Vibration acceleration signals are often measured from case surface of a running machine to monitor its condition. If the measured vibration signals display to have periodic impulse components with a certain frequency, there may exist a corresponding local fault in the machine, and if further extracting the periodic impulse components from the vibration signals, the severity of the local fault can be estimated and tracked. However, the signal-to-noise ratios (SNRs) of the vibration acceleration signals are often so small that the periodic impulse components are submersed in much background noises and other components, and it is difficult or inconvenient for us to detect and extract the periodic impulse components with the current common analyzing methods for vibration signals. Therefore, another technique, called singular value decomposition (SVD), is tried to be introduced to solve the problem. First, the principle of detecting and extracting the signal periodic components using singular value decomposition is summarized and discussed. Second, the infeasibility of the direct use of the existing SVD based detecting and extracting approach is pointed out. Third, the approach to construct the matrix for SVD from the signal series is improved largely, which is the key program to improve the SVD technique; Other associated improvement is also proposed. Finally, a simulating application example and a real-life application example on detecting and extracting the periodic impulse components are given, which showed that the introduced and improved SVD technique is feasible.
