The characteristics of acoustic emission signals generated in the process of rock deformation and fission contain rich information on internal rock damage. The use of acoustic emissions monitoring technology can analyze and identify the precursor information of rock failure. At present, in the field of acoustic emissions monitoring and the early warning of rock fracture disasters, there is no real-time identification method for a disaster precursor characteristic signal. It is easy to lose information by analyzing the characteristic parameters of traditional acoustic emissions to find signals that serve as precursors to disasters, and analysis has mostly been based on post-analysis, which leads to poor real-time recognition of disaster precursor characteristics and low application levels in the engineering field. Based on this, this paper regards the acoustic emissions signal of rock fracture as a kind of speech signal generated by rock fracture uses this idea of speech recognition for reference alongside spectral analysis (STFT) and Mel frequency analysis to realize the feature extraction of acoustic emissions from rock fracture. In deep learning, based on the VGG16 convolutional neural network and AlexNet convolutional neural network, six intelligent real-time recognition models of rock fracture and key acoustic emission signals were constructed, and the network structure and loss function of traditional VGG16 were optimized. The experimental results show that these six deep-learning models can achieve the real-time intelligent recognition of key signals, and Mel, combined with the improved VGG16, achieved the best performance with 87.68% accuracy and 81.05% recall. Then, by comparing multiple groups of signal recognition models, Mel+VGG-FL proposed in this paper was verified as having a high recognition accuracy and certain recognition efficiency, performing the intelligent real-time recognition of key acoustic emission signals in the process of rock fracture more accurately, which can provide new ideas and methods for related research and the real-time intelligent recognition of rock fracture precursor characteristics.

Studying the characteristics of tensile and shear cracks in a rockburst event is crucial for understanding the rockburst mechanism. In this study, several specimens were tested under biaxial compression, and the released acoustic emissions (AEs) of the resulting tunnel rockburst event were recorded. Based on the AE energy and dominant frequency parameters, the different AE characteristics and dominant frequency evolution law of tensile and shear cracks during the rockburst were analysed. According to the results, there are four types of AE signals in a rockburst event: low-frequency and low-energy signals, low-frequency and high-energy signals, medium-frequency and low-energy signals and high-frequency and low-energy signals. The low-frequency and low-energy signals are mainly generated by tensile cracks, and the low-frequency and high-energy signals originate from both tensile and shear cracks. The medium-frequency and low-energy signals and high-frequency and low-energy signals are often generated by tension cracks. The rockburst process can be classified into four stages: calm period, small-grain ejection period, sheet-like peeling period and violent ejection period. Before the occurrence of rockburst, both tensile and shear cracks produce low-frequency and high-energy signals. The low-frequency and high-energy (amplitude) AE signals are crucial for monitoring and early warning procedures for rockburst events.

Unequal flaws are widespread in rock engineering, in which the extensions induced by excavation and unloading are the main factors leading to engineering instability. Rock deformation and failure are essentially the results of energy transmission. In order to study the influence of flaw type and angle on the mechanical properties and energy characteristics of the flawed rock mass, uniaxial compression tests were conducted using particle flow numerical analysis software. The results show that the crack initiation strength and peak strength of the sample increase with the increase in the flaw angle. For smaller flaw angles, the sample shows obvious plastic deformation during uniaxial compression. For larger flaw angles, the sample shows elastic–brittle properties, whose storage energy is larger. The peak strength of samples with flaws of unequal length presents an obvious decrease compared with that of flawed samples with flaws of equal length, and the extent of the reduction is 6.15% on average. Unequal flaws decrease the ability of the flawed sample to absorb strain energy. Compared with flawed samples with equal flaws, flawed samples with unequal flaws decrease the boundary energy and elastic strain energy by 9.29% and 9.95% on average, respectively. Flaws of unequal length in this sample can weaken the performance of the flawed sample.