节点文献
基于模糊支持向量机的多级二叉树分类器的水轮机调速系统故障诊断
FUZZY SVM-BASED MULTILEVEL BINARY TREE CLASSIFIER FOR FAULT DIAGNOSIS OF HYDROTURBINE SPEED REGULATING SYSTEM
【摘要】 在传统支持向量机(C-SVM)的基础上,通过集成模糊聚类技术和支持向量机算法,构造了一种适合于故障诊断的多级二叉树分类器,并首次应用于水轮机调速系统故障诊断,取得了良好效果。该方法首先利用模糊聚类技术求取每类样本聚类中心,再对各聚类中心逐次二分,从而确定了一棵二叉树,然后在二叉树的每个节点处,根据样本聚类中心把相应样本分成两类,构造出SVM 子分类器。实验结果表明,对于k 类别故障诊断问题,只需构造k-1 个SVM 子分类器,简化了分类器结构,避免了不可区分区域的出现,且节省了内存开销,故障诊断正确率高。
【Abstract】 Based on conventional Support Vector Machine(C-SVM) and through integrating fuzzy clustering technique and SVM algorithm, a multilevel binary tree classifier which is suitable for fault diagnosis task is presented and applied to fault diagnosis task of hydroturbine speed regulating system for the first time. Firstly, this method computes the clustering centers of each class by using fuzzy clustering technique and all clustering centers are divided into two successively. So, a binary tree is constructed. Then, it reconstructs SVM sub-classifier according to the clustering centers and the samples which belong to those clustering centers in each node of binary tree. The experimental results show that, in k-class fault diagnosis task, only k-1 SVM sub-classifiers needs to be constructed. In this way, the structure of the classifier is simplified and the unclassifiable region is avoided. Also, less memory is needed and a high rate of correct fault diagnosis is obtained.
【Key words】 Hydroturbine; Fuzzy clustering; Support Vector Machine (SVM); Binary tree; Hydroturbine speed regulating system; Fault diagnosis;
- 【文献出处】 中国电机工程学报 ,Proceedings of the Csee , 编辑部邮箱 ,2005年08期
- 【分类号】TV738
- 【被引频次】77
- 【下载频次】766