基于模糊K近邻支持向量数据描述的水电机组振动故障诊断研究

发布时间：2018-01-05 17:01

本文关键词：基于模糊K近邻支持向量数据描述的水电机组振动故障诊断研究　出处：《中国电机工程学报》2014年32期 　论文类型：期刊论文

【摘要】：水电机组振动故障诊断中常面临样本稀缺及分布不均匀、不平衡等问题,严重影响诊断结果。针对此类问题提出一种基于模糊K近邻(K nearest neighbor,KNN)支持向量数据描述(support vector data description,SVDD)的故障诊断模型。首先利用核变换将故障样本映射到高维特征空间,并采用SVDD提取不平衡故障样本域的边界支持向量样本,构建基于相对距离模糊阈值和KNN的决策规则,最终在此基础上建立机组故障诊断模型。用该模型对经过不平衡处理的国际标准测试数据样本进行测试实验,并与支持向量机(support vector machine,SVM)及目前应用较多的SVDD模型的分类结果进行对比,结果表明该模型可有效解决不平衡样本分类倾斜性问题。最后,将模型用于某水电厂机组振动故障诊断,取得了较高的诊断精度,证明了该方法的有效性。
[Abstract]:The vibration fault diagnosis of hydropower units often faces the problems of scarcity of samples, uneven distribution and imbalance. This paper presents a K nearest neighbor based on fuzzy K nearest neighbor. KNN) support vector data description for support vector data description. Firstly, the fault samples are mapped to high dimensional feature space by kernel transformation, and the boundary support vector samples in unbalanced fault sample domain are extracted by SVDD. The decision rules based on relative distance fuzzy threshold and KNN are constructed, and finally, the fault diagnosis model of unit is established. The model is used to test the samples of international standard test data after unbalanced processing. And compared with the classification results of support vector machine support vector machine (SVM) and the SVDD model which is widely used at present. The results show that the model can effectively solve the problem of the classification inclination of unbalanced samples. Finally, the model is applied to the vibration fault diagnosis of a hydropower plant, and a high diagnostic accuracy is obtained, and the effectiveness of the method is proved.
【作者单位】：华中科技大学水电与数字化工程学院;
【基金】：国家自然科学基金项目(51239004,51079057) 高等学校博士学科点专项科研基金(20100142110012)~~
【分类号】：TM312;TP18
【正文快照】： 0引言水电机组故障诊断实质上是一种模式识别问题,即通过已有故障样本建立诊断模型,并以此对未知机组状态进行识别。目前,诊断分类方法主要有:模糊推理[1]、贝叶斯决策[2]、神经网络[3]和支持向量机[4](support vector machine,SVM)等。模糊推理故障诊断方法采用模糊关系矩阵

【参考文献】