 위한 실시간 관측 기반의 해양환경 인공지능 예측 기술 검증
Verification of Artificial Intelligence Prediction Technology of Marine Environment Based on Real-Time Observation for Maritime Autonomous Surface Ship
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
자율운항선박 등 스마트 선박에서 항로상의 해양환경 상태를 관측·예측하는 과정은 필수 요소이며, 선박의 관측자료만을 이용하여 의사결정이 가능하도록 해양환경 정보를 생산하는 기술이 필요하다. 제약된 선박 장비 상황 속에서 실시간 수집 가능한 해양환경정보는 파랑, 해상풍이 일반적이다. 본 연구에서는 4종의 인공지능 알고리즘(ANN, RNN, Conv-LSTM, GAN)과 2개의 학습 구성 방안(초단기-비연속, 장기-연속)을 접목하여 항로상의 시계열 관측 정보만으로 다음 시간대의 해양환경 예측 정보를 생산할 수 있는 기술을 개발하였다. 개발된 알고리즘은 기상청에서 운용 중인 ‘기상1호’의 실 선박 관측자료 1년 치를 활용하여 계절별로 비교·분석하였다. 초단기-비연속 학습 구성에서는 지도 학습모델 중 RNN이 가장 좋은 성능을 보였고 비지도 학습 모델은 낮은 성능을 보인 반면, 장기-연속 학습 구성에서는 모든 모델이 모든 계절에서 높은 정확도를 보였다. 학습 구성 방안에 따른 알고리즘별 예측 정확도 검증에서는 초단기-비연속에 비해 장기-연속 학습 구성이 파랑, 해상풍 모두에서 계절별 예측 정확도가 우수하였다. 본 연구를 통해 자율운항선박 내 데이터 기반 해양환경 예측 정보를 제공하기 위한 기술의 유의미성과 가능성을 확인하였으며 향후 범용수로데이터모델(S-100)과의 연계, 선박 운동성 및 항로 추정 등과 연계되어 보다 실질적인 의사결정 정보 생산에 기여할 것으로 기대된다.
Abstract
For smart ships, such as autonomous vessels, technology that can predict and observe the state of the marine environment along the route in real-time is essential. This requires observational data from ships. The capability to generate marine environmental information for effective decision-making is crucial. Typically, the information that can be collected in real-time under limited equipment conditions on ships includes wave and offshore wind data. In this study, we developed an algorithm that produces marine environment prediction information for subsequent time periods based solely on time series observation data from the route. This prediction information integrates four types of artificial intelligence algorithms (ANN, RNN, Conv-LSTM, and GAN) and two learning structure methods (ultra-short-term - discontinuous and long-term - continuous). The algorithm was compared and analyzed by season using one year of actual ship observation data from 'Meteorological No. 1,' operated by the Korea Meteorological Administration. Among the supervised learning models, RNN demonstrated the best performance in the ultra-short-term and discontinuous learning structure. While unsupervised learning models exhibited lower performance, all models achieved high accuracy in the long-term continuous learning structure across all seasons. In terms of prediction accuracy by algorithm based on learning structure, the long-term continuous learning structure outperformed the short-term continuous learning structure in seasonal prediction accuracy for both waves and sea surface winds. This study confirms the significance and potential of technology for providing data-based marine environment prediction information within autonomous ships. In the future, this technology is expected to facilitate the generation of more practical decision-making information, rather than simply linking it to next-generation waterway products, ship mobility, and route estimation.
참고문헌 (20)
- Ahn, Y. S., "Validation of Building Power Consumption Prediction Accuracy According to Frozen Number of Layers based on Transfer Learning", 40, pp.151~160 (2020)
- Azeem, A., "Electrical load fore-casting models for different generation modalities : A review", IEEE Access, 9, pp.1422391~1142263 (2021)
- Al-Absi, M. A., "IoT Technology with Marine Environment Protection and Monitoring", International Conference on Smart Computing and Cyber Security, pp.81~89 (2021), Springer
- Brophy, E., "Quick and Easy Time Series Generation with Established Image-based GANs", CoRR, abs, 2019 (1902)
- Hong, G. Y., "Development of Real-Time Forecasting System of Marine Enviromental Information for Ship Routing", Journal of the Korean Society for Marine Enviromental Engineering, 8, pp.46~52 (2005)
- Issa, M., "Maritime Autonomous Surface Ships : Problems and Challenges Facing the Regulatory Process", Sustainability, 14(23), pp.15630~ (2022)
- Karlis, T., "Maritime law issues related to the operation of unmanned autonomous cargo ships", WMU J Marit Affairs, 17(1), pp.119~128 (2018)
- Kang, D. H., "A Study on the HDF5 Data Model Design for Gridded Marine Weather Information Based on S-100", Journal of Navigation and Port Research, 46(3), pp.158~167 (2022)
- Kim, D. J., "An Evaluation Technique for the Path-following Control Performance of Autonomous Surface Ships", Journal of Navigation and Port Research, 47(1), pp.10~17 (2023)
- Kim, J., "Technology Trend and Preparations for Maritime Autonomous Surface Ship (MASS)", 56(4), pp.4~7 (2019)
- Khan, R. U., "Risk Assessment for Autonomous Ships Using an Integrated Machine Learning Approach", Engineering Proceedings, 46(1), pp.9~ (2023)
- Lee, G. S., "Wave prediction in a Harbour using Deep Learning with Offshore Data", Journal of Korean Society of Coastal and Ocean Engineers, 33(6), pp.367~373 (2021)
- Lee, J., "Real Ship Experimental Analysis on the Influence of Fuel Oil Consumption Considering Ship Speed and Ocean Environment", Journal of the Society of Naval Architects of Korea, 60(4), pp.203~212 (2023)
- Li, S., "Maritime autonomous surface ships(MASS) : implementation and legal issues", Maritime Business Review, 4(4), pp.330~339 (2019)
- Park, G. I., "Introduction of Optimum Navigation Route Assessment System Based on Weather Forecasting and Seakeeping Prediction", Journal of Korean Navigation and Port Research, 28(10), pp.833~841 (2004)
- Qi, S., "An efficient GAN-based predictive framework for multivariate time series anomaly prediction in cloud data centers", The Journal of Supercomputing, 80(1), pp.1268~1293 (2024)
- Ravuri, S., "Skillful precipitation nowcasting using deep generative models of radar", Nature, 597, pp.672~677 (2021)
- Thombre, S., "Sensors and AI techniques for situational awareness in autonomous ships : a review", IEEE Transactions on Intelligent Transportation Systems, 23, pp.64~83 (2022)
- Tsujimoto, M., "Evaluation on Ship Performance in Actual Seas-Ship Speed, Fuel Oil Consumption and Main Engine Power", Journal of Kansai Society of Naval Architects, 229, pp.69~78 (1998)
- Yim, J. B., "Effect of Turning Characteristics of Maritime Autonomous Surface Ships on Collision", Journal Navigation Port Research, 45(6), pp.298~305 (2021)