NEURAL-NET COMPUTING FOR DYNAMIC POSITIONING OF VESSELS AT SEA

Dynamic positioning or station-keeping is a technique which uses ship-mounted propulsors and lateral thrusters to counteract environmental forces acting on the vessel due to wind, wave and current, thereby maintaining it as closely as possible at some desired position in the horizontal plane. The authors suggest that it may be useful to train a neural-net so that it constitutes a neural network model of the action of the waves on the ship. It can then be used to map measured wave motion into drift forces exerted by the waves on the ship. Feedforward control can then be used to counteract those forces. The feasibility of this approach using hydrodynamically calculated wave drift forces for network training is discussed. The net was indeed able to act as a wave force estimator. Dynamic positioning was demonstrated with simulation. A training data acquisition scheme is also described.

• Supplemental Notes:
  • Hydrodynamics: Computations, Model Tests and Reality; ARIN Workshops on: Advanced Vessels Station Keeping, Propulsor-Hull Interaction and Nautical Simulators; 11-14 May 1992; Wageningen, The Netherlands. Pubs by Elsevier Sc. Pubs. Procs. Workshop B, p 329 [15 p, 12 ref, 6 tab, 5 fig]
• Authors:
  • Gu, M X
  • Pao
  • Y-H Yip, P P
• Publication Date: 1992

Language

• English

Subject/Index Terms

• TRT Terms: Drift; Dynamic positioning; Mathematical models; Ship operations
• Uncontrolled Terms: Numerical models
• Old TRIS Terms: Drift forces; Station keeping
• Subject Areas: Geotechnology; Marine Transportation;

Filing Info

• Accession Number: 00704515
• Record Type: Publication
• Source Agency: British Maritime Technology
• Files: TRIS
• Created Date: Aug 14 1995 12:00AM