船舶螺旋桨推力的在线高精度辨识方法  

Online high-precision identification method for ship propeller thrust

作　　者：马相龙 吴昊[1] 邹冬林 薛林[2,3] 塔娜[2,3] 饶柱石 MA Xianglong;WU Hao;ZOU Donglin;XUE Lin;TA Na;RAO Zhushi(Shanghai Marine Equipment Research Institute,Shanghai 200031,China;State Key Laboratory of Mechanical System and Vibration,Shanghai Jiao Tong University,Shanghai 200240,China;Institute of Vibration,Shock and Noise,Shanghai Jiao Tong University,Shanghai 200240,China;Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration,Shanghai 200240,China)

机构地区：[1]上海船舶设备研究所,上海200031 [2]上海交通大学机械系统与振动全国重点实验室,上海200240 [3]上海交通大学振动、冲击、噪声研究所,上海200240 [4]高新船舶与深海开发装备协同创新中心,上海200240

出　　处：《中国舰船研究》2024年第3期310-318,共9页Chinese Journal of Ship Research

基　　金：国家自然科学基金资助项目(52375110);中国船舶集团有限公司–上海交通大学海洋装备前瞻创新联合基金资助项目(GC3270001/022);中央高校基本科研业务费专项基金资助项目(AF0200296)。

摘　　要：[目的]螺旋桨推力在轴系上产生的微弱形变信号易被干扰噪声所淹没,导致难以准确测量其推力,故提出一种在低信噪比下仍具有很高精度的螺旋桨推力测量方法。[方法]综合考虑测量数据误差的影响以及数据中潜藏的力学机制,基于Kalman滤波结合推力–位移状态方程建立轴系推力辨识模型,并以恒定转速、变转速及低频波动转速3种工况为例开展轴系推力测量研究,分析不同信噪比下该方法的推力辨识精度与鲁棒性。[结果]研究结果表明,当信噪比仅为20 dB时,推力辨识的最大相对误差仅为3.56%,验证了其在低信噪比下的高精度辨识能力与鲁棒性。[结论]研究成果可为实船环境下螺旋桨推力及轴系状态的在线实时高精度监测提供参考。[Objective]The weak deformation signal generated by propeller thrust on a shaft system is easily overshadowed by interfering noise,making it challenging to accurately measure the thrust.Therefore,this study proposes a new method for accurately measuring the thrust of propellers under low signal-to-noise ratio conditions.[Methods]By taking into consideration the influence of measurement data errors and the under-lying mechanical mechanisms,a thrust identification model is established based on a combination of Kalman filtering,and the thrust-displacement state equation.Thrust measurement research is conducted under three dif-ferent operating conditions:constant rotational speed,variable rotational speed and low-frequency fluctuating rotational speed.The precision and robustness of the proposed method are then analyzed under different signal-to-noise ratios.[Results]The results demonstrate that even at a signal-to-noise ratio of only 20 dB,the maxi-mum relative error in thrust identification is only 3.56%.Thus,the proposed method exhibits highly accurate and robust thrust identification performance under low signal-to-noise ratio conditions.[Conclusion]The findings of this study can provide valuable insights for the real-time and high-precision monitoring of pro-peller thrust and shaft system status in actual ship environments.

关 键 词：KALMAN滤波 推力辨识 推进轴系 位移测量 在线监测 

分 类 号：U664.21[交通运输工程—船舶及航道工程]