摘要
重载的铁路货车滚动轴承极易出现损伤类故障,因故障的突发性使得振动信号具有非线性、非平稳的特性,且振动信号极易受非检测部位振动的干扰,而使得有效信号不易被提取,传统的基于傅里叶变换的检测方法无法有效将非平稳的振动信号中的干扰去除,造成最终的检测准确率不高的问题。为了提高检测准确率,提出基于小波包的检测方法,首先针对振动信号的非平稳特性采用小波滤波有效去除其中的干扰,然后通过最小二乘支持向量机提取出非平稳振动信号中的故障特征信号并完成故障检测。实验表明,这种方法能够有效去除振动信号中的干扰,并准确检测重载铁路货车上的滚动轴承损伤故障。
The research on the railway freight trains overloading of rolling bearing fault of the accurate detection of damage the heavy railway wagon rolling bearing fault of appear easily injury, due to a failure of the vibration signal of sudden made with nonlinear non-stationary characteristics, and vibration signal by the testing site easily vibration interference, and make effective signal not easily by extraction. The traditional based on Fourier transform detection method can not be nonstationary vibration signal of will remove the interference, cause the final testing accuracy is not high problems in order to improve the detection accuracy, based on the wavelet packet inspection method, firstly the nonstationary vibration signal characteristics of wavelet filter remove one of the interference, and then through the least square support vector machine （SVM） extract the nonstationary vibration signal of the fault characteristic signal and complete the fault detection experiments show that this method can effectively remove the vibration signal interference, and accurate detection of overloading of the rolling bearing on the railroad freight damage fault.
出处
《科技通报》
北大核心
2012年第7期95-97,117,共4页
Bulletin of Science and Technology
基金
四川省教育厅重点项目(10ZC036)
关键词
滚动轴承
振动信号
小波包
rolling bearing
vibration signal
wavelet packet
