基于SVM的2PTMC多分类算法在船舶柴油机故障诊断中的应用

Application of SVM-based 2PTMC multi-classification algorithm in fault diagnosis of marine diesel engine

文章针对船舶柴油机故障特征变得越来越复杂,且大量的故障样本难以获得的状况,引用基于小样本的支持向量机算法(SVM)分类器。采用基于SVM的2PTMC算法,该方法根据故障优先级的不同,将不同类故障逐层分类,相比于传统的一对一(OVO)和一对多(OVA)多分类策略,该方法具有模型简单、重复次数较少的优点。文章构建了一个四级SVM分类器,结果表明,该方法适用于船舶柴油机故障分类。

In this paper,the fault characteristics of marine diesel engines are becoming more and more complex,and a large number of fault samples are difficult to obtain.The Swpport Vector Machine(SVM)classifier based on small samples is cited.The SVM based 2PTMC algorithm is adopted.This method classifies different types of faults layer by layer according to different fault priorities.Compared with the traditional OVO and OVA muti-classification strategies,this method has the advantages of simple model and fewer times.This paper constructs a four-level SVM classifier,and the results show that the method is suitable for ship diesel engine fault classification.