Related references
Note: Only part of the references are listed.
Article
Computer Science, Information Systems
Ranran Lou et al.
Summary: Machine learning has become a popular research method in various fields and is applied to analyze and apply ocean data. It has great potential and significance in protecting the ocean environment, predicting ocean elements, exploring the unknown, and responding to extreme weather.
MULTIMEDIA SYSTEMS
(2023)
Article
Engineering, Marine
Wei Hao et al.
Summary: This study proposes an EMD-LSTM method combining the advantages of the LSTM model and EMD for accurate wave forecasting. Experimental results show that this method can effectively reduce prediction errors and increase forecast lead time.
Article
Engineering, Marine
Kumars Mahmoodi et al.
Summary: Accurate prediction of extreme wave height occurrence is crucial for offshore and onshore structures. A new hybrid NOF-ELM method is proposed in this study, showing that radial basis ELM performs best in extreme wave detection based on meteorological data.
SHIPS AND OFFSHORE STRUCTURES
(2022)
Article
Engineering, Marine
Xinfang Li et al.
Summary: This study proposes a method for short-term and long-term ocean wave forecasting using GRU network and conducts case studies on 6 stations along the coast of China. The results show that this method outperforms other methods in 1-hour forecasts and exhibits more robustness compared to the LSTM algorithm in 3-hour forecasts. This method is essential for ocean safety and exploitation.
Article
Energy & Fuels
F. A. R. Abdullah et al.
Summary: Significant wave height (SWH) is important for marine operational and maritime activities. A novel modeling approach based on long short-term memory (LSTM) neural network model is proposed for SWH prediction in Indonesian waters. The single-step predictions show accurate results, but the multi-step predictions are inconsistent. The length of data significantly affects the performance of the LSTM model in single-step prediction, while fluctuations and data complexity influence the performance in multi-step prediction.
JOURNAL OF OCEAN ENGINEERING AND MARINE ENERGY
(2022)
Article
Computer Science, Theory & Methods
Fan Meng et al.
Summary: This study develops a real-time tropical cyclone wave height forecasting system using an adaptive time-frequency decomposition method and a mixture of deep learning methods. The system performs well in predicting wave heights caused by tropical cyclones and achieves excellent results in recent super typhoon events.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Meteorology & Atmospheric Sciences
Brandon J. Bethel et al.
Summary: A LSTM neural network model is proposed for predicting hurricane-forced significant wave heights in the Caribbean Sea. The model shows faster and more accurate predictions compared to numerical wave models, with lower computational cost and expertise required. The results demonstrate high accuracy in forecasting hurricane-forced wave heights and suggest that LSTM can supplement numerical wave models for extreme wave height forecasting.
Article
Meteorology & Atmospheric Sciences
Tao Song et al.
Summary: This paper proposes a novel deep learning model for tropical cyclone track prediction. The model can excavate the historical track information in a deeper and more accurate way, and performs well in mid- to long-term track forecasting.
JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
(2022)
Article
Oceanography
Song Gao et al.
Summary: A forecasting model for wave heights in the Bohai Sea based on LSTM neural network showed improved prediction accuracy compared to conventional numerical wave models, especially in the range of significant wave heights of 3-5 meters. The numbers of hidden neurons, buoys used, and time length of training samples all impacted the prediction accuracy.
ACTA OCEANOLOGICA SINICA
(2021)
Article
Mathematical & Computational Biology
Ahmed Ali et al.
Summary: With advancements in ocean observation technology, the increase in marine data volume presents an opportunity for training better predictive models. Efforts to predict marine data have primarily focused on machine learning, deep learning, and statistical predictive models. Experimental results indicate that deep learning models slightly outperform machine learning models in predicting marine features.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2021)
Article
Engineering, Marine
Tao Song et al.
Summary: This study utilized machine learning to optimize unstructured triangular meshes in the global ocean to improve mesh quality. Experiment results showed an increase in the proportion of equilateral triangles and a decrease in the proportion of long, narrow triangles in the optimized unstructured meshes.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2021)
Article
Oceanography
Liangshi Zhao et al.
Summary: The 21st-century Maritime Silk Road initiative provides a new direction for the high-quality development of China and the marine economy along the route. The marine economic efficiency of countries along the MSR showed an overall upward trend from 2007 to 2017, with clear differences between developing and developed countries. Factors such as resource dependence, openness, economic development levels, and industrial structures have significant impacts on marine economic efficiency.
OCEAN & COASTAL MANAGEMENT
(2021)
Article
Multidisciplinary Sciences
Hector Lobeto et al.
Summary: The study shows that extreme waves will undergo significant changes in the future under different climate change scenarios, with the Southern Ocean projected to experience robust increases in extreme wave heights while the tropical north Pacific is expected to see substantial decreases. Additionally, there are notable divergences between the projected behavior of mean and extreme wave conditions in certain ocean regions, indicating that caution should be exercised when extrapolating changes in mean wave conditions to extremes in these areas.
SCIENTIFIC REPORTS
(2021)
Article
Environmental Sciences
Nawin Raj et al.
Summary: Utilizing advanced deep learning algorithms to forecast significant wave height of coastal sea waves can provide crucial information for planning and decision-making in events such as search and rescue and wave surges. The study developed a highly accurate hybrid algorithm for forecasting, with the EEMD-BiLSTM model outperforming other models in terms of accuracy. Pearson's correlation and relative mean square error values demonstrated the superior performance of the EEMD-BiLSTM model compared to traditional methods.
Article
Engineering, Marine
Sargol Memar et al.
Summary: The study aimed to predict maximum seasonal wave height using new integrative data-driven methods, specifically ANFIS-BWO and SVR-BWO. Results showed that the use of BWO and PSO algorithms increased the accuracy of ANFIS and SVR by approximately 18.45%. Furthermore, SVR-BWO was identified as the most accurate model among the developed methods.
Article
Engineering, Marine
Fan Meng et al.
Summary: A bidirectional Gated Recurrent Units (BiGRU) network is introduced for predicting wave height during tropical cyclones, utilizing buoy and TC data for training, the model's performance is found to be stable and superior compared to traditional machine learning methods, especially for 24-hour advance prediction and long-term forecasting.
Article
Meteorology & Atmospheric Sciences
Haoguo Hu et al.
Summary: This study utilized machine learning methods to predict wave height and period in Lake Erie, with the XGBoost model performing the best and LSTM model following. Compared to the WW3 model, ML methods showed advantages in performance and computational cost.
Article
Chemistry, Analytical
Chih-Chiang Wei et al.
Summary: This study aimed to predict wind speeds and wave heights near major ports in Taiwan during typhoon periods. By combining gated recurrent unit (GRU) neural networks and convolutional neural networks (CNNs), multiple prediction models were designed to enhance accuracy. Results showed that the new models performed well in extracting spatial image features and providing useful information from time-series data.
Article
Engineering, Marine
Shuyi Zhou et al.
Summary: This study combines EMD and LSTM networks to significantly improve forecasting accuracy for significant wave height, especially for forecasts exceeding 24 hours. EMD-LSTM responds faster to large waves than LSTM, and it is found that LSTM lacks the ability to adequately predict high-frequency significant wave height intrinsic mode functions, leaving room for further improvements.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2021)
Article
Meteorology & Atmospheric Sciences
Soroush Sorourian et al.
Article
Multidisciplinary Sciences
Yanan Guo et al.
Article
Engineering, Marine
Shuntao Fan et al.
Article
Green & Sustainable Science & Technology
Mumtaz Ali et al.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2020)
Article
Meteorology & Atmospheric Sciences
Stephan Rasp et al.
JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS
(2020)
Review
Green & Sustainable Science & Technology
Mumtaz Ali et al.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2019)
Article
Energy & Fuels
Zheng Yi et al.
PETROLEUM EXPLORATION AND DEVELOPMENT
(2012)
Article
Engineering, Mechanical
Yaguo Lei et al.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2009)
Article
Engineering, Marine
J. Mahjoobi et al.
Article
Mathematics, Interdisciplinary Applications
Zhaohua Wu et al.
ADVANCES IN DATA SCIENCE AND ADAPTIVE ANALYSIS
(2009)