Machine learning-based wavelet method for estimating the parameters in ship roll damping models using Legendre polynomials

Abstract

ABSTRACT In this paper, an efficient Machine Learning-based Legendre wavelet method (ML-LWM) is utilised to estimate the parameters in ship damping models. Damping is critical for the roll motion response of a ship in waves. To the best of our knowledge, there is no Legendre Wavelet Method (LWM) has been reported to estimate the damping and restoring moments in ship dynamics models. LWM is applied to estimate roll angle, damping coefficients and restoring moments. Some numerical examples are given to demonstrate the validity and applicability of the proposed ML-LWM. The ML-LWM results are compared with the results obtained by the Homotopy Perturbation Method (HPM). Also, the proposed results are validated with the experimental data. Satisfactory agreement with experimental and HPM is noticed. The efficiency of the proposed wavelet method is confirmed by CPU runtime.