Experimental Study on Breathing Crack Detection and Evaluation Under Random Loading With Entropy

Abstract

Breathing crack detection and evaluation on a beam under random loading is experimentally studied and realized using entropy measures. During testing, the beam is subjected to random excitations which is the most evident excitation type experienced by most of the engineering structures. Frequency response function is employed to erase the random frequency components due to the excitation from the response of the structure to pick up the vibration characteristics of the beam structure itself with/without the breathing crack. Based on experimental results, the proposed methodology can clearly discriminate a crack with 25% depth of the total thickness of the beam from the healthy case. It is also capable of distinguishing a 50% crack depth from a 25% crack depth with distinct entropy values.

Breathing crack damage diagnostic strategy using improved MFCC features

“…It is also highly difficult to produce breathing cracks even using an accelerated fatigue machine. In view of this, in the present work, the breathing fatigue cracks are constructed by bonding three aluminum alloy beams together; the same technique has been used by various researchers across the globe such as Prime and Shevitz (1995), Douka and Hadjileontiadis (2005), Yelve et al (2014) and Wimarshana et al (2017, 2018) to simulate breathing cracks experimentally. The fabrication of the beam with a breathing crack is illustrated in Figure 18.…”

“…Only fewer techniques (Asnaashari and Sinha, 2014; Wimarshana et al, 2017, 2018) use ambient vibration data for breathing crack identification. The reason behind this is that the presence of noise may sometimes be wrongly construed as nonlinearity and vice versa.…”

“…However, the method is highly dependent on input excitation position, crack position, crack depth and number of modes being considered. Wimarshana et al (2017, 2018) developed wavelet entropy-based approach for breathing crack identification using random vibration data. However, the work is limited to confirming the presence of a breathing crack in the structure only.…”