摘要
为了有效的实现滚动轴承的故障诊断,提出基于近似等距投影和支持向量机的滚动轴承故障诊断方法。该方法首先使用高斯随机投影矩阵对数据进行降维投影得到压缩数据,根据近似等距投影性质压缩数据能够保持原始信号的结构;然后从压缩数据中提取压缩域特征并作为支持向量机的输入,建立滚动轴承故障诊断模型,实现轴承的故障诊断。使用不同状态的轴承实测数据进行验证,结果表明该方法能够获得准确的结果。
A new method based on near-isometric projection and support vector machine was proposed for fault diagnosis of rolling bearings.Firstly,Gaussian random projection matrix was utilized to do dimension reduction projection for the signal data to obtain the compressed data.According to the near-isometric projection property,the compressed data kept the structure of the original signals.Then the compressed domain features were extracted from the compressed data,they were taken as the input of a support vector machine to establish the fault diagnosis model of rolling bearings and realize fault diagnosis of rolling bearings.The actual measured data of rolling bearings in different faulty states were used to verify the new method.Results demonstrated the correctness and effectiveness of the proposed method.
作者
刘畅
伍星
刘韬
柳小勤
LIU Chang;WU Xing;LIU Tao;LIU Xiaoqin(College of Mechanical & Electrical Engineering, Kunming University of Science & Technology, Kunming 650500, China)
出处
《振动与冲击》
EI
CSCD
北大核心
2018年第5期234-239,共6页
Journal of Vibration and Shock
基金
国家自然科学基金资助项目(51405211)
关键词
滚动轴承
故障诊断
近似等距投影
支持向量机
rolling bearing
fault diagnosis
near-isometric projection
support vector machine