Detrend of Vibration Signals Based on EMD-SSSC Decomposition
A new method for removing the trend of vibration signals based on Empirical Mode Decomposition and Soft Sifting Stopping Criterion(EMD-SSSC)is proposed.EMD-SSSC improves IMF' s aliasing problem through the soft screening stopping criterion and adaptively controls the screening process.By EMD-SSSC,the accuracy and efficiency of EMD are raised,and the trend terms in the vibration signals are effectively removed.The precision,feasibility and effectiveness of the proposed method are verified by an analytical function and a 2-DOF spring-mass-damper system.The method is also utilized to remove the trend of monitoring acceleration signals of a practical quayside container crane.The results show that the proposed EMD-SSSC is capable for removing the trend much more accurately than the least-squares method,and can remove the trend caused by numerical integration,as well as the trend of practical monitoring signals.