摘要

To study the precursory failure infrasonic characteristics of coal samples, coal rock stress loading system and infrasonic wave acquisition system were adopted, and infrasonic tests in uniaxial loading process were made for the coal samples in the studied area. Wavelet filtering, fast Fourier transform, and relative infrasonic energy methods were used to analyze the characteristics of the infrasonic waves in the loading process, including time domain characteristics, and relative energy. The analysis results demonstrated that the frequencies of the infrasonic signals in the loading process mainly distribute within 5-10 Hz, which are significantly different from noise signals. The changes of the infrasonic signals show clear periodic characters in time domain. Meanwhile, the relative energy changes of the infrasonic wave also show periodic characters, which are divided into two stages by the yield limit of coal samples, and are clear and easy to be recognized, so that they can be used as the precursory characteristics for recognizing coal sample failures. Moreover, the infrasonic waves generated by coal samples have low frequency and low attenuation, which can be collected without coupling and transmitted in long distance. This study provides an important support for the further in-situ prediction of coal rock failures.