基于支持向量机的柴油机磨损模式识别方法  

Wear Mode Recognition Method of Diesel Engines Based on Support Vector Machine

作　　者：杨绍卿[1] 王宪成[1] 赵文柱[1] 王杰 

机构地区：[1]装甲兵工程学院机械工程系,北京100072 [2]辽宁石油勘探局振兴公用事业公司,盘锦124010

出　　处：《四川兵工学报》2015年第8期96-99,共4页Journal of Sichuan Ordnance

摘　　要：针对现有柴油机磨损模式判别方法中存在的不足,将支持向量机算法应用到柴油机磨损模式的识别中,建立了基于支持向量机的柴油机磨损模式判断模型,并对部分实验柴油机油液样本进行了评估,并与广义贴近算法、模糊聚类算法和专家评判结果进行了比较,证明了支持向量机能够准确、有效地识别柴油机磨损模式。In view of the existing shortcomings on the course of wear mode identifying method on diesel en- gines, the support vector machine （SVM） algorithm was applied to the wear mode identifying. The recog- nition mode for the wear of diesel engine which was based on the SVM algorithm was established, and sev- eral oil samples of experimental diesel engines were estimated. The result was compared with results of generalized closeness degree algorithm, fuzzy clustering algorithm and expert assessment, and we proved that the wear mode of diesel engine can be effectively and accurately recognized by the SVM method.

关 键 词：支持向量机 磨损模式 柴油机 油液分析 

分 类 号：TK421.2[动力工程及工程热物理—动力机械及工程] TJ8[兵器科学与技术—武器系统与运用工程]