Estimating maneuvering coefficients using system identification methods with experimental, system-based, and CFD free-running trial data

Predicting ship maneuverings in calm water is an important topic: however, it is very expensive to run numerous maneuvering simulations using CFD. Despite its short computation time, system-based simulations need many captive model tests to estimate the hydrodynamic and rudder maneuvering coefficients used in the system-based mathematical model and the results have difficulty to achieve quantitative agreement with experimental data. Herein system identification techniques (extended Kalman filtering and constrained least square method using generalized reduced gradient algorithm) are used to predict the maneuvering coefficients from several EFD, systems based and CFD free-running trials. CFD gives not only the ship motions but also the total and component hydrodynamic forces/moments during the free-running simulations, which is helpful for estimating the maneuvering coefficients. Several types of free-running (turning circle, zigzag and large angle zigzag) tests are conducted using EFD and CFD to examine which free-running trial gives the best maneuvering coefficients using system identification. Also several free-running data are combined together for parallel processing. The set of maneuvering coefficients estimated by the constrained least square method using combined CFD free-running trial data show the most generalized results which covers a wide range of maneuvers. (C) 2012 Elsevier Ltd. All rights reserved.