New identification algorithm for ship model parameters based on sinusoidal function processing innovation

Abstract

ObjectiveWhen few ship test data samples are available, it is difficult to identify ship model parameters quickly and accurately.MethodsOn the basis of the original least-square system identification algorithm, an innovative sinusoidal function process is introduced, and a new ship model parameter identification method based on the nonlinear sinusoidal function is proposed.The ship Yukun, a teaching and training ship of Dalian Maritime University, is selected for the identification experiment. With only 26 test data samples, the identification effects of the original least square method and improved least square algorithm are compared. ResultsThe simulation results show that the parameter identification accuracy of the algorithm is improved by about 15%, and the effectiveness of the algorithm is verified using the ship Yupeng.ConclusionThis algorithm provides valuable references for the parameter identification of ship models with few test data samples.

Keywords

• ship model
• parameter identification
• nonlinear innovation
• least square method