Abstract

Control rod drive mechanism (CRDM) is the crucial component within floating power station associated with the reliability and safety operation. It is challenge to monitor and understand the operating status of CRDM, as characteristic vibration signals of fault are nonlinear and non-stationary. To obtain the key information and de-noise the vibration signal, a combination of semi-soft wavelet threshold (SWT) and Hilbert-Huang transform (HHT) is proposed for roller status diagnosis and de-noising, and local mean decomposition (LMD) method accompanied with support vector machine (SVM) is used for improvement. Hilbert transform is used to further extract fault characteristics of roller vibration signal. By using Semi-soft wavelet threshold method the noise interference is decreased and the effect of endpoint effect on empirical mode decomposition (EMD) is reduced. An evaluation model was established by LMD method to extract the characteristic features from the faults and SVM to collect and analyse the data. The experimental results show that the method can effectively eliminate the interference of noises and realize the status diagnosis of the roller.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call