Article
Engineering, Geological
Ling Yang, Bangbiao Wu, Ying Xu, Yan Fu, Kaiwen Xia
Summary: A new method is proposed in this study to study the damage evolution of rocks during dynamic compression by continuous wavelet analysis of acoustic emission signals. Based on the characteristics of the time-frequency spectrums, the rock exhibits a mixed damage mode shifting from tensile to shear damage. The increase in the loading rate leads to an increase in the amount of tensile damage.
ROCK MECHANICS AND ROCK ENGINEERING
(2023)
Article
Geochemistry & Geophysics
Yalei Wang, Jinming Xu
Summary: This study investigates the deformation and failure process of rocks under different pressures using acoustic emission (AE) features. The results indicate that unloading direction of rocks and increased confining pressure can lead to significant dilatation deformation and severe rock burst. Furthermore, the unloading failure of rocks involves a mixed failure mode of tensile and shear, and the occurrence of AE events is concentrated at the top and bottom of the rocks.
Article
Geosciences, Multidisciplinary
Sijie Liu, Haijun Zheng, Guoqing Chen, Yitao Hu, Kai Meng
Summary: Rock failure can have serious consequences, therefore obtaining precursor information using techniques like acoustic emission (AE) is important. AE detection technology can monitor changes inside rocks in real time and predict the process of failure. An experimental study on red sandstone found that AE parameters go through different stages during failure, and the occurrence moment and clarity of the frequent period determine the reliability and priority of the failure precursor information. The reliability and priority of AE precursor information increase with the rise of intermediate principal stress. Comparatively, AE precursor information occurs before thermal infrared precursor information.
FRONTIERS IN EARTH SCIENCE
(2023)
Article
Engineering, Geological
Yangyang Di, Enyuan Wang, Tao Huang
Summary: This study aims to overcome the difficulties in identifying microseismic, acoustic emission, and electromagnetic radiation interference signals through on-site monitoring and deep learning methods. Identification models based on ResNet-50 convolutional neural network and recurrent neural network were constructed for microseismic waveforms and AE/EMR interference signals respectively, using the original data. The proposed method can effectively eliminate interference signals and improve the reliability of monitoring data, thus effectively monitoring rock burst disasters.
INTERNATIONAL JOURNAL OF ROCK MECHANICS AND MINING SCIENCES
(2023)
Article
Engineering, Geological
Wenkai Wan, Charlie C. Li
Summary: This study investigates the progressive microcracking processes in burst-prone Class II rock (Kuru granite) and non-burst-prone Class I rock (Fauske marble) to understand the physics of rock burst and the difference in burst-proneness between the two classes. The study finds that extensional intergranular cracking dominated the damaging process in Kuru granite, while intragranular shear cracking in calcite dominated the damaging process in Fauske marble. This suggests that the fracturing process in Kuru granite dissipates less strain energy and releases more energy for rock ejection compared to Fauske marble.
ROCK MECHANICS AND ROCK ENGINEERING
(2022)
Article
Engineering, Geological
Yangyang Di, Enyuan Wang, Zhonghui Li, Xiaofei Liu, Tao Huang, Jiajie Yao
Summary: A comprehensive early warning method for rock burst in coal mines based on deep learning algorithm is proposed in this paper. It uses LSTM-RNNs and CNN to intelligently identify the precursor signals of rock burst risk, analyzes the data to obtain the risk coefficient, predicts the risk coefficient using multi-input CNN, and completes the comprehensive early warning of rock burst. Field verification shows that this method can respond positively to rock burst risk and capture the information in advance.
INTERNATIONAL JOURNAL OF ROCK MECHANICS AND MINING SCIENCES
(2023)
Article
Engineering, Geological
Zhenghu Zhang, Ke Ma, Hua Li, Zhiliang He
Summary: In this study, a micro analysis of rock direct tensile failure was conducted using quantitative analysis of acoustic emission (AE) waveforms. The results show that the complexity of mineral composition corresponds to the dispersion degree of rock tensile strength. L-type waveforms carry more energy than H-type waveforms under tension. The peak strength of rock specimens shows a downward trend with increasing energy ratios of L-type waveforms. Micro shear fractures were observed during the macro tensile failure process. The determined tensile strength can be used as a conservative design parameter for rock engineering.
ROCK MECHANICS AND ROCK ENGINEERING
(2022)
Article
Environmental Sciences
Yi Zhang, Guangliang Feng, Manqing Lin, Xianfu Li, Chengcheng Gao, Xiaoshuai Liang
Summary: Rock mass failure is becoming more common in geotechnical engineering projects. This paper analyzes the probability density distributions of initial and peak frequency events in the acoustic emission (AE) data from two types of rock undergoing failure. The study finds that events with an initial frequency of 1,000 kHz and peak frequency of 625 kHz have higher probabilities. The cumulative probability distributions (CPDs) of the characteristic events in the AE data are also investigated, revealing their evolutionary behavior as the rocks fail.
FRONTIERS IN ENVIRONMENTAL SCIENCE
(2022)
Article
Geochemistry & Geophysics
Rui Yang, Lv Jiakun, Bo Zhou, Depeng Ma
Summary: The mechanical response characteristics and occurrence mechanism of coal and rock under unloading conditions are crucial for evaluating the stability and control of rock excavation in engineering. Fractal characteristics of coal and rock acoustic emission time series were analyzed to predict the unloading failure of coal and rock. In addition, the HURST index was calculated to determine the unloading and fracture process of rock samples.
Article
Mining & Mineral Processing
Longjun Dong, Yongchao Chen, Daoyuan Sun, Yihan Zhang
Summary: This study investigated the qualitative relationship between rock instability precursors and principal stress direction through wave velocity in rock acoustic emission (AE) experiments. The results show that the anisotropic characteristics of wave velocity variations can help identify the principal stress direction, and both the AE event rate and wave velocity are effective monitoring parameters for rock instability. The anisotropic characteristics of wave velocity variation and AE event rate are beneficial complements for identifying the rock instability precursors and determining the principal stress direction, providing a new analysis method for stability monitoring in practical rock engineering.
INTERNATIONAL JOURNAL OF MINING SCIENCE AND TECHNOLOGY
(2021)
Article
Chemistry, Analytical
Lin Sun, Lisen Lin, Xulong Yao, Yanbo Zhang, Zhigang Tao, Peng Ling
Summary: The characteristics of acoustic emission signals generated in the process of rock deformation and fission contain rich information on internal rock damage. This paper proposes a method based on speech recognition and spectral analysis to extract features from acoustic emissions of rock fracture. Six intelligent real-time recognition models were constructed using deep learning techniques, achieving high accuracy and efficiency in the recognition of key signals. The Mel+VGG-FL model performed the best with an accuracy of 87.68% and a recall of 81.05%.
Article
Engineering, Geological
Jun Zhu, Jianhui Deng, Po Ning, Ziguo Fu, Xuankun Li, Ronald Y. S. Pak
Summary: This study investigates the effects of water saturation on the tensile strength of rocks through experiments and acoustic emission (AE) techniques. The results reveal the reduction in tensile strength caused by water saturation and establish the correspondence between AE waveform types and micro-failure patterns of rocks.
ROCK MECHANICS AND ROCK ENGINEERING
(2023)
Article
Chemistry, Multidisciplinary
Jui Hsiang Kao
Summary: A new method for calculating radiating pressures of objects at interior resonance frequencies is proposed in this study, with robustly convergent results confirmed through examples.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Geological
Z. W. Ding, X. F. Li, X. Huang, M. B. Wang, Q. B. Tang, J. D. Jia
Summary: In this study, the AE signal characteristics of coal and rock samples were extracted using the MFCC method, and a stress state criterion based on signal features was constructed. By employing a BP neural network for deep learning, the recognition, classification, and prediction of coal and rock materials were successfully achieved.
INTERNATIONAL JOURNAL OF ROCK MECHANICS AND MINING SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Weinan Liu, Guojun Zhang, Yu Huang, Wenyuan Li, Youmin Rong, Ranwu Yang
Summary: This paper investigates the complex interaction between nanosecond laser and glass, focusing on the generation of defects during laser scribing such as micro-cracks and fractures. The acoustic emission (AE) technique is employed to monitor the laser scribing of float glass and a strong correlation between AE signals and the width of laser ablation area is observed. The study also explores the correlation between AE signals and the quality of laser scribing and applies time-frequency spectrum analysis for defect diagnosis.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)