A data-driven method for real-time forecasting of ship motion attitude at sea

Abstract

The invention discloses a data-driven real-time prediction method for the motion attitude of a ship at sea, which performs information screening, information decomposition, information processing and information fusion on the information of the six-degree-of-freedom motion of the ship, and establishes a ship motion prediction model based on multi-dimensional data to predict the future of the ship. The six-degree-of-freedom motion is predicted, so as to perform a comprehensive ship motion attitude prediction. The present invention directly predicts roll angle, pitch angle θ and yaw angle ψ; for longitudinal position X, lateral position Y and yaw angle ψ, the surge velocity u, sway velocity v and yaw angular velocity r are firstly forecasted, Then the predicted values ​​of longitudinal position X, lateral position Y and yaw angle ψ are obtained. The invention makes full use of the measured ship motion state time series information, through the information mining and time series forecast of the information, and utilizes the nonlinear fitting ability of the radial basis function neural network, and can obtain more accurate maritime ship motion forecast.

Description

technical field [0001] The invention relates to a technology for predicting the motion attitude of a ship, in particular to a data-driven real-time prediction method for the motion attitude of a ship at sea. Background technique [0002] The ship motion prediction method based on artificial neural network refers to taking the historical data of ship motion in the past period of time as the input data of the network, and through repeated training of the network, the threshold value of each neuron and the connection weight between each layer of the network are The value is optimal, so that the knowledge stored in the network learning is closer to the output of the actual system (forecast output) when used for reasoning, that is, an approximate forecast value is obtained for the new network input data. The biggest advantage of this method is that the network can independently find out the mapping rules of the given samples through training, which saves the process of data analy...

AI Technical Summary

Problems solved by technology

[0004] (1) It is difficult to obtain an accurate mathematical model of ship motion at sea

Due to the interference of the natural environment such as wind, waves and currents on the movement of the ship at sea, the role of propeller and rudder propulsion and control force, the change of the ship's motion characteristics caused by the environmental load and the change of the ship's own load, it is difficult to establish a Accurate Ship Motion Mathematical Model for Ship Motion Prediction

[0005] (2) The model cannot reflect the influence of time-varying factors on ship motion

[0006] (3) Lack of a comprehensive online prediction model for the six-degree-of-freedom motion of ships at sea

The six-degree-of-freedom ship motion prediction model has realistic requirements for the safe operation and manipulation of ships at sea, but previous research lacks a comprehensive and universal six-degree-of-freedom online motion prediction model

However, when a ship is sailing at sea, it is affected by environmental factors such as wind, waves, and currents, and the swaying motion of the ship has complex characteristics such as nonlinearity and dynamic time-varying, which increases the difficulty of its prediction.

Images