次序二叉树支持向量机多类故障诊断算法研究

Multi-class fault diagnosis based on support vector machines with sequenced binary tree archtecture

构建二叉树支持向量机时,如果随机地将分类器分布在二叉树的各个结点上,是不能充分发挥其性能的。考虑到样本的分布情况对分类器推广能力具有较大影响,提出一种次序二叉树支持向量机多类算法,采用样本分布半径和样本分布距离估算各个类别的样本在高维特征空间中的分布情况,把分布半径较大的类别或者分布距离较大的类别较早地分出来,并且在特征空间中给其划分较大的分类区域。转子多故障诊断实验表明,该算法的诊断速度快,故障识别率高,推广能力强,更加适合于实际故障诊断应用。

If classifiers of a support vector machine with binary tree architecture are arrayed randomly in the binary tree,their performance is not the best.A sequenced method in consideration of the sample range was proposed to rationally array classifiers of a support vector machine with binary tree architecture.A sample distribution radius and a sample distribution distance were introduced to estimate sample range of all classes in high-dimension characteristic space.The classes with bigger sample range were classified earlier in the higher nodal point of the binary tree architecture,and were given wider classificatory areas in the characteristic space.The experiment of multi-class fault diagnosis of a rotor showed that the proposed method distinctly improves the fault recognition accuracy,the diagnosis speed and the generalization,and it is suitable for practical application of multi-class fault diagnosis.