基于核主成分的船舶柴油机磨损信息特征提取方法研究  

Research on Feature Extraction Method of Marine Diesel Engine Wear Information Based on Kernel Principal Component

作　　者：石新发 贺石中 谢小鹏[2] 崔策 孙宇航 SHI Xinfa;HE Shizhong;XIE Xiaopeng;CUI Ce;SUN Yuhang(Guangzhou Mechanical Engineering Research Institute Co.Ltd.,Guangzhou 510530,China;School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou 510640,China;National and Local Joint Engineering Center of Industrial Friction and Lubrication Technology,Guangzhou 510530,China)

机构地区：[1]广州机械科学研究院有限公司,广州510530 [2]华南理工大学汽车与机械工程学院,广州510640 [3]工业摩擦润滑技术国家地方联合工程研究中心,广州510530

出　　处：《武汉理工大学学报（交通科学与工程版）》2022年第6期1039-1043,共5页Journal of Wuhan University of Technology（Transportation Science & Engineering）

基　　金：广东省科技计划项目(2020B1212070022);广州机械科学研究院有限公司博士后专项(17300065)。

摘　　要：文中基于某型船舶柴油机跟踪监测获取的55个油样磨损监测信息,应用核主成分理论对以上油样原子发射光谱数据和直读铁谱数据进行特征提取,使原始信息从8维压缩至2维,建立了特征信息描述模型,基于所选择的两阶主成分得分和统计界限值,对55个油样所对应的柴油机磨损状态进行了评价,并进行了状态区域划分,实现了磨损故障的分类诊断与识别,结合历史检修记录及分析式铁谱、X荧光能谱和PQ数据,对分类与识别结果进行分析,进一步验证了模型有效性.Based on the wear monitoring information of 55 oil samples obtained from the tracking monitoring of a certain marine diesel engine,the atomic emission spectrum data and direct-reading ferrography data of the oil samples were extracted by using the kernel principal component theory,and the original information was compressed from 8 dimensions to 2 dimensions,and the characteristic information description model was established.Based on the selected two-order principal component scores and statistical limit values,the wear states of 55 oil samples were evaluated,and the state regions were divided,thus realizing the classification,diagnosis and identification of wear faults.Combined with historical maintenance records,analytical ferrography,X-ray fluorescence spectrum and PQ data,the classification and identification results are analyzed,and the validity of the model is further verified.

关 键 词：船舶柴油机 磨损故障 油液监测 核主成分 特征提取 

分 类 号：U672[交通运输工程—船舶及航道工程]