Neural Network Application on Ship Motion Prediction

The swing motion of the ship, which is mainly caused by sea wind and sea wave while on cruise, has significant impacts on completing the task of the ship. The paper analyzed the six degree of freedom ship motion. Put forward the characteristic of time series about ship motion methods. Based on the Back-Propagation neural network which has the characteristic that it can approach any nonlinear function in theory. Established the neural network model which is used to analyze the time series, and provided a Back-Propagation algorithm. Calculated an example. The calculation and results show that neutral network prediction is better than time series analysis. Therefore, it provides a new method for short-term ship motion prediction.