基于Newton法优化ARMA模型参数的船舶升沉运动预测研究

为解决波浪补偿系统中时延现象导致的控制性能下降问题,通过建立Newton-ARMA模型提前预测船舶升沉运动来消除时延现象。首先设计卡尔曼滤波器对船舶升沉运动加速度信号进行降噪滤波处理;然后使用加速度二次积分模块将加速度信号转换为位移信号;最后建立自回归滑动平均(ARMA)模型,并使用牛顿(Newton)法对模型参数进行优化,得到船舶升沉运动的Newton-ARMA预测模型。仿真结果表明,Newton-ARMA模型对船舶升沉运动的预测时间可达10 s,预测误差随着预测时间的增加而增大; Newton-ARMA模型对二级海况、三级海况和四级海况下的船舶升沉运动平均预测精度分别达到89.43%、88.53%以及87.78%,远高于ARMA模型对船舶升沉运动预测的精度,说明采用Newton法优化ARMA模型参数可以显著提高船舶升沉运动的预测精度,也即Newton-ARMA模型对控制波浪补偿系统时延具有较好的补偿效果。

Research on prediction of vessel's heave motion based on Newton method for optimizing ARMA model parameters

In order to solve the problem of control performance degradation caused by the delay phenomenon in the wave compensation system, the Newton-ARMA model is established to predict the vessel''s heave motion in advance to eliminate the delay phenomenon. Firstly, the Kalman filter is designed to reduce the noise of the vessel''s heave motion acceleration signal; then the acceleration quadratic integral module is used to convert the acceleration signal into the displacement signal; finally, the auto-regressive moving average (ARMA) model is established, and Newton method is used to optimize the model parameters, then obtain the Newton-ARMA prediction model for the vessel''s heave motion. The simulation results show that the Newton-ARMA model can predict the vessel''s heave motion up to 10 s, and the prediction error increases with the increase of the prediction time; the Newton-ARMA model has an average prediction accuracy of 89.43%, 88.53% and 87.78% for the level-two sea states, level-three sea states and level-four sea states, which is much higher than the prediction of the vessel''s heave motion by the ARMA model, indicating that using Newton method to optimize the ARMA model parameters can significantly improve the prediction accuracy of the vessel''s heave motion, and the Newton-ARMA model has a better compensation effect on the control wave compensation system delay.