Ship manoeuvrability prediction using neural networks

Wang, Yie (1996) Ship manoeuvrability prediction using neural networks. Masters thesis, Memorial University of Newfoundland.

[img] [English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (6Mb)
  • [img] [English] PDF - Accepted Version
    Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.
    (Original Version)

Abstract

This thesis is divided into three parts. The first two parts deal with two different methods for predicting the manoeuvring characteristics of ships using a neural network technique. The third part deals with the application of the random decrement concept to the coupled sway-yaw motions. -- In the first part of this thesis, a new predictive method is presented for the estimation of the hydrodynamic characteristics of a ship performing certain standard manoeuvres. This method uses the static neural network technique to predict the nonlinear hydrodynamic forces of the ship during its motion in the horizontal plane. The neural network model uses a steepest descent search to find the neural network weights. In this thesis, a back propagation algorithm is used to calculate the slope of the sum-of-squared-error curve as a function of the different weights. Data for training the neural network consists of the data from a 35-35 degree zigzag manoeuvre. Surge, sway, yaw velocities and rudder angles are used as input to the predictive model. The target output data are the lumped nonlinear hydrodynamic functions. -- The generalization of the trained neural network model is checked by simulating the manoeuvres of the ship in a situation different from the one used in the training of neural network. A moderate 20-20 degree zigzag manoeuvre, a 25 degree turning (starboard) and a 20 degree Dieudonne spiral manoeuvre are selected to check the validity of the neural network model. -- In the second part of this thesis, another approach to predict ship turning manoeuvres is proposed. This model maps the relationship between sway velocities and yaw rates during the circular manoeuvre using a neural network technique. This method reduces the number of equations to be used in the prediction to a single yaw equation. This new yaw equation can then be used for predicting turning manoeuvres. -- In the last part of the thesis work, the extension of the random decrement approach to the nonlinear sway-yaw motions is presented. The random waves are simulated based on the ITTC spectrum formula. The linear system and the nonlinear system of sway and yaw motion equations are discussed. The autocorrelation functions of the response of sway and yaw velocities in random waves are obtained. A method for using these functions to identify the hydrodynamic characteristics of the coupled sway-yaw motions is suggested.

Item Type: Thesis (Masters)
URI: http://research.library.mun.ca/id/eprint/6577
Item ID: 6577
Additional Information: Bibliography: leaves 119-122.
Department(s): Engineering and Applied Science, Faculty of
Date: 1996
Date Type: Submission
Library of Congress Subject Heading: Ships--Maneuverability--Computer simulation; Neural networks (Computer science)

Actions (login required)

View Item View Item

Downloads

Downloads per month over the past year

View more statistics