登录    注册    忘记密码    使用帮助

详细信息

基于多源时频特征融合对称点模式的船舶舵机故障诊断    

Marine Steering Gear Fault Diagnosis Based on Multi-Source Time-Frequency Features Fusion Symmetric Dot Pattern

文献类型:期刊文献

中文题名:基于多源时频特征融合对称点模式的船舶舵机故障诊断

英文题名:Marine Steering Gear Fault Diagnosis Based on Multi-Source Time-Frequency Features Fusion Symmetric Dot Pattern

作者:廖志强[1,2,3];梁观龙[1];刘乔[1,2,3];黄振德[1];宋雪玮[1,2,3];贾宝柱[1,2,3]

机构:[1]广东海洋大学船舶与海运学院,广东湛江524088;[2]广东海洋大学广东省船舶智能与安全工程技术研究中心,广东湛江524088;[3]广东海洋大学省市共建南海海洋牧场智能装备广东省重点实验室,广东湛江524088

年份:2025

卷号:47

期号:5

起止页码:67

中文期刊名:船舶工程

外文期刊名:Ship Engineering

收录:北大核心2023、、北大核心

基金:国家自然科学基金(52201355,52401418,52071090);广东海洋大学科研启动资金计划(060302132304,060302132101);湛江市非基金科学与科学研究项目(2022B01049,2023B01046)。

语种:中文

中文关键词:舵机故障诊断;故障特征增强;多源时频特征融合;对称点模式;改进SE-Res Net18

外文关键词:marine steering gear fault diagnosis;fault feature enhancement;multi-source time-frequency feature fusion;symmetric dot pattern;improved SE-Resnet18

中文摘要:[目的]针对船舶舵机结构复杂、故障形式多样且具有隐蔽性、单一信号源无法充分表征舵机系统状态特征等问题,提出一种基于多源时频特征融合对称点模式(SDP)的船舶舵机故障诊断方法。[方法]使用振动和电流信号监测船舶舵机状态,利用时频融合SDP方法将不同尺度的多源信号转换为二维图像。通过自适应优化算法寻找最优SDP参数,以放大舵机不同故障时图像之间的差异、增强故障特征,最后将图像输入到一种改进的SE-Res Net18网络进行故障诊断。[结果]通过船舶舵机数据试验验证表明,所提方法对船舶舵机故障诊断的准确率达到100%,证明所提方法的可行性和有效性。通过与单一数据源SDP转换方法进行对比表明,所提方法具有最高的舵机故障诊断准确率,验证了所提方法的卓越性。[结论]研究结果可为舵机故障诊断提供一种高效可行的诊断方法。

外文摘要:[Purpose]Aiming at the problems of marine steering gear complex structure,multi fault states,failure features are not obvious and the single signal can not fully express the state features of the marine steering gear,a multisource time-frequency feature fusion symmetric dot pattern(SDP)is proposed for fault diagnosis method of marine steering gear.[Method]Multi-source signals are used to monitor the state of marine steering gear,and converts multi-source signals with different scales into 2-dimensional images using time-frequency fusion SDP.The adaptive optimization algorithm is executed to find the optimal SDP parameters to enlarge the difference between different steering gear faults,so as to enhance the fault features.The SDP images are input to an improved SE-ResNet18 network for fault diagnosis.[Result]Through the experimental verification of the marine steering gear data,the results show that the accuracy of the fault diagnosis of the based on the proposed method is 100%,which proves the feasibility and effectiveness of the method.By comparing with the single signal source experiment,the results show that the proposed method has the highest fault diagnosis accuracy,which validates the excellence of the proposed method.[Conclusion]The proposed method provides an efficient and feasible diagnostic method for marine steering gear.

参考文献:

正在载入数据...

版权所有©广东海洋大学 重庆维普资讯有限公司 渝B2-20050021-8 
渝公网安备 50019002500408号 违法和不良信息举报中心