On the use of neural network techniques for the identification of ship stability parameters at sea
- Memorial Univ. of Newfoundland, St. John`s, Newfoundland (Canada). Faculty of Engineering and Applied Science
In this work, neural network techniques are used to identify the stability parameters for a ship sailing in a random sea. The random decrement is calculated from the random roll response. This equation has been shown to resemble the differential equation describing the free rolling motion. The nonlinearities in the free role equation in addition to the linear damping term are lumped in one nonlinear function, F({phi},{dot {phi}}), in the role angle, {phi}, and velocity, {dot {phi}}. A feedforward network with a single hidden layer is then used to identify this general function. The function, F({phi},{dot {phi}}) can be used to identify the parameters in the righting moment using regression techniques. An example for applying this technique using model experiments for a series 60 block 60 model is presented. The agreement between curves predicted using neural network techniques and the actual curves is excellent.
- OSTI ID:
- 194298
- Report Number(s):
- CONF-950695-; ISBN 0-7918-1308-8; TRN: IM9611%%139
- Resource Relation:
- Conference: International conference on offshore mechanics and arctic engineering, Copenhagen (Denmark), 18-22 Jun 1995; Other Information: PBD: 1995; Related Information: Is Part Of OMAE 1995 -- Proceedings of the 14. international conference on offshore mechanics and arctic engineering. Volume 2: Safety and reliability; Guedes Soares, C. [ed.] [Technical Univ. of Lisbon, Lisboa (Portugal)]; PB: 567 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
Stochastic gradient method for training of a class of recurrent neural nets
Simulation of dynamic processes with adaptive neural networks