Early‐warning signal recognition methods in flawed sandstone subjected to uniaxial compression

Abstract

To investigate the precursor information related to rock fracturing, a series of indoor uniaxial compression tests are conducted on flawed sandstone. Four methods of identifying potential precursors to rock catastrophic failure, namely, cumulative acoustic emission (AE) activity, AE b-value, inter-event time function, and critical slowing down, are discussed. The results show that the four precursory identification methods can reasonably identify the earlywarning signal, subcritical failure, and final catastrophic failure of flawed sandstone involving the whole loading process. The minimum b-value is correspondingly obtained when the specimen approaches failure. The inter-event rate of AE multi-parameters shows a significant increase, following a power law distribution. By comparison with the other three precursor identification methods, the time intervals of AE parameters-based variance between the precursory signal and its corresponding tripping points become increasingly shortened. Additionally, the precursory characteristics of variance of AE multiparameters are more accurate and reliable than that of autocorrelation coefficient.

Temporal‐frequency distribution and multi‐fractal characterization of acoustic emission of rock materials containing two parallel pre‐existing flaws

“…The corresponding ligament angles are 0°, 30°, 60°, 90°, 120°, and 150°. To facilitate the description, the samples are designated by a combination of letters and numbers, in which SN15‐150, SG15‐60, and SI are taken as an example, where S, N, I, 15, and 150 represent sandstone, non‐infilled, intact, flaw inclination angle, and ligament angle, respectively 33 …”

Research on modulation recognition method of electromagnetic signal based on wavelet transform convolutional neural network

“…However, this method does not preprocess the electromagnetic signal, and is affected by the noise of the electromagnetic signal, which leads to the low accuracy of extracting the characteristics of the electromagnetic signal. Lei, R. et al extracted the signal identification method under uniaxial compression [11], which used principal component analysis to extract the characteristics of electromagnetic signals and complete the modulation identification of electromagnetic signals. However, this method has a large amount of calculation in the process of data analysis, which leads to a long time for electromagnetic signal modulation recognition.…”

Creep Behavior and Its Prediction of Saturated Weakly Cemented Medium-Grained Sandstone under Multiaxial Loadings

“…In addition, the stability evaluation of water-resisting strata is important to ensure the safety of mining. The signs of rock failure can be obtained by abnormal deformation or external monitoring [18][19][20][21][22], but they are not suitable for failure prediction considering rock anisotropy. The creep model can be used to describe the relationship between rock deformation and time, and rock deformation is related to the development of cracks [23].…”