基于NAR神经网络的船舶运动姿态短期预测

Short-term Prediction of Ship Motion Attitude Based on NAR Neural Network

船舶在海上作业的过程中,不可避免地会受到风浪的扰动影响,风浪扰动可以分解到船舶运动的六个自由度上,而船舶的六自由度运动是一个复杂的非线性过程。借助预测算法可以对船舶短时间后的运动状态进行预测,从而更好辅助在船舶上的工作活动。为了提升船舶运动姿态的预测精度,建立了非线性自回归(NAR)神经网络模型,并利用NAR模型对船舶运动姿态进行预测仿真,将仿真结果与AR预测法的结果进行对比。仿真结果分析表明,基于NAR神经网络模型的预测算法与传统的基于AR模型的预测算法相比,精度更高,更具有实用价值。

In the process of offshore operations,the ship will be affected by the disturbance of wind and waves inevitably.Wind waves can be decomposed into six degrees of freedom of movement,and the six-degree-of-freedom movement of a ship is a complex nonlinear process.The prediction algorithm can predict the motion state of the ship after a short time,so as to better assist the working activities on the ship.In order to improve the prediction accuracy of the ship's motion attitude,a nonlinear autoregres⁃sive(NAR)neural network model is established,and the NAR model is used to predict the ship's motion attitude,and the simula⁃tion results are compared with the results of the AR prediction method.The simulation results show that the prediction algorithm based on NAR neural network model has higher precision and more practical value than the traditional AR model based prediction algorithm.