摘要
工程机械液压系统发生冲击故障时,液压冲击引起的振动信号包含了大量的故障信息。该文针对液压冲击产生的振动信号,通过EEMD方法计算有效IMF分量的能量分布作为振动信号的特征向量,研究了基于SVM分类预测的典型冲击振动信号的高维大样本的分类识别,比较了不同SVM分类器的分类识别效果。结果表明:基于EEMD和SVM相结合的方法可有效进行高维大样本条件下液压系统冲击振动信号分类识别,能实现液压系统冲击振动信号的智能诊断。
When the hydraulic system of construction machinery has an impact fault,the vibration signal caused by the hydraulic impact contains a lot of fault information.In this paper,aiming at the vibration signal generated by hydraulic impact,the EEMD method was used to calculate the energy distribution of the effective IMF component as the feature vector of the vibration signal,and the SVM learning prediction function was used to study the classification and recognition of large samples of typical shock and vibration signals in high dimension,and the classification and recognition effects of different SVM classifiers were compared.The results show that the method based on the combination of EEMD and SVM can effectively classify and identify the impact vibration signals of the hydraulic system under the condition of high dimensional and large sample size,and can realize the intelligent diagnosis of the impact vibration signals of the hydraulic system.
作者
高立龙
蒋文峰
黄鹤
赵钰淞
GAO Li-long;JIANG Wen-feng;HUANG He;ZHAO Yu-song(Noncommissioned Officer Academy,Army Artillery and Air Defense Academy,Shenyang 110000,China)
出处
《液压气动与密封》
2021年第8期14-18,共5页
Hydraulics Pneumatics & Seals
