在船舶航行与海上作业中，船舶升沉运动受复杂海况影响呈现显著非平稳特性，传统AR模型难以满足复杂场景下的升沉运动预测需求，开展高精度预测方法研究具有重要意义。本文首先构建自适应AR模型，通过引入遗忘因子实现参数动态更新，有效应对海况突变时的预测偏差。在此基础上进一步构建带外部输入的ARX模型，融合海浪预报数据以增强建模，并结合误差驱动机制动态调整遗忘因子。最后通过仿真对模型进行验证，构建的ARX模型显著提升了船舶升沉运动的预测精度与复杂海况下的鲁棒性。
In ship navigation and offshore operations, the heave and sink movements of ships show significant non-stationary characteristics affected by complex sea conditions. Traditional AR models are difficult to meet the prediction requirements of heave and sink movements in complex scenarios. Therefore, it is of great significance to carry out research on high-precision prediction methods. This paper first constructs an adaptive AR model and realizes the dynamic update of parameters by introducing a forgetting factor, effectively dealing with the prediction deviation when the sea conditions change suddenly. On this basis, the ARX model with external input is further constructed. The wave forecast data is integrated to enhance the modeling, and the forgetting factor is dynamically adjusted in combination with the error-driven mechanism. Finally, the model was verified through simulation. The constructed ARX model significantly improved the prediction accuracy of the ship's heave and sink motion and the robustness under complex sea conditions.
2025,47(15): 177-180 收稿日期：2025-3-30
DOI：10.3404/j.issn.1672-7649.2025.15.030
分类号：U662.2
基金项目：山东省教育规划课题（2023ZC391）
作者简介：颜鲁晓(1992-),女,硕士,讲师,研究方向为数学与应用数学
参考文献：
[1] 赵所, 林立, 李震, 等. 基于BP神经网络的甲板运动预报与补偿技术[J]. 北京航空航天大学学报, 2024, 50(9): 2772-2780.
[2] 黄彦文, 褚德英, 谢颖. 基于船舶运动数据的AR(p)模型参数估计和适用性验证[J]. 船舶工程, 2022, 44(S1): 500-504.
[3] 王允峰. 船舶纵横摇和升沉运动预报方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2010.
[4] 刘东, 武裕鑫, 孙树政. 基于不同海浪谱船舶极限海况运动统计特征值预报分析[J]. 舰船科学技术, 2024, 46(9): 60-65.
[5] 唐刚, 唐溥, 邵辰彤, 等. 基于IPESN的船舶升沉运动预报方法[J]. 船舶工程, 2021, 43(4): 43-47.
[6] 周利, 段玉响, 任政儒, 等. 主动式升沉补偿控制系统及运动预报[J]. 华中科技大学学报(自然科学版), 2021, 49(3): 98-104.
[7] 张大兵, 彭智力, 段江哗, 等. 基于混沌理论和改进极限学习机的船舶升沉预报(英文)[J]. 船舶力学, 2021, 25(10): 1322-1330.
