基于两阶段学习的半监督SVM故障检测方法

PDF(1689 KB)

振动与冲击 ›› 2012, Vol. 31 ›› Issue (23) : 39-43.

论文

基于两阶段学习的半监督SVM故障检测方法

陶新民1 曹盼东1 宋少宇1 付丹丹1

作者信息 +

Fault Detection based on Two Stage Learning Semi-Supervised SVM

TAO Xin-min1, CAO Pan-dong 1, SONG Shao-yu 1 , Fu Dan-dan 1

Author information +

文章历史 +

摘要

本文提出了一种基于两阶段学习的半监督SVM故障检测方法。该方法首先使用标识传递算法给未标识样本赋予初始伪标识，并通过近邻图对比样本点标识值，将可能是噪声的样本点识别并剔除；然后将去噪处理后的样本集输入到SVM中，使得SVM在训练时能兼顾整个样本集的信息，从而提高SVM的故障检测性能。实验中将本文方法同支持向量机(SVM)、模糊支持向量机(FSVM)、直推式支持向量机(TSVM)及拉普拉斯支持向量机算法(LapSVM)进行比较，结果表明本文方法在不同数目标识样本集合的情况下，检测精度较其他算法有较大幅度提高，同时本文方法还比较了不含测试样本和含测试样本训练条件下的故障检测性能，结果表明结合测试样本可进一步提高算法的故障检测性能。

Abstract

A novel semi-supervised Support vector machine fault detection model based on two stage learning is presented in this paper. First of all, propagation algorithm is applied to provide pseudo labels for the unlabeled samples. Secondly k-near-neighbor graph is utilized to distinguish and delete the noisy samples. Then the denoised samples are input into the support vector machine(SVM), so that the global information of the whole samples can be considered by SVM to enhance the fault detection accuracy. The comparisons with Support vector machine, Fuzzy Support vector machine, Transductive support vector machine and Laplacian support vector machine fault detetion algorithm are performed.The experiments show that the proposed approach has a more substantial detection accury than other algorithms under different labeled sample sets. The proposed fault detection methods with test samples and without test samples are compared.The results illustrate the investigated techniques with test samples as unlabeled samples can outperform the one without test samples as unlabeled samples.

导出引用

陶新民曹盼东宋少宇付丹丹. 基于两阶段学习的半监督SVM故障检测方法[J]. 振动与冲击, 2012, 31(23): 39-43

TAO Xin-min;CAO Pan-dong;SONG Shao-yu ;Fu Dan-dan . Fault Detection based on Two Stage Learning Semi-Supervised SVM [J]. Journal of Vibration and Shock, 2012, 31(23): 39-43