Abstract

In order to effectively remove the noise interference in the vibration response signal, Smoothing Filtering, a filtering method commonly used in the field of geophysical data processing for improving data signal-to-noise ratio is introduced. The method is mainly realized by triangular filtering, and the triangular filter is constructed by two rectangular filters, which has high computational efficiency. Meanwhile, triangular filtering can also be regarded as a good approximation method of Gaussian filtering. When the trigonometric function is multi-convolved with itself, mathematically, it can approximate the Gaussian function. After multiple iterations, it can also implement an approximate Gaussian filter. Through the processing of the measured vibration signal, the de-noising performance of the median filtering method and the smoothing filtering method is compared. And the advantages of the smoothing filtering method are further testified by using time-frequency analysis. The results show that the smoothing filtering method can better preserve the details of the effective signal while suppressing the noise interference, and effectively improve the data signal-to-noise ratio. The effectiveness and feasibility of the smoothing filtering method applied to the noise suppression of vibration response signals are verified.

How to Cite:

Liu, N., (2019) “Research on vibration signal noise suppression method based on smoothing filtering”, Review of Progress in Quantitative Nondestructive Evaluation .