In order to precisely control ships, her engines and so on, it is very important to previously estimate the behaviors (output signals) of these objectives when the manipulation action (input signal) was given to the relevant objects. In other words, it is necessary to have some models which can simulate the behaviors, that is, the characteristics, of the object systems. Usual models such as a mathematical one, or others, however, seem to be unsuitable in the case of the field of ship engineering because the characteristics are complex, and sometimes have the characteristics of time-varying as well as nonlinear dynamic. New techniques such as neural network (NN) is said to be superior to simulate such complex characteristics. Therefore, the ability of NN, especially, the ability of adaptiveness for time-varying characteristics was investigated by using frequency response method. The results have shown that the ability of NN is enouth against the relatively slow time-varying object systems in the field of ship engineering.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Applications of neural network for ships


    Additional title:

    Einsatz von neuronalen Netzen auf Schiffen


    Contributors:
    Morita, S. (author)


    Publication date :

    1995


    Size :

    6 Seiten, 11 Bilder, 3 Quellen


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    Development of a neural network-based controller for ships

    Grigoryev, Vladislav / Rauh, Andreas / Aschemann, Harald et al. | Tema Archive | 2010




    A Neural Network Controller for Optimal Guidance of Surface Ships

    Burns, R. S. / Polkinghorne, M. N. | British Library Conference Proceedings | 1993


    Neural network-based fuel consumption estimation for container ships in Korea

    Le, Luan Thanh / Lee, Gunwoo / Park, Keun-Sik et al. | Taylor & Francis Verlag | 2020