Bayesian and Neural Networks for Preliminary Ship Design
Authors: Clausen, H. B.; Lützen, M.; Friis-Hansen, A.; Bjørneboe, N.
Source: Marine Technology, Volume 38, Number 4, 1 October 2001 , pp. 268-277(10)
Abstract:To ease the determination of the main particulars of a ship at the initial design stage it is convenient to have tools which, given the type of ship and a few other parameters, output estimations of the remaining dimensions. To establish such a tool, a database of the characteristics of about 87 000 ships is acquired and various methods for derivation of empirical relations are employed. A regression analysis is carried out to fit functions to the data. Further, the data are used to learn Bayesian and neural networks to encode the relations between the characteristics. On the basis of examples, the three methods are evaluated in terms of accuracy and limitations of use. For different types of ships, the methods provide information on the relations between length, breadth, height, draft, speed, displacement, block coefficient and loading capacity. Thus, useful tools are available to the designer when he chooses the preliminary main characteristics of a ship.
- Marine Technology is dedicated to James Kennedy, 1867-1936, marine engineer, and longtime member of the Society, in recognition and appreciation of his sincere and generous interest in furthering the art of ship design, shipbuilding, ship operation, and related activities. The Technical papers in this quarterly flagship journal cover a broad spectrum of research on the latest technological breakthroughs, trends, concepts, and discoveries in the marine industry. SNAME News is packed with Society news and information on national, section, and local levels as well as updates on committee activities, meetings, seminars, professional conferences, and employment opportunities. For access to Volume 47 Issue 2 and later, please contact SNAME
- Information for Authors
- Membership Information
- Volume 47 Issue 2 and later
- ingentaconnect is not responsible for the content or availability of external websites