Open AccessImproving deterministic pitch motions estimation using bivariate sequential wave input
Author(s) -
Yucheng Liu,
Qiumeng Zheng,
Wenyang Duan,
LiMin Huang
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/688/3/033017
Subject(s) - nonlinear system , bivariate analysis , ship motions , response amplitude operator , series (stratigraphy) , computer science , motion (physics) , point (geometry) , term (time) , control theory (sociology) , engineering , marine engineering , artificial intelligence , mathematics , geology , physics , machine learning , paleontology , geometry , control (management) , quantum mechanics , hull
As the ship navigates through waves, it will sway continuously with six degrees of freedom, which will adversely affect the offshore operations. Accurate real-time estimation of deterministic ship motions under wave excitation is a key to ship motion prediction and assistance decision-making. In the actual marine environment, ocean waves and ship motions are often nonlinear. Therefore, an effective nonlinear estimation model to accurately estimate the real-time response of the wave-induced ship motions is of great concern. Due to its unique advantages in dealing with nonlinear time series, Long Short-Term Memory network can provide a powerful method for the estimation of nonlinear wave-induced ship motions. Pitch motions as an oscillating motion put forward higher requirements for model input. Based on the Long Short-Term Memory network model using the wave time history information as input to estimate the ship pitch motion, this paper proposes a pitch estimation model received a bivariate sequential wave time series as input. With the use of the nonlinear wave generated by numerical simulation and the corresponding ship motion time history data, the feasibility of the new model is verified and compared with the corresponding single-point sequential wave input model, determined its superiority.