Deterministic and probabilistic fatigue prognosis of cracked specimens using acoustic emissions

A novel closure based approach for fatigue crack length estimation using the acoustic emission technique in structural health monitoring applications

“…Verification of the derived model was conducted using separate AE data sets recorded from tests 8 and 9 performed with Δσ of 27 MPa and stress ratio of 0.1. Figure 12 shows the averaged points in the normalised stress range where AE signals were generated at various crack lengths throughout the tests, along with the model from equation (13). It can be observed that the results also follow the exponential trend.…”

“…This diagram represents AE signals produced in the lower 70% of the stress range. An exponential best-fit model was used to characterise their relationship as also illustrated in figure 10 and expressed in equation (13). It can be seen that the majority of data points are located within the confidence bounds with some outliers in the period of crack growth between crack lengths of 10-15 mm.…”

“…It was therefore suggested that this group of AE signals were more closely related to the fatigue processes occurring around the crack tip and plastic zone [12]. This relationship has been further explored in several studies [13][14][15][16]. Yu et al (2012) [14] showed that this approach was dependent on stress ratio (minimum load / maximum load) as different constants were obtained for the regression models derived from AE data recorded in fatigue tests performed at different stress ratios on samples of identical geometry.…”

“…A comparison was made between the novel approach described above for crack length estimation and another empirical approach based on AE signal features as described in [11][12][13][14][15][16][17]. In this alternative approach, correlations made between the rates of change of AE signal counts (dc/dN) and absolute energy (de/dN) with fatigue crack growth rates (da/ dN) and crack tip stress intensity range (ΔK) values applying the expressions in equations ( 7) and ( 9).…”

“…It is well established that fatigue crack growth rates for part of the growth rate range, exhibit a relationship with crack tip stress intensity range (ΔK) as expressed in equation ( 7) [11]. However, Morton et al [12] and Zarate et al [13] also showed linear correlations between AE signal Counts (number of threshold crossings of the signal) and Absolute Energy rates per loading period for AE signals recorded close to the peak load with crack tip stress intensity range (ΔK) and crack growth rates (da/dN) as expressed in equations ( 8)- (9). It was therefore suggested that this group of AE signals were more closely related to the fatigue processes occurring around the crack tip and plastic zone [12].…”