支持向量机是一种基于统计学习理论的机器学习算法,它能在训练样本很少的情况下达到很好的分类效果。本文以双螺杆挤出机为例,介绍了基于支持向量机的多故障分类器,探讨了“成对分类”与“一类对多类”两种多类分类算法的应用。诊断实例表明,基于支持向量机的多故障分类器对设备故障具有很好的分类效果。 SVM is a machine learning algorithm based on statistical learning theory, which can achieve good classification results with few training samples. In this paper, a twin-screw extruder is taken as an example to introduce a multi-fault classifier based on support vector machines. The application of two kinds of multi-class classification algorithms, “paired classification” and “one class to multiple classes”, is discussed. The diagnostic examples show that SVM-based multi-fault classifier has a good classification effect on equipment failures.