To address the issue that the single-modal characteristics of acoustic emission signals in rock fracture research are not comprehensive, and the existing methods have insufficient feature mining and utilization, a non-steady-state extension stage identification method based on dual-modal acoustic emission data fusion was proposed. Through uniaxial compression experiments on siltstone, acoustic emission signal data were collected. Acoustic emission parameters were extracted and time-frequency diagrams based on wavelet packet transform were introduced to construct a dual-modal dataset combining acoustic emission parameters and time-frequency diagrams. The InceptionV4 network and 1D convolutional neural network-long short-term memory(1D CNN-LSTM) network were used to form a feature extraction module for dual-modal feature extraction. To fully integrate dual-modal features, a dual-path feature fusion module was designed, including a bimodal tensor fusion module, a cross-modal attention module, and a gating mechanism. This module, along with the feature extraction module and a classification prediction module based on a multi-layer perceptron, constituted the non-steady-state extended stage identification network. The experimental results show that the accuracy and F1 value of this recognition network reach 92.5% and 92.7% respectively, which are 2.33% and 2.72% higher than the optimal single-modal method (1D CNN-LSTM), and 1% and 0.97% higher than the optimal dual-modal fusion method. This result indicates that the proposed method can fully utilize and integrate the features of dual-modal data, achieving balanced optimization of the accuracy and reliability of non-steady-state extended stage identification, and providing a new idea for intelligent early warning of rock instability.
SUN Lin, ZHANG Yang-hua, LIANG Peng, YAO Xu-long, YU Guang-yuan.
Identification of Unsteady Propagation Stage of Rock Cracks Based on Dual-modal Acoustic Emission Fusion[J].
Science Technology and Engineering, 2026, 26(17): 7396-7406 DOI:10.12404/j.issn.1671-1815.2505170
参数集采用结构化数据的形式构建。为保证参数的充分利用,选择原始声发射15个核心参数(上升时间、计数、能量、持续时间、幅值、平均频率、均方根(root mean square,RMS)、平均信号电平(average signal level,ASL)、峰值频率、反算频率、初始频率、信号强度、绝对能量、中心频率以及峰频,并引入上升时间/最大振幅(rise time/amplitude,RA)、振铃计数/持续时间(average frequency,AF)[33]及其衍生比率特征RA/AF值。为消除特征量纲差异并提升模型训练效率,所有数据均进行Z-score标准化处理。该处理严格以训练集的统计量(均值与标准差)为基准,将验证集和测试集的特征分布同步转换至零均值、单位方差的空间。
LiB, WangX, ZhaoH, et al. A multi-scale observation and crack statistics based method for analyzing failure mechanism of pre-flawed rock under true triaxial stress[J]. PLoS One, 2025, 20(5): e0323809.
[2]
LiuR, WangZ, ZhangY, et al. Research on rock fracture evolution prediction model based on Adam-ConvLSTM and transfer learning[J]. Discover Applied Sciences, 2025, 7(3): 1-26.
[3]
LiB, WangE, LiZ, et al. Automatic recognition of effective and interference signals based on machine learning: a case study of acoustic emission and electromagnetic radiation[J]. International Journal of Rock Mechanics and Mining Sciences, 2023, 170: 105505.
ZengJingjiang, ZhaoJun, WangXiaofei. Experiment on time delay destructive characteristics of granite true triaxial rock burst[J]. Science Technology and Engineering, 2025, 25(25): 10850-10860.
[6]
HuangJ, HuQ, SongZ, et al. Classification of cracking sources of different engineering media via machine learning[J]. Fatigue & Fracture of Engineering Materials & Structures, 2021, 44(9): 2475-2488.
[7]
DongL, ZhangY, BiS, et al. Uncertainty investigation for the classification of rock micro-fracture types using acoustic emission parameters[J]. International Journal of Rock Mechanics and Mining Sciences, 2023, 162: 105292.
DaiXin, HuBin, LiJing, et al. Acoustic emission characteristics and fractal features of shear failure in carbonaceous shale[J]. Science Technology and Engineering, 2024, 24(12): 4909-4915.
SunJiehao, GuoBaohua, TianShixuan, et al. Experimental study of direct shear failure characteristics of sandstone joints based on characteristic parameters of acoustic sandstone joints based on characteristic parameters of acoustic[J]. Rock and Soil Mechanics, 2024, 45(S1): 167-177.
LeiRuide, GuQingheng, HuChao, et al. Acoustic emission signal characteristics and precursory recognition of rock failure in fractured sandstone[J]. Rock and Soil Mechanics, 2025, 46(7): 2023-2038.
GaoAnsen, QiChengzhi, FengYanjie, et al. Damage evolution pattern and failure characteristics of red sandstone under uniaxial compression load[J]. Science Technology and Engineering, 2023, 23(23): 9875-9880.
[16]
GaoA, QiC, ShanR, et al. Identification and early warning method of key disaster-causing factors of AE signals for red sandstone by principal component analysis method[J]. Ain Shams Engineering Journal, 2023, 14(10): 102205.
[17]
FukunX, HouranW, GangL. Study on multiparameter precursory information identification of the fracture of yellow sandstone[J]. Advances in Civil Engineering, 2019, 2019(1): 7676801.
LiuJian, ZhouZonghong, ZhangJing, et al. Acoustic emission characteristics of diorite at varying unloading rates and identification of its unsteady phases[J]. Rock and Soil Mechanics, 2025, 46(1): 225-232, 243.
[20]
RenF, ZhuC, YuanZ, et al. Recognition of shear and tension signals based on acoustic emission parameters and waveform using machine learning methods[J]. International Journal of Rock Mechanics and Mining Sciences, 2023, 171: 105578.
[21]
DuanD, FengX, ZhangR, et al. Research on recognition of quiet period of sandstone acoustic emission based on four machine learning algorithms[J]. Geofluids, 2022, 2022(1): 2133607.
[22]
DaiJ, LiuJ, ZhouL, et al. Crack pattern recognition based on acoustic emission waveform features[J]. Rock Mechanics and Rock Engineering, 2023, 56(2): 1063-1076.
[23]
HuangY, DengJ, ZhuJ. An experimental investigation of moisture-induced softening mechanism of marble based on quantitative analysis of acoustic emission waveforms[J]. Applied Sciences, 2019, 9(3): 446.
[24]
YangL, WuB, XuY, et al. Damage evolution during rock dynamic compression revealed by wavelet analysis of acoustic emission signals[J]. Rock Mechanics and Rock Engineering, 2024, 57(2): 1527-1535.
[25]
李玄昆. 基于声发射频谱特征的大理岩损伤力学特性研究[D]. 成都: 成都大学, 2023.
[26]
LiXuankun. Experimental study on damage behavior of marble based on the analysis of acoustic emission spectrum characteristics[D]. Chengdu: Chengdu University, 2023.
HuangHuasen, ZhangYongfa, WangChaolin, et al. Failure mechanism analysis of sandstone with cavity based on acoustic emission spectrum[J]. Science Technology and Engineering, 2025, 25(14): 6023-6034.
HuYing, XiaoYao, LiuHanxiang, et al. Rock fracture precursor identification method based on acoustic emission time-frequency characteristics[J]. Science Technology and Engineering, 2022, 22(34): 15259-15268.
ChangXinke, WuShunchuan, ChengHaiyong, et al. Full-waveform characteristics and frequency spectrum of acoustic emission for fine sandstone[J]. Journal of Changjiang River Scientific Research Institute, 2022, 39(7): 118-125.
[33]
NiuY, WangJ, HuY, et al. Time-frequency domain characteristics of intact and cracked red sandstone based on acoustic emission waveforms[J]. Geomechanics and Engineering, 2023, 34(1): 1-15.
[34]
YangH, WangE, SongY, et al. Identification and prediction method for acoustic emission and electromagnetic radiation signals of rock burst based on deep learning[J]. Physics of Fluids, 2024, 36(7): 0219409.
[35]
WangW. Recognition algorithm of AE signal of rock fracture based on multiscale 1DCNN-BLSTM[J]. Journal of Electrical and Computer Engineering, 2024, 2024(1): 3717867.
[36]
PuY, ChenJ, ApelD B, et al. Real-time identification of rock failure stages using deep learning: a case study of acoustic emission analysis in rock engineering[J]. Rock Mechanics and Rock Engineering, 2025, 58(5): 1-17.
[37]
SongZ, HuangJ, DengB, et al. Influence of acoustic emission sequence length on intelligent identification accuracy of 3-D loaded rock’s fracture stage[J]. Engineering Failure Analysis, 2024, 162: 108389.
[38]
ZhangL, LiK, WangH, et al. MFC-PINN: a method to improve the accuracy and robustness of acoustic emission source planar localization[J]. Measurement, 2024, 235: 114995.
[39]
ZhaoC, SunJ, MaZ, et al. A multimodal information fusion method for real-time monitoring of glass cutting with laser-induced thermal-crack propagation[J]. Optics & Laser Technology, 2025, 180: 111457.
[40]
WangR, ShanG, QiuF, et al. Corrosion state monitoring based on multi-granularity synergistic learning of acoustic emission and electrochemical noise signals[J]. Processes, 2024, 12(12): 2935.
ZhangYanbo, LiangPeng, SunLin, et al. Spectral characteristics of AE in the process of saturated granite fracture under uniaxial compression[J]. Rock and Soil Mechanics, 2019, 40(7): 2497-2506.
ZhangYanbo, LiangPeng, SunLin, et al. Frequency characteristics of acoustic emission signal during rupture process of siltstone based on HHT[J]. Journal of Mining and Safety Engineering, 2016, 33(1): 179-184.
DongLongjun, ZhangYihan, SunDaoyuan, et al. Stage characteristics of acoustic emission and identification of unstable crack state for granite fractures[J]. Chinese Journal of Rock Mechanics and Engineering, 2022, 41(1): 120-131.
[47]
SzegedyC, IoffeS, VanhouckeV. Inception-v4, Inception-ResNet and the impact of residual connections on learning[J]. arXiv: 1602. 07261, 2016.
[48]
ZadehA, ChenM, PoriaS, et al. Tensor fusion network for multimodal sentiment analysis[J]. arXiv Preprint, arXiv: 1707. 07250, 2017.
[49]
VaswaniA, ShazeerN, ParmarN, et al. Attention is all you need[J]. Advances in Neural Information Processing Systems, 2017, 30: 6000-6010.