A dominant frequency-based hierarchical clustering method for evaluating the mode-II fracture mechanisms of shale using the acoustic emission technique

The fracture toughness of rocks is a crucial fracture parameter in unconventional geo-energy exploitation and plays a significant role in the fields of fracture mechanics and hydraulic fracturing. This research first introduces the processing approaches of AE (acoustic emission) signals for further elucidating the fracture mechanisms. To measure the mode-II fracture toughness of Longmaxi shale, the PTS (punch-through shear) specimen is employed because it can accomplish both mode-II loading and mode-II fracturing. Interestingly, the Gaussian function provides a superb description of the distribution of AE amplitudes, and the discrepancies in AE amplitude distribution are conspicuous for PTS shale specimens with typical bedding angles. The conventional RA-AF analysis of AE signals is more appropriate for qualitative analysis of failure mechanisms. When relatively higher RA values are monitored, implying that the partial stress drop and large-scale or more intense cracks take place, which can adequately furnish precursory information for engineering failure diagnosis. Based on the correlation between damage mechanisms (or fracture modes) and the characteristic bands of AE dominant frequency distribution, the dominant frequency-based hierarchical clustering method is developed by this work to quantitatively identify the damage or fracture modes, then the demarcation of AE dominant frequency ranges is determined as low, intermediate, and high, with corresponding tensile, tensile-shear, and shear cracks.