Abstract
Abstract Ship motion prediction is applied to the ship maneuvering to keep the motion attitude stable all the time, which is of great practical significance to the safety and efficiency of the sailing and shipboard equipment operation. However, ship motion is a complex time-varying nonlinear process, which varies with sailing conditions and environmental factors. Long short-term memory (LSTM) deep learning model provides a potential way for nonlinear ship motion prediction due to its powerful capability in nonlinearity processing. Determination of a reasonable dimension of the input vector and output vector is critical in training the LSTM model. The input vector represents the hidden time autocorrelation in random ship motion, and the output vector is used to evaluate the prediction performance. Therefore, a ship motion prediction synthesis method based on multi-layer LSTM model is proposed. In the Bohai Sea, China, the original motion data are obtained and used as the basis for analysis and demonstration. Inevitably, there is high-frequency noise interference and low-frequency certain trend in the measured data. Low pass filtering and empirical mode decomposition (EMD) methods are used to remove noise. Based on the multi-layer LSTM model, the ship motion data under different speed conditions are analyzed, and the optimal relationship between input vector and output vector is investigated. Numerical results show that this method can improve the prediction result, and realize the prediction function for 20 timestep. This research is of great practicability for ensuring the safety of navigation and improving the efficiency of maritime operations and the intellectualization level of ships.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.