摘要
故障特征信息提取是滚动轴承故障诊断的关键问题之一。针对轴承故障信号的提取,提出了改进的DCT和EMD相结合的轴承故障诊断方法。首先对故障信号进行DCT变换得到多个DCT系数,对DCT系数进行阈值处理后,重构故障信号。然后对重构信号进行EMD分解,得到多个IMF,并对轴承故障频率所对应的IMF做频谱分析。此方法减小了噪声对EMD的影响,提高了信噪比,能够对轴承故障进行准确的诊断。
The extraction of fault features is one of the key issues for rolling bearing fault diagnosis. For bearing fault signal extraction, the method for bearing fault diagnosis is given, which combines improved DCT and EMD. DCT transform is used to process fault signals to get a number of DCT coefficients, and the fault signals are reconstructed after threshold treatment of DCT coefficients. Then EMD is used to the reconstructed signals, a number of intrinsic mode functions （IMF） are got, spectrum analysis is used to IMF corresponding to the bearing fault frequency. This method reduces the effect of noise on EMD and improves SNR, which is able to diagnose bearing fault accurately.
出处
《轴承》
北大核心
2013年第3期53-56,共4页
Bearing
