基于多传感器信息融合的船舶动力机械设备故障自动化监测方法  

Automated Monitoring Method for Faults in Ship Power Machinery Equipment Based on Multi-sensor Information Fusion

作　　者：郑钧 盛善智[2] ZHENG Jun;SHENG Shanzhi(Yantai Salvage Bureau of the Ministry of Transport,Yantai 264000,China;School of Ocean,Yantai University,Yantai 264000,China)

机构地区：[1]交通运输部烟台打捞局,烟台264000 [2]烟台大学海洋学院,烟台264000

出　　处：《自动化与仪表》2024年第12期114-118,共5页Automation & Instrumentation

摘　　要：对船舶动力机械设备和姿态进行了研究,采用了多传感器信息融合与分类算法的关键技术和方法,对采集到的信息进行融合和分类,实现故障信号的检测和识别。以船体为参考点的坐标系,描述船体在运动中的位置、速度和加速度等状态。经实验测试,电机正常状态、电机故障和齿轮箱故障的特征频率分别为180 Hz、610 Hz和505 Hz。相比于单传感器测量数据,数据融合后可将故障识别率显著提升,平均识别率达到了95.6%。The equipment and attitude of ship power machinery are studied.The key technology and method of multi-sensor information fusion and classification algorithm are adopted to fuse and classify the collected information and realize the detection and recognition of fault signals.The position,velocity and acceleration of the ship in motion are described with the coordinate system of the ship as the reference point.The characteristic frequencies of motor normal state,motor fault and gearbox fault are 180 Hz,610 Hz and 505 Hz respectively.Compared with the single sensor measurement data,the fault recognition rate can be significantly improved after data fusion,and the average recognition rate is 95.6%.

关 键 词：船舶 机械设备 故障自动化监测 多传感器 信息融合 

分 类 号：TH165[机械工程—机械制造及自动化]