摘要
针对舰船航态的非线性特征，提出了采用非线性自回归模型描述舰船海上航行极短期运动过程的新方法利用正交化的手段得到了辅助模型，进而通过对辅助模型的辨识推导出舰船航态超前多步预报器仿真结果表明本文提出的预报方法比线性自回归模型描述的预报方法有效，能提高预报精度、加长预报时间另外，本文给出的非线性预报器的辨识方法，可以加快预报时计算速度。
Being aimed at characters of navigating posture of warship,a new method is presented in this paper.This method can describe extremely short time motion process for warship at sea by using the nonlinear autoregressive (NAR) model.An auxiliary model is obtained by using orthogonalization and a multi-step forward prediction is derived through identification of this model.Simulation results show that the method is more effective than linear autoregressive (AR) model,for it improves the prediction accuracy and enlarges the prediction time range.Moreover,the identification method of nonlinear predictor presented in the paper can speed up computing during predicting and is suitable for real prediction.
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
1996年第4期1-7,共7页
Journal of Harbin Engineering University
关键词
舰船
极短期预报
自回归模型
航行姿态
navigating posture for warship
extremely short time prediction:nonlinear autoregressive model
orthogonalization
