Real-time online forecasting model of ship rolling motion based on chaotic online LSSVM

Considering the chaotic characteristics of ship rolling time series, the new real-time forecasting method is proposed to further enhance accuracy and real-time of the prediction model based on support vector machines in the prediction of ship rolling motion, which utilizes phase space reconstruction theory of chaotic systems and online LSSVM. Delay time and delay time window are estimated by C-C method, and then the chaotic online LSSVM real-time prediction model is established. The experiments of ship rolling time series prediction are done. The simulation results indicate the real-time prediction method can more effectively improve the convergence rate and the prediction precision and extend prediction time at the same time, which is compared to the combination prediction model based on support vector machine and neural network.