Experimental study of infrasonic signal generation during rock fracture under uniaxial compression

Abstract The low frequency infrasonic signals generated during micro fracture process of six types of rocks (including granite, mudstone, sandstone, phyllite, limestone and red sandstone) under uniaxial compression were detected. A specialized data collection system was developed for infrasonic signals sensing and analog-to-digital conversion. The wavelet-based de-noising method, Short-Time Fourier Transform (STFT) time-frequency analysis and accumulative ring-down count (ARC) of events were used to process and analyze digital infrasonic signals in order to investigate the characteristics of infrasonic emission (IE) event during rock micro fracture process. The experimental results indicate obvious infrasonic emission events before rock failure. The characteristic frequency typically ranges from about 2.0 to 8.0 Hz for the six kinds of rocks and it has no necessary relationship with structure or lithology. According to the ARC estimate, hard rocks (such as granite, limestone) can generate much more infrasonic emission events than soft ones (mudstone, phyllite), in other words, the brittle fracture is more prone to generate infrasonic signals. In addition, infrasound is an inaudible and low-attenuate signal that cannot be absorbed easily by water or air, and can propagate avoiding huge obstacles due to its long wavelength. The research results and the advantages of the infrasound will provide a beneficial supplement for our further studies on rock instability or rocky landslide in field combined with other advanced methods.