在环境风、浪的作用下，船舶在泊船过程中的运动姿态会发生显著变化，导致遭遇角出现偏差，从而无法按预期调整航向，进而降低泊船过程的安全性。因此，提出基于运动姿态数据挖掘的泊船遭遇角控制方法。利用FP-growth算法挖掘出运动姿态与遭遇角控制间的关联关系。通过RBF神经网络构建泊船系统函数，由此使用Nomoto传递函数描述船舶航向响应特性，并将其输入至三阶闭环增益成形算法中，通过增益成形实时调整船舶运动状态，抑制外部环境干扰造成的遭遇角偏差，实现泊船遭遇角的精准控制。结果表明，所提方法各个时刻遭遇角控制结果与实际期望控制值高度吻合，可以稳定船舶的运动姿态，保证泊船过程的安全性。
Under the influence of environmental wind and waves, the motion posture of the ship during berthing will undergo significant changes, resulting in deviation of the encounter angle and inability to adjust the heading as expected, thereby reducing the safety of the berthing process. Therefore, a ship encounter angle control method based on motion attitude data mining is proposed. Use the FP growth algorithm to mine the correlation between motion posture and encounter angle control. Based on the obtained correlation, a berthing system function is constructed using an RBF neural network. The ship heading response characteristics are described using the Nomoto transfer function, which is then input into a third-order closed-loop gain shaping algorithm. Through gain shaping, the ship's motion state is adjusted in real time to suppress the encounter angle deviation caused by external environmental interference and achieve precise control of the berthing encounter angle. The experimental results show that the proposed method's encounter angle control results at each moment are highly consistent with the actual expected control values, which can stabilize the ship's motion posture and ensure the safety of the berthing process.
2025,47(13): 162-166 收稿日期：2024-7-13
DOI：10.3404/j.issn.1672-7649.2025.13.028
分类号：P731.2
基金项目：山东省自然科学基金面上项目（ZR2024MF024）
作者简介：李宗锋(1981-),男,硕士,副教授,研究方向为信息管理、数据挖掘
参考文献：
[1] ZHANG K , HUANG L , HE Y , et al. A real-time multi-ship collision avoidance decision-making system for autonomous ships considering ship motion uncertainty[J]. Ocean engineering, 2023, 278(6): 114205.1-114205.19.
[2] 陶力, 周凡. 基于动态时间规整的尾气估算船舶燃油硫含量方法[J]. 传感技术学报, 2023, 36(5): 769-775.
[3] KIM I T , KIM S , PAIK K , et al. Free-running CFD simulations to assess a ship-manoeuvring control method with motion forecast in waves[J]. Ocean Engineering, 2023, 271(3): 1.1-1.9.
[4] 苏文学, 孟祥飞, 张强. 输入饱和约束下自适应RBF神经网络非线性反馈船舶航向控制[J]. 上海海事大学学报, 2024, 45(2): 14-19.
[5] 黄立文, 刘进来, 贺益雄, 等. 考虑舵机延时的船舶最优航向控制器设计[J]. 武汉理工大学学报, 2023, 45(8): 60-67.
HUANG L W, LIU J L, HE Y X, et al. Design of ship optimalcourse controller considering the delay of steering gear[J]. Journal of Wuhan University of Technology, 2023, 45(8): 60-67.
[6] 王伟, 王勇, 周晨光, 等. 基于模糊神经网络PID的无人艇航向控制器研究[J]. 合肥工业大学学报(自然科学版), 2023, 46(4): 458-462.
WANG W, WANG Y, ZHOU C G, et al. Research on unmanned surface vehicle heading controller based on fuzzy neural network PID[J]. Journal of Hefei University of Technology(Natural Science), 2023, 46(4): 458-462.
[7] 乔阳阳, 王丽娟. 数据点位置并行FP-Growth挖掘算法仿真[J]. 计算机仿真, 2023, 40(5): 501-505.
QIAO Y Y, WANG L J. Simulation of parallel FP growth mining algorithm for data point location[J]. Computer Simulation, 2023, 40(5): 501-505.
[8] 姜建武, 王博. 高维数据组合关联关系挖掘方法[J]. 科学技术与工程, 2023, 23(4): 1615-1624.
JIANG J W, WANG B. Combinatorial association mining method for high-dimensional data[J]. Science Technology and Engineering, 2023, 23(4): 1615-1624.
[9] 陈辉, 刘建湖. 舰船水平向冲击环境测量仪研制[J]. 中国测试, 2024, 50(5): 100-105.
CHEN H, LIU J H. Instrument development for measuring horizontal shock environment of ships[J]. China Measurement & Testing Technology, 2024, 50(5): 100-105.
[10] 常波, 杨鑫, 郑建朋. 应用RBF模型的高压熔断器缺相运行状态多点监测[J]. 电子器件, 2024, 47(3): 804-808.
[11] 王欢, 曾庆华, 张宗宇, 等. 基于改进PSO优化RBF的压力扫描阀温度补偿研究[J]. 传感技术学报, 2023, 36(3): 449-455.
[12] 高伟男, 杨涛, 柴天佑. 基于自适应动态规划和梯度下降法的自适应最优输出调节[J]. 控制与决策, 2023, 38(8): 2425-2432.
[13] 李从跃, 胡以怀, 沈威, 等. 基于小波变换和CNN的船用机械故障诊断[J]. 中国测试, 2024, 50(3): 183-192.
[14] 苏义鑫, 公成龙, 张丹红. 考虑推进器饱和特性的动力定位船舶递归滑模动态面控制[J]. 振动与冲击, 2023, 42(8): 206-214.
[15] 赵洁, 任晋宇. 混合海浪作用下无人船泊船姿态自动控制方法[J]. 舰船科学技术, 2022, 44(12): 71-75.
ZHAO J, REN J Y. Automatic attitude control method of unmanned ship under mixed wave[J]. Ship Science and Technology, 2022, 44(12): 71-75.
