Effect of Noise and Measurement Distance on Wavelet Transform-Based Damage Detection

Abstract

Wavelet Transform (WT) has been proven to be a reliable method of damage detection in structures. This is partly due to its ability to detect local damage. However, the effect of noise may result to false detection of damage, thus affecting the reliability of the method. Moreover, the location of sensors also plays an important role in providing good quality dynamic response data of structure, where a small number of sensors may lead to inadequate data for accurate estimation of the structural damage, while a large number increases the financial cost of the monitoring system and causes higher computational time. In this view, this study investigates the effect of noise and different sensor distance to Continuous Wavelet Transform (CWT) for damage detection through the decomposition of mode shape differences. In this study, a numerical model of a plate with all four sides fixed is used as an example. To analyse the noise effect, various noise levels are added to the mode shape difference signal before decomposition. The effect of the sensor distance is done by using different sensor measuring distances. The results indicate that the increase of noise reduced the detectability of damage. It is also observed that excessive sensor distance increment significantly effects damage detectability.